Titans Of Tomorrow

Adam Grimes argues retail traders should avoid high-frequency strategies and deceptive order flow, favoring 2-5 day holds and timeless mean reversion.

1h 32m Listen to the original Episode page
Retail TradingOrder FlowMean ReversionTechnical AnalysisTrade ManagementMarket MicrostructureAlpha DecayDiscretionary Trading

Summary

Adam Grimes and a guest discuss the limitations of retail trading infrastructure and the deceptive nature of order flow. Grimes asserts that technical indicators like Fibonacci levels lack statistical significance when tested against random noise. He recommends a two-to-five-day holding period as the optimal timeframe for retail traders to avoid institutional speed advantages. The speakers emphasize that momentum and mean reversion are timeless forces, while specific chart patterns decay. They also cover trade management, warning against the 'timeframe justification system' and advocating for dynamic risk adjustments based on recent performance.

Key takeaways

  1. 1
    Grimes claims Fibonacci levels, moving averages, and VWAP perform no better than random lines when tested against historical data, indicating they lack inherent predictive power.

    Unidentified speaker (speaker_1) · 1:14

Episode sections

  1. 0:00 – 1:14

    Introduction and the Trader's Unique Data Scope

    Adam Grimes introduces his background as a former NYMEX and SMB Capital trader who has analyzed price data dating back to the Phoenicians. He contrasts his long-term approach with high-frequency firms like Jane Street, noting that he cannot compete on speed or co-location advantages.

  2. 1:14 – 6:11

    Testing Technical Indicators Against Random Noise

    Grimes argues that Fibonacci levels, moving averages, and VWAP are not statistically special because they perform no better than random lines when tested against historical data. He suggests that if an indicator were truly significant, its predictive power would be visible in the data compared to dummy levels.

  3. 6:11 – 11:07

    The Optimal Timeframe for Retail Traders

    Grimes identifies a holding period of two to five days as the sweet spot for retail traders. He explains that shorter timeframes are dominated by institutional infrastructure, while longer timeframes are dominated by fundamental data access, leaving the medium-term as the least contested space.

  4. 11:07 – 13:45

    Market Evolution and Alpha Decay

    Grimes believes the market is at a historical inflection point similar to the arrival of the internet or electricity. He observes that specific trading edges are decaying and suspects that new forms of AI or digital market structures are reshaping market dynamics.

  5. 13:45 – 19:04

    Quantitative Testing of Fibonacci Levels

    To test if Fibonacci levels cause reversals, Grimes proposes comparing price behavior at Fibonacci levels against behavior at randomly drawn lines. He found that the distributions of outcomes are statistically indistinguishable, indicating the levels are not special.

  6. 19:04 – 26:00

    Mean Reversion in Stock Indexes vs. Continuation in FX

    Grimes notes that stock indexes exhibit a natural tendency toward mean reversion, whereas interest rate-sensitive instruments and FX tend toward continuation. This distinction affects the probability of profitability when buying after significant downward moves.

  7. 26:00 – 28:10

    The Utility of the Wyckoff Framework

    Grimes finds the Wyckoff framework useful for discretionary traders as a narrative tool to understand institutional intent, such as accumulation. He views it as a helpful story to guide decisions, even if the specific labels are not statistically proven edges.

  8. 28:10 – 29:31

    Deception in Order Flow

    The discussion touches on how large traders and institutions may deceive retail traders reading the tape. Grimes mentions having been on both sides of this deception, implying that order flow contains significant noise and intentional manipulation.

  9. 29:31 – 30:03

    Order Flow, Deception, and Tape Reading

    Speaker 1 discusses how market positioning and lack of selling pressure can signal moves, noting that quietness at highs is informative. He explains that while order flow provides information, it is heavily deceptive due to tactics like iceberg orders and spoofing used by large traders to hide intent.

  10. 30:03 – 50:02

    Timeless Market Forces: Momentum and Mean Reversion

    Speaker 0 asks if momentum and mean reversion are timeless forces that will persist. Speaker 1 agrees, stating that while specific patterns may decay, the underlying forces of momentum and mean reversion are fundamental to functioning financial markets.

  11. 50:02 – 54:07

    The Illusion of Confluence and Correlation

    Speaker 1 argues that stacking correlated technical indicators (like Fibonacci, trendlines, and EMAs) does not increase probability because they measure the same underlying price structure. He contrasts this with quantitative approaches that require uncorrelated alphas to be averaged together for a stronger signal.

  12. 54:07 – 56:21

    Invalidation and Urgency in Trade Management

    Speaker 1 defines invalidation not just by price levels but by lack of urgency or participation. He prefers setups where the market shows immediate follow-through and dislikes consolidations with ragged upper boundaries (rejections), viewing a lack of urgency as a sign to kill the trade.

  13. 56:21 – 58:33

    Trade Management and Time Decay

    Speaker_1 discusses exiting positions when time decay occurs without invalidation, noting that daily timeframe trading has become less profitable for holding positions compared to the past three years. Speaker_0 asks about exiting based on time elapsed rather than price action.

  14. 58:33 – 1:31:39

    Market Probing and Liquidity Rejection

    Speaker_0 describes an edge involving price sweeping previous highs/lows and rejecting, inviting breakout traders and triggering stops. Speaker_1 confirms this as a fundamental market dynamic where the market probes for real interest and liquidity.

Assets discussed

AssetKindStanceContext
Crude oilcommodityneutralIf you think you're trading oil on fundamentals, maybe you are.

“Stance” is how the speakers talked about the asset, not a recommendation.

Claims worth verifying

Grimes has analyzed price data dating back to the Phoenicians and medieval Europe.

Unidentified speaker (speaker_1) ·0:14

How to verify: Verify via Adam Grimes' book 'The Art and Science of Technical Analysis' or his institutional research publications which cite historical data sources. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Jane Street is co-located and competes on nanosecond timeframes using copper cables.

Unidentified speaker (speaker_1) ·0:34

How to verify: Verify via Jane Street's public disclosures or industry reports on high-frequency trading infrastructure and co-location services. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Grimes is a former trader at the New York Mercantile Exchange and SMB Capital.

Unidentified speaker (speaker_0) ·0:42

How to verify: Verify via Adam Grimes' LinkedIn profile or professional biography on his website. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Fibonacci levels, moving averages, and VWAP are not special and perform like random lines.

Unidentified speaker (speaker_1) ·1:14

How to verify: Verify via Adam Grimes' published quantitative research or backtesting results available on his website or in his books. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

The optimal holding period for retail traders is two to five days.

Unidentified speaker (speaker_1) ·7:45

How to verify: Verify via Adam Grimes' research on alpha decay and retail trader performance statistics. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Stock indexes have a natural tendency for mean reversion.

Unidentified speaker (speaker_1) ·19:04

How to verify: Verify via academic finance literature on mean reversion in equity indices or Grimes' published data. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

FX and bond futures exhibit more continuation than mean reversion.

Unidentified speaker (speaker_1) ·19:17

How to verify: Verify via academic finance literature on momentum and continuation in currency and bond markets. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Grimes believes the market is at a historical inflection point comparable to the internet.

Unidentified speaker (speaker_1) ·11:07

How to verify: Verify via the podcast transcript or Grimes' public commentary on market structure changes. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

The 200-day moving average is not magical but captures volatility regimes.

Unidentified speaker (speaker_1) ·21:14

How to verify: Verify via Adam Grimes' quantitative tests comparing 200-day MA performance against other volatility proxies. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker 1 was part of the second generation of traders using volume profile, which was originally pitched to floor traders in Chicago.

Unidentified speaker (UNKNOWN) ·32:41

How to verify: Check historical accounts of volume profile adoption in Chicago trading floors; verify Adam Grimes' biography or publications for his stated role in volume profile development. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker 1 has personally used deceptive order flow tactics, such as iceberg orders and fake sell orders, to execute large volumes in thin stocks.

Unidentified speaker (speaker_1) ·33:40

How to verify: Verify via Adam Grimes' book 'The Art and Science of Technical Analysis' or his interviews where he discusses institutional execution tactics and market microstructure. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker 0 has been trading EUR/USD and GBP/USD exclusively for six years.

Unidentified speaker (speaker_0) ·38:42

How to verify: Verify via Speaker 0's (Waqar Asim) public trading journals, social media posts, or website content from the period between 2026 and 2026. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker 0's win rate floats between 35% and 55% with 1:3 risk-reward ratios.

Unidentified speaker (speaker_1) ·39:16

How to verify: Verify via Speaker 0's published trade records, backtests, or performance reports shared on his platform or social media. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker 1 recently had his 50th birthday.

Unidentified speaker (speaker_1) ·44:44

How to verify: Verify via Adam Grimes' public biography, LinkedIn profile, or social media posts indicating his birth year (approx. 2026-2026 given the 2026 transcript date). For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

The market has changed in the past three years such that it is no longer paying to be sticky on the daily timeframe.

Unidentified speaker (speaker_1) ·57:42

How to verify: Backtest daily timeframe mean reversion strategies over the last 3 years (2026-2026) and compare performance to prior periods using historical price data. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker_0 made 30% of his profits in the last year from trades where he was directionally wrong but broke even fast and let the rest run.

Unidentified speaker (speaker_0) ·1:03:26

How to verify: Review Speaker_0's trade journal or performance records from the last year to calculate the percentage of profit derived from trades with incorrect initial direction. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker_1 uses MATLAB, Python, and R for his quantitative framework.

Unidentified speaker (speaker_1) ·1:14:10

How to verify: Check Speaker_1's public code repositories, professional bio, or published papers for mentions of these specific programming languages in his trading infrastructure. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker_1 has zero confidence in AI answering questions about trade stats or edges.

Unidentified speaker (speaker_1) ·1:14:10

How to verify: Review Speaker_1's public statements, interviews, or social media posts regarding AI capabilities in trading to confirm his stance on AI for ideation vs. execution. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Mean reversion traders need to take profits very quickly, while momentum traders can be stickier.

Unidentified speaker (speaker_1) ·1:04:47

How to verify: Backtest mean reversion vs. momentum strategies to compare average holding periods and profit-taking speeds across different market regimes. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker_1 traded the same way for decades, focusing on basic structure and trade management.

Unidentified speaker (UNKNOWN) ·56:49

How to verify: Review Speaker_1's published books (e.g., 'The Art and Science of Technical Analysis') and compare the strategies described in earlier editions to later editions or recent interviews. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

The intraday low of the British Pound did not hold for years.

Unidentified speaker (speaker_1) ·1:05:23

How to verify: Analyze historical price data for the British Pound to check if any intraday low served as a long-term support level for multiple years. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker_1's quantitative framework was written in MATLAB, Python, and R.

Unidentified speaker (speaker_1) ·1:14:10

How to verify: Inspect Speaker_1's GitHub repositories or professional disclosures for code samples or documentation listing these languages. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker_0 trades between M15 and M1 timeframes.

Unidentified speaker (speaker_1) ·1:10:35

How to verify: Review Speaker_0's trading logs or public disclosures of his trading style to confirm the timeframes he primarily uses for execution. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Speaker_1 visualizes higher timeframes while trading on one time frame.

Unidentified speaker (UNKNOWN) ·1:10:48

How to verify: Review Speaker_1's interviews or books for descriptions of his charting setup and how he incorporates multi-timeframe analysis into his single-chart trading. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Testing random levels will yield some that appear to work due to non-random action and statistical probability.

Unidentified speaker (speaker_1) ·1:24:37

How to verify: Statistical analysis of multiple hypothesis testing; verify by running backtests on random price levels and observing the distribution of p-values. For recent events, verify 2026-08-17 through 2026-11-17. Use another period only when the speaker explicitly names it.

Do your own research

  • Verify the statistical significance of Fibonacci levels by comparing reversal rates at these levels against random intervals using historical price data.
  • Review Andrew Lowe's ecosystem metaphor and Grimes' claims about retail vs. institutional infrastructure advantages in trading timeframes.
  • Analyze the correlation between stock index mean reversion and FX continuation behavior using historical volatility and trend data.
  • Investigate the prevalence of iceberg orders and spoofing in current market microstructure reports from major exchanges.
  • Evaluate the 'timeframe justification system' risk by backtesting trades held on lower timeframes with higher timeframe invalidation levels.
  • Check the performance of dynamic risk management techniques that adjust exposure based on recent win rates versus static risk models.
  • Review academic literature on alpha decay to understand how published trading strategies degrade over time.
  • Test the validity of using AI for code generation in quantitative frameworks versus direct statistical analysis, as advised by Grimes.
Full transcriptexpand
0:00
Unidentified speaker (speaker_0): This quant traded orders big enough to move the market, spent decades researching 700 years of price data, and sold what he found to some of the biggest hedge funds on Wall Street. And today you get it here for free.
0:14
Unidentified speaker (speaker_1): One thing that makes my work almost unique is I've looked at the full scale of available data in financial markets all the way back to the Phoenicians. The way that I trade, that's a fundamental principle that has existed as long as we've had financial. markets. On the other hand, if I were to decide to compete with Jane Street, now I have a problem because the length of a copper
0:34
Unidentified speaker (speaker_1): cable, the speed of light through that cable, matters. They're co-located. I'm at home. I brought the wrong tool for the job, and I can't compete on that timeframe.
0:42
Unidentified speaker (speaker_0): Introducing Adam Grimes, former New York Mercantile Exchange and SMB Capital Trader, author of The Art and Science of Technical Analysis. And he's written and published institutional research every day for nearly a decade. People
0:56
Unidentified speaker (speaker_1): have been looking at order flow ever since markets went electronic, and surely there's information there. But there's also a tremendous amount of deception on the tape. I've been on both sides of that. I've been reading the tape, and I've also been the guy trying to screw the people reading the tape. I know the tricks to do that.
1:11
Unidentified speaker (speaker_0): What does that mean, the second part?
1:12
Unidentified speaker (speaker_1): Well, it means that
1:14
Unidentified speaker (speaker_1): let's talk about Fibonacci specifically. Does the market reverse at that level, first of all? That's the question. How would we test that? We draw a bunch of random lines on the chart and we test those random lines against the Fibonacci levels. They should look different. What I have found overwhelmingly, Fibonacci levels are not special. Moving averages are not special.
1:34
Unidentified speaker (speaker_1): VWAP is not special. They just should look different. And if they don't look different, I don't believe that something can be really significant in the market and also invisible in the data.
1:44
Unidentified speaker (speaker_0): In this episode, Adam reveals exactly what those tricks are: the one holding period where his data says a retail trader has the best chance of winning, and why stacking six confluences on your entry just makes you more confident, not more right. We started off talking about correlation versus causation, signal versus noise, the evidence. evolution and alpha decay and where the world is
2:04
Unidentified speaker (speaker_0): headed. I feel like maybe timeless things would be momentum, volatility, trend, and direction. Would this stand the test of time and therefore be better places to build an edge around as opposed to when you open up a technical analysis book and what you may find in the
2:17
Unidentified speaker (speaker_1): I think I have pretty good data on this. Um
2:24
Unidentified speaker (speaker_0): Adam, so I've been on a US tour right now and I'm I want to say pretty confused. And the reason for that, and I want to look for greater truths with yourself. We went for dinner last night, and you strike me as a very introspective and philosophical person. And then the pursuits for today's conversation is to find the truth when I find successful traders on both sides uh of of the fence, let's say.
2:44
Unidentified speaker (UNKNOWN): I
2:44
Unidentified speaker (speaker_0): have traders that tell me, I look, mean, the market is super complicated because you have the smartest people in the world with the best uh resources and money. and everything to like carve out an edge. And now it's to the point where you have Jane Street executing in nanoseconds, and it's really difficult, and you're always subject to alpha decay, so you've got to move fast. Then you have the other camp, which is also verified, made profits, which is
3:03
Unidentified speaker (speaker_0): timeless principles, keep it simple, a trend line and a doji and a trend is enough. Uh and they're they're on opposite sides. So where would you lie? And
3:13
Unidentified speaker (UNKNOWN): yeah,
3:13
Unidentified speaker (speaker_0): what is your beliefs on this topic? of alpha decay and just the evolution of the market? Or do timeless principles edge from the 90s can still exist today?
3:21
Unidentified speaker (speaker_1): Both of those things can be true, and I think they are. I think so for decades I've structured my trading around pullbacks and consolidations and trends. And you know, market makes a big move, and we know we can take this apart l later, you know, exactly what the forces are behind. it. Market makes a big move and then makes a pause.
3:41
Unidentified speaker (speaker_1): And then based on reading how the patterns develop, you find the scales tilt a little bit in the direction of a continued move in that initial direction. So that is a fundamental principle. And you said from the 90s, I would say much deeper. You know, if we go back
4:01
Unidentified speaker (speaker_1): the entire history of the of recorded price data. We go back to we have price data from Europe in the Middle Ages. We have barley prices from the 1300s. Uh, we have some of those early clay tablets the Phoenicians had were accounting tablets and records of prices from merchants. And academics have reconstructed Constructed
4:21
Unidentified speaker (speaker_1): those prices. And of course, you know, no nobody was keeping tick data then. They would uh not record daily prices, they would record when big changes happen. But I've inspected those price series. I think that one thing that makes my work almost unique is I've looked at the full scale of available
4:41
Unidentified speaker (speaker_1): data in financial markets all the way back to the Phoenicians. And I would tell you the way that I trade, looking for a market to make a big move, maybe get a little bit ahead of. itself, maybe consolidate, make a pullback or flag or whatever we're going to call it, and then make a continued move in that direction, that's a fundamental principle that has existed as long as
5:01
Unidentified speaker (speaker_1): we've had financial markets. On the other hand, uh, if I were to fire up my computer at home and decide to compute compete to compete with Jane Street, now I have a problem because the length of a copper cable. the speed of light through that cable matters. And they're co-located. I'm at home. I'm playing a
5:21
Unidentified speaker (speaker_1): game where I, you know, I brought the wrong tool for the job, and I can't compete on that timeframe. So I think both of these things are true. And I also think Alpha's decay. And I think that the specific structure that I built for decades around this pullback, uh, I think. the way that I was trading it 10,
5:41
Unidentified speaker (speaker_1): 15, 20 years ago, I think we're starting to see some shifts in the statistics. And I don't know, you know, I I I don't know exactly why the market is changing, but it does change.
5:57
Unidentified speaker (speaker_0): Quotes Where if I'm trying to compete against J Streets, Street, I'm never going to win. But if I go to the other approach, I have a better chance What would be a retail independent trader's best chance in the markets?
6:07
Unidentified speaker (speaker_1): I think I have pretty good data on this. Uh
6:11
Unidentified speaker (speaker_1): and it's not going to be an exciting, sexy answer. It's not going to be day trading. Uh, the best chance, I think, is a time frame where the fewest people are trying to extract money.
6:25
Unidentified speaker (UNKNOWN): The
6:25
Unidentified speaker (speaker_1): markets, you know, you use whatever metaphor you want. The markets are a battle, they're a war. I think they're an ecosystem. I didn't come up with that. Andrew Lowe talks about markets adapting and thinking about different market participants like predator and prey, and they evolve and they change over time. So if you're competing on a very short time
6:45
Unidentified speaker (speaker_1): frame, you're competing with people who have infrastructure, data. that you do not have. If you're competing on the longer term timeframe, you're competing with people that have access to fundamental data. If you think you're doing fundamentals at home, well, you're up against firms who are
7:05
Unidentified speaker (speaker_1): sending analysts to the companies. You know, they're spending time in the factories, they're talking to management, they know what whatever you think you know, you don't know what they know. And you don't have access to that information. If you think you're trading oil on fundamentals, maybe you are. But do you have access to the data feeds that show you where the tankers are in the ocean, how full
7:25
Unidentified speaker (speaker_1): they are? When they're do you do it is that the game you're going to compete? Is a trader at home? Probably not. I think a simple, concrete answer is looking for holds that are probably a little bit longer than you're going to be inclined for. So not day trades. You do need that overnight risk, but somewhere in the you know, let's say average holding period two to five days
7:45
Unidentified speaker (speaker_1): is probably a sweet spot for a retail trader.
7:48
Unidentified speaker (speaker_0): You mentioned how there are certain principles that you could trace back to hundreds of years. And I would suppose that common denominator between then and now is human emotion. Um greed, optimism, uh euphoria, then despair XYZ, which makes total sense. And that would be reflected naturally in price action until you go digital and until you go
8:07
Unidentified speaker (speaker_0): algorithmic. And now, if I'm not wrong, 85 or 90 percent of price movements or orders are algorithmically generated or executed.
8:17
Unidentified speaker (UNKNOWN): When
8:17
Unidentified speaker (speaker_0): I see the comments online, it's like, well, yeah, but humans coded it, so it's it the emotions are a baked big in. difference. I'm not quite sure, but nonetheless, does human emotion reflect when the majority of the market is algorithmic?
8:28
Unidentified speaker (speaker_1): So far, yes. So, you know, we're recording at a point in time and the markets may evolve very quickly. And I may say things today that look very foolish five years from now. So you know, I will reserve the right to say that I was wrong. And sure I'm gonna be wrong in some things, but I think so. I think this argument that people make that algorithms
8:48
Unidentified speaker (speaker_1): so first of all, many algorithms are execution algorithms. That's that's all they are. You know, people think like there's some algo that determines price. No, you know, like if if I and you know, I've been this trader where if I need to execute a lot relative to the volume in a market, I will either farm it to a trader
9:08
Unidentified speaker (speaker_1): who can do it algorithmically or to an algo. And that algo is gonna, you know, it's gonna be benchmarked against VWAP over the time. period. But it's the algorithm is just figuring out how to nibble my order without really moving the market because you know if I push the button, the the market moves. Many algorithms are just that. And what they're
9:28
Unidentified speaker (speaker_1): looking to do is buy when the market is, and you know, can we can flip It for selling, but you know, they're they're looking to extract some minimal value, and it's not something that's completely disconnected from the way that humans trade, the way that humans think. And the uh opportunistic algorithms, if you will,
9:48
Unidentified speaker (speaker_1): well, what are they looking for? They're looking for extr in most cases, extremes and some kind of mean reversion. That's by and large. So what are they looking for? How do did we get to those extremes? Either error or emotion, right? And in financial markets, maybe emotion is error. I don't know. Um
10:06
Unidentified speaker (speaker_1): so I
10:08
Unidentified speaker (speaker_1): we are seeing, I'm seeing in my quantitative work, uh, some alphas decaying. I'm seeing the markets change in some fundamental ways. It may be exactly what you're asking about. It may be uh new kinds of AI driving market action. But I think over the, you know, as the world's been more and more digital, going
10:28
Unidentified speaker (speaker_1): back to the late 80s, I think at least up to this point where we're having this conversation, I think the idea of emotion is still valid.
10:36
Unidentified speaker (speaker_0): It's interesting. You mentioned before, if we reached out to you and said, hey, let's do a podcast, you would have
10:42
Unidentified speaker (UNKNOWN): said,
10:43
Unidentified speaker (speaker_0): no, because I've got a system, it's working, I'm making money, and that would be a boring conversation.
10:46
Unidentified speaker (UNKNOWN): It's not that
10:47
Unidentified speaker (speaker_1): interesting, yeah.
10:47
Unidentified speaker (speaker_0): Which sounds more interesting to me. But now we're at a time where you're saying, well it seems like we're at an inflection point in history where you're sensing change, you're sensing decay, and you're seeing the emergence of quantitative power that may be reshaping the landscape. And you're saying my beliefs today might look foolish in five years. Are we at a inflection point in history right now in the markets?
11:07
Unidentified speaker (speaker_1): I believe we are at an inflection point in history. Full stop. Uh I suspect
11:14
Unidentified speaker (UNKNOWN): you know,
11:15
Unidentified speaker (speaker_1): in my lifetime, I saw the arrival of the internet, which changed everything, the internet and email. And I know generations before us saw electricity, industrial revolution. We've seen these major inflection points. I suspect today where it's something that
11:32
Unidentified speaker (UNKNOWN): we
11:32
Unidentified speaker (speaker_1): haven't like this is maybe invention of writing, this is maybe agriculture. This this could be one. of those civilization level inflections. And I think there are concerns outside of I have many concerns outside of financial markets, how we navigate this as a species and as a culture and and
11:52
Unidentified speaker (speaker_1): as humans.
11:53
Unidentified speaker (speaker_0): With the objective, I guess, today of understanding truths of the markets.
11:57
Unidentified speaker (UNKNOWN): I'd
11:57
Unidentified speaker (speaker_0): like to know where I should begin because it seems like you've had a journey in your career of styles of trading, what you use from day trading. M1 to where you are now.
12:06
Unidentified speaker (UNKNOWN): A
12:06
Unidentified speaker (speaker_0): lot of traders begin at strategies and and that would be like broad schools of thought, whereas Elliot Wave Theory, Wickoff theory, uh Smart Money Concepts and Fibonacci support resistance, and then you'd try and put that to fit into the market. The other approach would be, uh, though I guess more algorithmic approach, which is past price could predict future price, and as a result, I'm
12:26
Unidentified speaker (speaker_0): just gonna go through data to find things. Most people start at the other side, which is schools of thoughts, because it's it feels more approachable. Uh, but nonetheless, the ones that do or don't work, when I think of, let's say, a Fibonacci, is there something magical about a Fibonacci level that ends up as as a reflect uh uh rejection point? Or is it kind of like a bell curve and
12:46
Unidentified speaker (speaker_0): you have these areas of increased probability? I don't know. Uh but what I'm I more concerned by myself with is this a correlation or causation? Does the market reser reverse because of the Fibonacci or this nicely reflects a reversal point? If I could start to think in first principles and start to thinking causation, where would I begin?
13:04
Unidentified speaker (speaker_1): So
13:05
Unidentified speaker (UNKNOWN): th
13:06
Unidentified speaker (speaker_1): the way you framed that, and this is a n another way to rephrase this is the self self-fulfilling fulfilling prophecy. Is the Fibonacci level, moving average VWAP, whatever we're talking about, a self-fulfilling prophecy? Uh my answer to that, I do care deeply about the truth. We're going to come back to that. But my answer to is it a self-fulfilling
13:25
Unidentified speaker (speaker_1): prophecy? Is I don't think I care. What I want to know, first of all, is, and let's talk about Fibonacci specifically. Uh, I want to know, does the market reverse at that level, first of all? That's the question. So how would we test that?
13:43
Unidentified speaker (speaker_1): How how it how it it
13:45
Unidentified speaker (UNKNOWN): it's not
13:45
Unidentified speaker (speaker_1): so easy. How how would we structure a test?
13:47
Unidentified speaker (UNKNOWN): Well,
13:47
Unidentified speaker (speaker_0): my first concern would be how do I separate let's say a broken clock? Why my watch is stuck at six o'clock, it's correct twice a day, six a.m., six PM. The rest of the time is wrong. But the times is correct. If I give if I attribute meaning to it, I'm attributing uh meaning to noise. So I'd have to separate signal from noise.
14:04
Unidentified speaker (UNKNOWN): I'll
14:05
Unidentified speaker (speaker_0): I'll pass it back to you and help me navigate that one.
14:07
Unidentified speaker (speaker_1): Well, there's also and then there's also the
14:11
Unidentified speaker (speaker_1): cognitive issue that you're going to remember things that look and feel significant. You know, if if you look at your watch and it says six o'clock, zero, zero, and then you look at it again, it says seven o'clock, you're gonna be like, oh, two zero, you know, I looked at my clock. So you're it and you know that's meaningless. but it's uh it's a data
14:28
Unidentified speaker (UNKNOWN): point
14:28
Unidentified speaker (speaker_1): that jumps out. Uh so the question is
14:33
Unidentified speaker (speaker_1): if I understand what you're saying, I have a Fibonacci level. First of all, let's define what we think should happen. Price should reverse more from the Fibonacci level. Is that is that what we're saying? Uh because we do kind of need to think around the corners because that's one possibility. Another possibility is maybe price is drawn to the
14:53
Unidentified speaker (speaker_1): Fibonacci level. Maybe price engages like, you know, that could be another way to think about it. But regardless, both of those, your question is, how do I know it's not the stopped clock being right twice a day? Right. There's a very simple technique for that, which is we draw a bunch of random lines on the chart and
15:13
Unidentified speaker (speaker_1): we test those random lines.
15:15
Unidentified speaker (speaker_1): against the Fibonacci levels. And this is a, you know, it's funny. Uh, I didn't have a quantitative background. And I didn't know how to test any of this stuff. When I when I started trading out of college, I was a musician. I had no math classes in my pocket. Uh so I had to teach
15:35
Unidentified speaker (speaker_1): myself everything. And then,
15:37
Unidentified speaker (UNKNOWN): you
15:37
Unidentified speaker (speaker_1): know, initially, you get very excited because you see things that make money. Uh, for instance, it's quite easy to develop a system that buys stocks that makes money because stocks on average go up over time, right? So, in fact, if you develop a system where you're buying stocks and doesn't make money, that's maybe very interesting to me today because maybe you
15:57
Unidentified speaker (speaker_1): found an edge. But the question is, am I just getting the index drift or am or am I getting am I getting alpha? You
16:04
Unidentified speaker (UNKNOWN): know,
16:05
Unidentified speaker (UNKNOWN): we know what we're
16:06
Unidentified speaker (speaker_1): talking about. Are we getting alpha? So what you can do is test against these dummy levels. And I wouldn't even, you know, with the Fibonacci level, I wouldn't even say,
16:18
Unidentified speaker (UNKNOWN): I
16:19
Unidentified speaker (speaker_1): wouldn't even define the bounce. It would be enough for me initially to say, I have my interactions with Fibonacci levels. And what happens after? those. I would go through and I would look every time we touch a level, and then you know, you you ask me, well, do you really need to touch it? And I say, okay, we can put a little band around it, whatever. Every time we get close to a level,
16:39
Unidentified speaker (speaker_1): uh, let's look what happens going forward. And then because we're doing a lot of tests, let's also test exact engagements and we have a bunch of levels and you know, should we test 50%? It's not really Fibonacci level, but the Fibonacci traders use it. We should talk about Fibonacci and sacred geometry as this entirely separate topic. But sticking on the test now,
16:59
Unidentified speaker (speaker_1): so we look at what happens, we look at the outcomes from those engagements, and then we look at the outcomes from the random lines. And they should look different. And of course, you know, one one of the we have to use the right kind of statistics because if you look closely enough, you'll see differences where there aren't differences. So we have to understand the
17:18
Unidentified speaker (speaker_1): variability. But what I have found overwhelmingly, and this is uh the the this is where people probably start throwing things at the screen or or spiking their phone on the ground, uh, all of these things that people think work. Fibonacci Bonacci levels are not special. Moving averages are not special, VWAP is not special.
17:39
Unidentified speaker (speaker_1): Simple test of a moving average, by the way. You want to tell me a 50 period or 200 period moving average is special. Fine. Let's test your 200 period. Test it against 210 and 109. Yeah. And again, they just should look different. And if they don't look different, I don't believe that something can be really significant in
17:58
Unidentified speaker (speaker_1): the market and also invisible in the data. It might be hard to see. It might be subtle. It might be slight and it might not show up. You know, I don't know. Like you, you told me you think price bounces off a Fibonacci level. Maybe it does. Maybe it accelerates through a level. Our test would have found that. Maybe after the level, volatility gets
18:18
Unidentified speaker (speaker_1): smaller. Our test would have seen that because we're looking, And all we're saying is, you know, we're saying, I don't know. I'm very naive. I have these two sets of distributions. which are the you know forward distributions. Let me look and see if they look different. So that's the test.
18:32
Unidentified speaker (speaker_0): I'd love to express this conversation towards it. Seems like you've done this test that you're describing to me on Fibonacci, and you've probably done it on everything else, trend lines
18:40
Unidentified speaker (UNKNOWN): and
18:41
Unidentified speaker (speaker_0): supporters, X, Y, Z.
18:42
Unidentified speaker (UNKNOWN): What
18:43
Unidentified speaker (speaker_0): conclusions did you derive?
18:44
Unidentified speaker (speaker_1): Uh there are no, with very few exceptions. And you want to know, I'll I'll tell you where some of them are. With very few exceptions, there are no special levels. Uh in general, the from a quantitative perspective, the and it it varies according to asset class a little bit. Uh for instance, stock
19:04
Unidentified speaker (speaker_1): indexes have a natural tendency for mean reversion.
19:08
Unidentified speaker (UNKNOWN): All
19:08
Unidentified speaker (speaker_1): markets have mean reversion. Stock indexes tend to have a little bit more. Interest rate sensitive instruments, FX, you know, bond futures,
19:17
Unidentified speaker (speaker_1): less mean reversion, more continuation. So if I had to, if if you told me uh a market went down a lot today
19:26
Unidentified speaker (UNKNOWN): and
19:26
Unidentified speaker (speaker_1): you're going to buy it on the close, if you told me it's a stock index or it's a currency, you have a different probability of having a profitable trade. You have a much better chance of making money on the stock index. So one thing we will see.
19:39
Unidentified speaker (speaker_1): is if you're looking at moves down to moving averages in stock indexes, they naturally bounce. But guess what? So do our random lines too. You also can, I've run tests where, you know, I said we should test the 200 against the 210 and the 190. Uh We also should test it against a move,
19:59
Unidentified speaker (speaker_1): how about a moving average that changes its period every every bar? So today it's a ninety period. Tomorrow it's a hundred and thirty hundred and thirty period. So it's it's a meaningless line on the chart.
20:18
Unidentified speaker (speaker_0): He's like, Yeah, one of my models is like once we're above the the two hundred moving average, if we have three down days, uh I'm gonna do X. I'm like, anything else? Okay, no, that's it. And then I speak to another person, he's like, Yeah, I just wait for a uh EMA cross, I wait for a trend line tap, or I wait for a breakout. I was like, oh, do you wait for the test? Do you use a fibology? Do you use a do like, no, just a
20:38
Unidentified speaker (speaker_0): breakout. Anything more? Nothing. And I and I then I have the other side of the camp and we have really deep technical conversations down to M1, More down to the philosophical side that we're touching upon. How can these both coexist? Which is back to the original question. Okay.
20:51
Unidentified speaker (speaker_1): they can both coexist in a few ways. So
20:54
Unidentified speaker (speaker_1): going back to the general behavior of stock markets, when stock markets are significantly off their highs, which would put you under the 200-day moving average. And the 200-day moving average is not a magical line in the sand, but in general, with stock indexes or individual stocks, if you're under the 200-day 200 day moving average, you're going to see a different kind of behavior. You're going to
21:14
Unidentified speaker (speaker_1): see a
21:16
Unidentified speaker (UNKNOWN): in
21:17
Unidentified speaker (speaker_1): the more recent data, you see more volatility. And this again, this is not true in currencies because there's not really an up and down. It depends, you know, what you're weighing it against.
21:29
Unidentified speaker (speaker_1): But with stocks
21:31
Unidentified speaker (UNKNOWN): under
21:31
Unidentified speaker (speaker_1): the 200 day moving average, you will see different kinds of behavior. Doesn't, it doesn't mean the 200 200-day day moving average itself is magical. but it does mean that when you're off of that high, and maybe uh,
21:43
Unidentified speaker (UNKNOWN): you
21:43
Unidentified speaker (speaker_1): know, if you're using in a systematic approach, you're using the 200-day moving average. Well, you have to use something. And you're capturing this aspect of behavior. Maybe it could be a 200-day median, which is actually quite comp, you know, computationally is a It's little bit more expensive to compute every day. Um, maybe you could just look at where you are in the yearly range. or something like
22:03
Unidentified speaker (speaker_1): that, or how far you you're going to get a similar effect.
22:07
Unidentified speaker (speaker_0): This is down to the self-fulfilling idea.
22:11
Unidentified speaker (speaker_0): But generally people would be behaving in a certain way below it.
22:14
Unidentified speaker (speaker_1): It's not it's not even that. You know, I I would be perfectly happy. Our test of Fibonacci levels, which I promise you shows no difference with the with the Fibonacci level, if they were self-fulfilling prophecy, we would see the difference. We would see the action there. And then I would be quite comfortable telling you, I don't really care why it's
22:34
Unidentified speaker (speaker_1): I don't I can't know why it's doing that, but it does that. And I see it in the data. I don't even see it in the data. If the 200-day, if funds were buying at the 50-day or the 200-day Day or you know, the nonsense that people peop that people say, you would see it at those averages. You do not see it at those averages. Now that can be true, and it can also be
22:54
Unidentified speaker (speaker_1): true that when stock indexes are significantly off their highs, you see a different kind of behavior, primarily with respect to volatility. I don't think there is a strong directional edge, uh, but there is uh you know, you'll you'll have more volatile events under the historically at least under the tender day moving out. now.
23:13
Unidentified speaker (speaker_0): So what I've personally done uh I'm with you. I I'm on the same page and I like to avoid looking at things that are correlated, like a head and shorter pattern. I don't know what that means and it could work but it is signal versus noise. I don't know if it's statistically relevant. Same for Bonaji, same for trend line support resistance XY list goes on. But I do like to think if in terms of causation. What could cause
23:31
Unidentified speaker (UNKNOWN): the
23:31
Unidentified speaker (speaker_0): market to move? And I find it
23:34
Unidentified speaker (speaker_0): as as as time goes on or as I'm projecting into further time horizons, uh I guess the crystal ball gets a bit hazy.
23:40
Unidentified speaker (UNKNOWN): So
23:40
Unidentified speaker (speaker_0): I think it it becomes harder to predict. Naturally, as a result of that, I'm doing these days or for the last decade, what you were initially doing and have moved away from, which is I like to be on the M1, I like to be on shorter time horizons, because I think the causation uh edge exists there. And what in the language I think of is supply and demand surely has to be causative. Market impact
24:00
Unidentified speaker (speaker_0): of large institutions can create a small causative event. Now, I don't know if that's going to lead to a full move, but it might give a reaction. Right. So I like to think in these terms. Am I on a better line than or a better direction in the industry than someone who was looking at Fibonacci and trend lines and these generic things?
24:16
Unidentified speaker (speaker_1): I think so. I mean, if you're trading on the one minute chart, uh one of the other things that we'll will test out well, which you're probably aware of previous days high and low, that's a real thing.
24:27
Unidentified speaker (speaker_1): And if we
24:30
Unidentified speaker (speaker_1): you can test this very easily. You can do the tests we were talking about, look at engagements of the high and low, look at random lines, throw your Fibonacci levels in there. And what you're going to see is that one of these things is different than the other. The high and low looks different than the Fibonacci and looks, you know, the the it's it non-random behavior there. So I I
24:51
Unidentified speaker (speaker_1): think you're correct there. I, you know, as humans, we're, we tell stories, right? We're we're ries. We're we're our
24:58
Unidentified speaker (speaker_1): culture is tied, our all of our cultures are tied together by the stories we tell. And we tell these stories about ourselves. And we're telling stories here today. And there's something that's very appealing about that. And as somebody who's done a decent amount of spiritual exploration stories are important to me. But as a
25:19
Unidentified speaker (speaker_1): quantitative trader, I know the market is, you know, I the story I would tell you about the market is the market is uh very fickle. The market, the the market can change, uh, can can can be making me feel one way, and then it can change very quickly. So I know that I can't get very attached to my stories.
25:39
Unidentified speaker (speaker_1): Uh I,
25:41
Unidentified speaker (speaker_1): you know, there's also at any point in the market, there's so many different influences and different people trading that I don't think I have,
25:50
Unidentified speaker (speaker_1): you know, in full humility, I don't think I understand I can understand those influences. But I will also tell you, I find the Wycoff framework very useful. I know it's naive and you know for the viewers. uh, where you think about these big buyers and big sellers driving the market. They may not even be real, but you can think
26:10
Unidentified speaker (speaker_1): what are buyers trying to do at this point? What are sellers trying to do at this point? And if you use those frameworks, even if these stories are completely fictitious, we are still saying something about the reality that has something, some, and you
26:26
Unidentified speaker (UNKNOWN): know,
26:26
Unidentified speaker (speaker_1): the story can be useful.
26:27
Unidentified speaker (speaker_1): Like when uh when people in a previous culture told a story about a goddess that went into the underworld, so all the plants died in the winter, and then you would know when to replant.
26:39
Unidentified speaker (UNKNOWN): I don't
26:39
Unidentified speaker (speaker_1): know if they thought that was real or not, but it guided their decisions for how they planned to get through the winter and when the spring came. And I think this this goes back to the question you asked before. Should you start, where should you start? Should you start analyzing data quantitatively, or should you start kind of trying to wrap your arms around the
26:59
Unidentified speaker (speaker_1): market? I would say start trying to wrap your arms around the market. But you know, that's my bias because I think at heart, despite all the quantitative work, I'm an artist, I'm a musician. Uh, you know, I'm I'm I'm a weird person because I have this quantitative side. but I also have this holistic artistic approach. And
27:19
Unidentified speaker (speaker_1): I think study, study patterns, uh, trade, you know, like you learn, don't lose much money, keep keep your losses small, but you can do all of the theoretical trading, you know, reading and back testing. It and it all changes when you get in the market, the market punches you in the face. and you oh,
27:39
Unidentified speaker (speaker_1): okay, they have there there's something there.
27:41
Unidentified speaker (speaker_0): You mentioned the that Wyckoff could be a little bit naive, but yeah, even looking at it, comparing it to the other things we've spoken about, that one still makes sense to me. Maybe
27:49
Unidentified speaker (UNKNOWN): not the
27:49
Unidentified speaker (speaker_0): stories, maybe not
27:50
Unidentified speaker (UNKNOWN): the
27:50
Unidentified speaker (speaker_0): labels and the uh the UTAD and the L L P S Y and all these uh let's say acronyms we give to it. But the philosophy of uh institutional buyers loading up, that's
28:00
Unidentified speaker (UNKNOWN): a
28:00
Unidentified speaker (speaker_0): cause. It's like an elastic band and pulling it back. If I let go surely it should. Yeah.
28:04
Unidentified speaker (speaker_1): It's
28:04
Unidentified speaker (speaker_0): a cause and effect relationship. So that makes more sense to me. What
28:08
Unidentified speaker (UNKNOWN): are
28:08
Unidentified speaker (speaker_1): they trying to do in the deception in the market?
28:10
Unidentified speaker (UNKNOWN): Now,
28:10
Unidentified speaker (speaker_0): naturally, there's going to be a lot of nuance to it. Like every not every buyer is equal. Someone is a speculator, someone is a utilizer of the currency. Some people are trying to hold for long-term, someone's short-term, someone's weak-handed, XYZ. So it does get nuanced, of course. But the general principle of I want to trade in the market from causative perspectives. accumulation feels causative.
28:29
Unidentified speaker (speaker_1): Uh
28:29
Unidentified speaker (speaker_0): one of the things could feel causative, and again, avoiding just storytelling, which could be a guiding principle, but I don't want to get into that because that can um yeah, it can it can get
28:38
Unidentified speaker (UNKNOWN): a bit weird.
28:39
Unidentified speaker (UNKNOWN): It's a little
28:40
Unidentified speaker (speaker_1): weird. Um I think this framework is very useful too. And I think particularly as a discretionary trader, when you you do all of the analysis and it still comes down to you making a decision. And I find that my decisions are best motivated by understanding what
29:00
Unidentified speaker (speaker_1): I think these groups of people are trying to do. If I think I see, you know, okay, what buyers are you said accumulating here. And I think that's a word that in some way, you know, it's overused. But you know, if I keep seeing that the market keeps getting depressed on this time frame, at uh like there's buying there. There's some and how much do I trust it? I don't know. I can't really
29:20
Unidentified speaker (speaker_1): trust anything that much, but it's something that is happening. There's somebody there that is buying. And then, like you said, you know, what happens when you move into you move above a resistance level
29:31
Unidentified speaker (UNKNOWN): and
29:31
Unidentified speaker (speaker_1): all of a sudden there's not the selling pressure to meet that, and then the market moves. And you saw it because you saw the positioning beforehand, or you know. if you're living in an intraday world, you know, you see you keep coming back to the high of the day and you can't really back off from it. And you go dull and dead and quiet at the high of the day, that's telling you
29:51
Unidentified speaker (speaker_1): something. You know, the they're they're not letting it back off much. So buy and I I think that whether or not this is actually a thing, it is for a discretionary trader a powerful tool for understanding price action.
30:03
Unidentified speaker (speaker_0): We started off talking about correlation versus causation, storytelling versus signal versus noise, and then also with the evolution and alpha decay and where the world is headed. I feel like maybe timeless things would be momentum, uh volatility, uh trend and direction. These I suppose would always be baked in.
30:20
Unidentified speaker (speaker_1): Admin reversion. Admin
30:21
Unidentified speaker (speaker_0): I
30:21
Unidentified speaker (UNKNOWN): mean reversion,
30:21
Unidentified speaker (speaker_0): yes. W Would would this stand the test of time and therefore be better places to build an edge around? as opposed to when you open up a technical analysis book and what you may find in the
30:31
Unidentified speaker (speaker_1): I think so, but I think if you open up a good tech not, you know, not to talk my book, but if you open up my book, my book is going to be centered around momentum and mean reversion.
30:40
Unidentified speaker (UNKNOWN): Okay.
30:40
Unidentified speaker (speaker_1): And the patterns are an expression of that. So I think, you know, barring our weird like civilizational financial market ending discussion earlier, which was pretty weird. Uh, as long as there are functioning financial markets, these forces of mean reversion and momentum will continue to play. I can't imagine financial markets where they will not
31:00
Unidentified speaker (speaker_1): play out, but they may play out in ways that mean the specific patterns, the ways we trade them, don't necessarily work. So we may need, but I don't and there are some other principles, right? We should talk about options traders doing things with volatility.
31:15
Unidentified speaker (speaker_1): doing things with time. There are things with relative value. You know, some people will tell you that the only thing that matters is price, you know, only price pays. But there is a there are other ways to trade, right? Yes. You can set up volatility structures. Uh you
31:30
Unidentified speaker (speaker_1): these things may, it's kind of a level of sophistication up from trading chart patterns. It's gonna going be to there some somehow. way, you know. and we may I I could imagine a world where the chart patterns are so heavily arbit, but they don't arm the volatility out of relative relationships. So,
31:50
Unidentified speaker (speaker_1): you know, there's something to be done with spreads, perhaps.
31:53
Unidentified speaker (UNKNOWN): I don't know. We
31:54
Unidentified speaker (speaker_0): spoke about dinner last night, uh, that the topic of the industry, or the buzzword in the industry right now, seems to be this year is it's all about order flow in the region. retail lens. A lot of influences are popping up talking about it, and it just seems to be everyone's looking for holy grail, mine's not working, let me strategy hop
32:09
Unidentified speaker (UNKNOWN): and
32:10
Unidentified speaker (speaker_0): try this one out. However, it doesn't feel like the other ones in the past that I've seen where it's just storytelling and something repackaged. It feels like something that could be causal, where uh I'm looking for big orders to come in. This could be institutional footprints and naturally self-fulfilling. If a large institution steps in, price should at
32:26
Unidentified speaker (UNKNOWN): least
32:26
Unidentified speaker (speaker_0): move. a bit in that direction. Is there any merit to linking price action to order flow behind it?
32:33
Unidentified speaker (speaker_1): I think so, yes. But uh where I would where I would take issue with this is with it being new. Because this is this is something
32:41
Unidentified speaker (UNKNOWN): you know,
32:41
Unidentified speaker (speaker_1): this is this is not new. Uh volume profile, I I was kind of in the second generation of that. It would if I understand critics. correctly, it was first pitched to the floor traders in Chicago who weren't really able to make much with it. So then it moved to a retail marketing. And it's been repackaged in various
33:01
Unidentified speaker (speaker_1): ways. And then order flow, people have been looking at order flow ever since markets went electronic. And surely there's information there. Uh, I think people do things with order flow that's probably not appropriate and correct. You know, the idea that um you see somebody made a big transaction at a
33:21
Unidentified speaker (speaker_1): price, does that mean they're gonna defend it when it comes back? And I, you know, that's one is one of many ways that people think about it. So
33:28
Unidentified speaker (speaker_1): I
33:30
Unidentified speaker (speaker_1): I think particularly for scalpers, there's a lot of information there. But there's also a tremendous amount of deception on the tape. You know, you don't see Yes.
33:40
Unidentified speaker (speaker_1): And I know because I've been I've been on both sides of that. I've been reading the tape, and I've also been the guy trying to screw the people reading the tape when I'm had to execute a lot of volume in thin stocks. You know, I I I I know the tricks to do that.
33:53
Unidentified speaker (speaker_0): What does that mean, the second plots?
33:54
Unidentified speaker (speaker_1): Well, it means that so say say there's a very just for instance, there's a very thin stock, or it could be an option, it could be a currency at a quiet time of the day. And there's not much liquidity. The bid ask is very wide. It's not trading very much. And you need to do a big order right now. If you just go and start clearing the book, you're gonna move it
34:14
Unidentified speaker (speaker_1): a lot. And instead, what you might do, there's all kinds of tricks. You if you have to buy, you know, whatever quantity, let's say you have to buy 20,000, you don't have to put 20,000 in the book and let everybody see, hey, you know, I uh you you can put in 100, you can put in 200, you can show that, you can iceberg the rest. You can uh if you're looking to
34:34
Unidentified speaker (speaker_1): buy, you might put an order in to sell. You might put a big order into sell way above the market, and then people reading the tape will think, oh, there's actually selling pressure there. Uh you're, you know, you're I I've got it flipped, but you know, uh the the
34:50
Unidentified speaker (speaker_1): so a lot of what you see on the tape is by nature deceptive. I don't think you can take it. Uh and then there's of course the uh argument for looking at a difference between the display, what you see displayed and what you see actually transacted, that you know the transactions are a level of reality that the displayed theoretical liquidity is not.
35:10
Unidentified speaker (speaker_1): There's something there. Uh I I think that the way that I read charts is
35:17
Unidentified speaker (speaker_1): effectively a kind of tape reading on the chart.
35:21
Unidentified speaker (UNKNOWN): You
35:21
Unidentified speaker (speaker_1): know, and I I think you probably, if you're reading a one-minute chart, I think, for instance, go back to that situation where you're up against the high of the day. And imagine that you can't really back off from the high of the day. And a few times that it's sold off, it leaves long, long wicks, long tails. You know that. those were probably just quick probes
35:41
Unidentified speaker (speaker_1): down. So you're reading the tape as you're reading the chart. And the argument the tape readers will make, well, you know, the chart's static. You're not seeing what's actually on the tape, but it's leaving the footprint in the same way on the chart sometimes, if you know how to read it.
35:56
Unidentified speaker (speaker_0): So what you're basically describing here is essentially what I do. And I don't want to make the a conversation just about my my beliefs, um, but that's where I found. as a retail trader some edge. And when I speak to others uh that are able to go into the wise like you are with me today, uh it seems to be the common answer is as a retail trader, the edges seem to lie on the lower time frames, on an intraday perspective, or scalping. Then at the
36:16
Unidentified speaker (speaker_0): beginning, you said day trading is uh is the wrong place to begin, or you don't like it.
36:20
Unidentified speaker (speaker_1): I
36:21
Unidentified speaker (speaker_1): I have seen more traders have more success on slightly longer time frames. Uh Um and it it it is interesting that you know we navigated around to basically what you do. Like, you know, you you didn't tell me that. I didn't know that it just just by talking about where the edges are, you have found it in practice. And you know, I found it in practice and and
36:41
Unidentified speaker (speaker_1): theory too. So you you are well aligned. But uh
36:46
Unidentified speaker (speaker_1): for me,
36:49
Unidentified speaker (UNKNOWN): there
36:49
Unidentified speaker (speaker_1): was a point where I was ex I was beyond obsessed with day trading. I would trade the SPs, I would do a little bit of music, I would, and then I would trade Europe at night, and I would get up and I don't know how I slept. Like, you know, I add up the days, the the hours I and it adds up to more than 24 hours. And I did that for a while. And then I realized, you know, it's not really how I want to be a
37:09
Unidentified speaker (speaker_1): human. We're here, we don't know how long. Like a plane could fall through this. right now while we're talking, right? It could happen. Um
37:18
Unidentified speaker (UNKNOWN): you
37:18
Unidentified speaker (speaker_1): you never know the future. I don't want to spend my entire life glued to the screen. And if I'm going to day trade well and every now and then I get I kind of get a thought, you know, maybe I'd like to day trade again. And my business partner, my wife are like, okay, we we've we've seen this before. Go ahead. Go ahead. Come back and tell me in a few months. And
37:38
Unidentified speaker (speaker_1): I did a like I this is not what I want to do. I hate it. I don't want to be glued to the screen 24 hours a day. I don't have to. I can make more money easier, certainly for a unit of my time on a longer timeframe. But I've also, I think
37:54
Unidentified speaker (speaker_1): that everything's good and bad, right? Everything's a double edged sword. And the advantage of day trading is you get many times at bat. You have to make many, many decisions throughout the day. Uh, you know, a disciplined day trader on a one minute chart, you're probably making what, on a slow day, three trades and maybe up to two dozen on a
38:13
Unidentified speaker (speaker_1): busy day. Is that roughly where you are?
38:16
Unidentified speaker (speaker_0): So uh j just to briefly give you context, I suppose it would help. Um, I found the sweet spot for me. I don't like look because exactly what you're saying, there is maybe an edge there, but it's at the expense of your time and health. and and and everything else. Uh so what what I like when I started, I was trading many asset classes and I realized I want to focus on two. And I I wanted to trade two things that are generally
38:35
Unidentified speaker (speaker_0): correlated, but they do depeg at times because I can give meaning to that. Okay. So I ended up trading euro dollar, pound dollar, and I've
38:41
Unidentified speaker (UNKNOWN): been
38:42
Unidentified speaker (speaker_0): doing that probably for six years now exclusively.
38:44
Unidentified speaker (UNKNOWN): I'll
38:44
Unidentified speaker (speaker_0): take one to four trades a week. Uh busier periods will get to uh maybe up to ten a week.
38:49
Unidentified speaker (speaker_1): Okay.
38:49
Unidentified speaker (speaker_0): Um naturally if I'm in a trend, then I can ride that trend. And what I found the advantages of doing so is because I I get what you're saying. Why don't I just turn that read into a swing position? But I found the benefit is if I can sense price could go and make a lower low in a bearish market, I can find the heightened pockets of reaction, pockets of probability.
39:07
Unidentified speaker (speaker_1): And
39:07
Unidentified speaker (speaker_0): instead of me catching one big one to five R, I can catch a few one to threes along the way. So my trade frequency is higher, my win rate is not the best. It floats around thirty five to fifty five. Yeah,
39:16
Unidentified speaker (speaker_1): I don't care about your winning.
39:17
Unidentified speaker (UNKNOWN): But my my
39:18
Unidentified speaker (speaker_0): my risk reward can be one to threes
39:20
Unidentified speaker (UNKNOWN): and
39:20
Unidentified speaker (speaker_0): my rate of return can be higher. And I and I have a good trade frequency, I think. And then I have my A plus setups and my B plus and my C plus XYZ, so I deploy them depending on recent performance and I've architectured it that way. Um I
39:33
Unidentified speaker (speaker_1): guess one of the challenges I would throw out there should you be doing your C trade, your B C trades at all? Or are you better off uh increasing leverage on your A trades? For many traders,
39:44
Unidentified speaker (speaker_0): yeah. Big debate I have is discretionary sizing. Do I size up on my high conviction plays uh or do I standardize? I've chosen to standardize
39:50
Unidentified speaker (UNKNOWN): and
39:50
Unidentified speaker (speaker_0): I see the merits in sizing up on on an A plus setup. But uh
39:54
Unidentified speaker (UNKNOWN): i
39:54
Unidentified speaker (speaker_0): we're talking about risk or exposure to the market. So there's two ways to have great greater exposure sizing up on a high conviction, or at certain scenarios in the markets, market, uh taking my C plus. setup, which has a lower win rate. But if I know
40:05
Unidentified speaker (UNKNOWN): it
40:05
Unidentified speaker (speaker_0): happened to me in September actually, uh I went through a hot period, not through skill, it's just the distribution of outcomes. A fifty percent win rate system can have a recent performance of eighty percent win because it just the wins aligned and now the losses may may follow. So as a result of that, if I have my uh trajectory of my expectancy and it's and it's just uh up into the top right,
40:25
Unidentified speaker (speaker_0): hopefully. And then I have a period where it overshoots. Yeah. Now it doesn't mean I've beaten my edge and now I'm gonna have a new like superstar edge. It's just I've overperformed. But I've realized when the t when the market is hot, or I'm hot at the moment, I can now add in a little bit more risk through riskier setups, which I wouldn't ordinarily take because my relative recent win rate
40:45
Unidentified speaker (speaker_0): is 80. It should be a 40. So now I can afford to bring it down gradually, and I know I'm calibrating. it. So when I start to hit the average of 50 in my last 15 trades, it back to my A plus setups.
40:56
Unidentified speaker (UNKNOWN): So
40:56
Unidentified speaker (speaker_0): that as I revert back to my mean of my trajectory of my expectancy, uh I can control when to deploy these higher risk trades. Uh instead of sizing up now I can actually get a greater rate of return through greater trade frequency temporarily. That's how I've navigated it. Okay. I'd love to hear your thoughts actually. Would I be better off just sizing up on high conviction and how having a lower general trade frequency?
41:17
Unidentified speaker (speaker_1): So
41:19
Unidentified speaker (UNKNOWN): if
41:20
Unidentified speaker (speaker_1): on the low side you're doing four trades a week, you're you're kind of tilted toward that longer holding period that I was talking about, I think, right? Or are are these are
41:32
Unidentified speaker (UNKNOWN): these ports?
41:33
Unidentified speaker (speaker_0): I'm trading FX.
41:34
Unidentified speaker (UNKNOWN): So it
41:34
Unidentified speaker (speaker_0): doesn't have the baked-in bias, so it doesn't necessarily serve me to just tur uh turn it into a long-term holding. hold. So I rarely hold a trade overnight.
41:41
Unidentified speaker (speaker_1): Okay.
41:41
Unidentified speaker (speaker_0): It would
41:42
Unidentified speaker (UNKNOWN): be
41:42
Unidentified speaker (speaker_0): usually a session reaction for my TP1, the average move of a session. And then my usually my TP two and final TP is average movement of the day. I might leave a runner if I see that there's a long-term trend in play, but generally not.
41:54
Unidentified speaker (speaker_1): I generally find that most traders, you know, this ABC trade book is a common way to think about it. Most traders are better off not doing their B and C trades.
42:04
Unidentified speaker (speaker_1): But what you're describing is exactly if you if your C trades are trades that only happen are more likely to happen in certain market environments, that you reach these environments. That's
42:15
Unidentified speaker (UNKNOWN): and I
42:15
Unidentified speaker (speaker_0): define it by lower win rates. It has a one to five risk road potential, but it it has a 40% win rate. I
42:23
Unidentified speaker (UNKNOWN): don't want to stop. That
42:24
Unidentified speaker (speaker_1): doesn't strike me as a C trade. You
42:25
Unidentified speaker (UNKNOWN): know,
42:25
Unidentified speaker (speaker_1): that's
42:25
Unidentified speaker (UNKNOWN): it. It still has it still has an
42:26
Unidentified speaker (speaker_0): edge, but it's just not as good as the other ones that might appear too twice a month. But it can give me a one to seven and it can give
42:31
Unidentified speaker (UNKNOWN): me a
42:31
Unidentified speaker (speaker_0): 80%, or not 80%, probably around 65% win rates. These are the ones I would love to only focus on.
42:36
Unidentified speaker (UNKNOWN): But
42:36
Unidentified speaker (speaker_0): if I exclusively trade them, my rate of return suffers because I'm taking a few trades a month.
42:42
Unidentified speaker (UNKNOWN): And if I have a
42:42
Unidentified speaker (speaker_0): bad month, it just to my personality, I don't like being wrong. Sure.
42:46
Unidentified speaker (speaker_1): Sure. No, I understand that. Um
42:49
Unidentified speaker (speaker_1): feels to me without like really getting into your numbers, like you've probably laid this out in a way that makes sense for you. I would be I would be very hesitant for you to come away from this interview and go make changes based based on anything that I'm saying, because you are you've already asked a lot of the questions that I would ask, and you seem to have good answers to them. And I think it sounds
43:09
Unidentified speaker (speaker_1): like you found what's working for you right now.
43:11
Unidentified speaker (UNKNOWN): The
43:12
Unidentified speaker (speaker_0): the interview is has kind of flipped, so let's say let's uh let's uh put it back into towards yourself. What would your equivalent be of what I've just described of your KPIs, the way you deploy risk, uh how you've built an edge around your temperament, which it seems like a a journey you had
43:25
Unidentified speaker (UNKNOWN): uh
43:25
Unidentified speaker (speaker_0): after your day trading uh era. Uh what does life look like now in the markets?
43:29
Unidentified speaker (speaker_1): My
43:30
Unidentified speaker (UNKNOWN): uh you
43:31
Unidentified speaker (speaker_1): know, I'm a I'm a person who
43:34
Unidentified speaker (speaker_1): r uh really feeds off of obsession. If I do something uh
43:39
Unidentified speaker (speaker_1): I've I have two speeds, either full stop or full speed ahead. And what I have found with the markets is while I have been involved and trading in the markets, I've gone through periods, I haven't been able to maintain for my entire career full focus. I think if I had, it would have been very unhealthy. And
44:00
Unidentified speaker (speaker_1): I have recently been in a period where the major of life transition where I haven't been trading quite as much. I haven't been taking as much risk. I've been uh doing more
44:11
Unidentified speaker (speaker_1): more simple trades, more sim uh only doing the trades that really jump out and say, you know, th this is a very clear pattern that you have to take. And also to be one thing we maybe should discuss is I don't think the visual prettiness of a chart picture is any
44:27
Unidentified speaker (speaker_1): judge of its outcome. Yes. I used to think that. Some people think that. But if you uh you know do do a few thousand trades and separate them into patterns that look really good and ones that don't, again, do they does the outcome look different? And you know, you can do tests like that when you have data. Um
44:44
Unidentified speaker (speaker_1): now my I think I'm looking at a future because I'm also thinking, you know, I recently had my 50th birthday. a few years ago, and I'm thinking about how am I going to trade in the future? And it's not going to look like how I traded in my 20s, and it's probably not going to look like how I traded 10 years ago. So my
45:04
Unidentified speaker (speaker_1): desire is to move into more of a
45:08
Unidentified speaker (speaker_1): I've always been a discretionary trader that has a quantitative foundation. You know, they the I I've asked these kinds of questions.
45:15
Unidentified speaker (speaker_1): the things that you ask about Fibonacci, things I've had, you know, every pattern I trade. If I see a market that makes a big move and then makes a consolidation, is there an edge to buying at the bottom of that consolidation? How can I define that? If I have, and some surprising things work. If I have a trend line, I I draw a trend line and then I
45:35
Unidentified speaker (speaker_1): draw a parallel trend line. Is there an edge to executing in a against that parallel trend line? There is. Okay. Why? I don't know. Volatility probably has something to do with random walk volatility, but there is and then the uh you know maybe the higher time frame influence of the pat I I I I do know it's there. It's probably but it's a uh you know
45:55
Unidentified speaker (speaker_1): issue of shorter time frame mean reversion plane into longer time frame momentum. Uh so that is how I have traded. But I think I'm looking at doing something like more of a weighted alpha kind of approach, where, and this is actually a loop that we should close. Your question about the 200-day moving average.
46:15
Unidentified speaker (UNKNOWN): For
46:16
Unidentified speaker (speaker_1): me to have something that I can trade as a standalone, it has to be quite a good edge, right? Uh It has to have a certain expectancy, has to occur often. You know, if it's a great trade that happens. once a year, I might have three bad years trading it, right? You you just don't so you have to have you have all these things that play into the system that's going to be the system you trade. One of the things
46:36
Unidentified speaker (speaker_1): you can do as a quantitative trader, though, and this uh for discretionary for retail traders, this I think this is going to be important because people don't on the retail side I typically think about this. I can have a handful of systems. edges that kind of suck,
46:53
Unidentified speaker (UNKNOWN): but
46:53
Unidentified speaker (speaker_1): they have a small edge. There's no way I could trade them. And you know, I might have it's like this. Uh let's say you have 50 friends who are kind of good at guessing the direction of the market, but none of them like would really make that much money. You call them up every morning and you say, Is the market going gonna to go up
47:13
Unidentified speaker (speaker_1): or down today? And they then you average them.
47:17
Unidentified speaker (speaker_1): And
47:19
Unidentified speaker (speaker_1): on days where most of them are telling you the market's going up, you would probably find you had a much stronger signal than you would from any one of those people. If you do what any one of those people tells you, you're not gonna make money. But you average them together. This is what happens in a lot of quant trading. You take these alphas and they have to be legitimate.
47:39
Unidentified speaker (speaker_1): alphas. And you said an important word earlier. They have to be uncorrelated. And the problem is not that many things are truly uncorrelated. Because let's say I want to trade mean reversion. Great. Okay. So a signal might be a certain RSI level, might be a momentum divergence, might be a big standard deviation move
47:59
Unidentified speaker (speaker_1): down. It might be crossing a Bullinger Bollinger band. It might be breaking under a five. or ten or twenty day high. All of these would be ways that we would measure a big move down, right? They're all highly correlated. So so if I if I think that I I just told you six different ways that I can put in my average alpha, no, I it's it's like me calling the same person six
48:18
Unidentified speaker (speaker_1): times in the average. So you have to understand, you have to find these edges and you have to they have to be uncorrelated. You average them together and you get a much stronger signal than you would from any of them. None of them individually have to be standout.
48:37
Unidentified speaker (speaker_0): I don't know why, but that bothers me from a perspective of
48:41
Unidentified speaker (speaker_0): why does that work?
48:42
Unidentified speaker (speaker_1): Yep. Well, it's uh
48:46
Unidentified speaker (UNKNOWN): I
48:46
Unidentified speaker (speaker_1): think it has to do with the wisdom of crowds. I don't know if you know that book, but like one of the since we're sitting here in the middle of farmland in America, uh, we have these county fairs. And one of the things that happens at these fairs, they would have an animal. They would have, you know, a bull or a pig. And people would
49:06
Unidentified speaker (speaker_1): put uh they would they would bet on the weight of the animal. If they get the weight right, they get the animal. And what you find is that if you average up all of those guesses, they're actually quite close in most cases to what the animal actually weighs. So what's actually happening there? You have people
49:26
Unidentified speaker (speaker_1): like me, and you know, I assume maybe you who have no idea what the so like we're just writing numbers down. My guess is not good, your guess is not good. But people like us who don't know anything, those guesses are going to average out to kind of see. zero. And then you have the handful of people who really know what they're looking at, who have intelligent guesses, and those actually so it's kind of
49:46
Unidentified speaker (speaker_1): like that with the alphas. You know, the ones that if they're weak working together, uh, the ones that they'll tend to average each other out until several of them point in the same direction.
49:58
Unidentified speaker (UNKNOWN): So
49:59
Unidentified speaker (speaker_0): you
49:59
Unidentified speaker (speaker_1): can't do this on a discretionary basis, to be clear.
50:02
Unidentified speaker (UNKNOWN): Oh.
50:10
Unidentified speaker (UNKNOWN): So
50:10
Unidentified speaker (speaker_0): let me cross that over with the trend line and an EMA and a doji candle and a head and shoulder batson. All of a sudden I've got a cocktail of confluences. Yeah.
50:17
Unidentified speaker (UNKNOWN): None
50:17
Unidentified speaker (speaker_0): of them were necessarily causative or had an edge, but in com combination, it might be.
50:22
Unidentified speaker (speaker_1): I I don't I've never seen that work. I be because first of all many of those things are correlated, right? Right. Uh the Fibonacci level and the trend line and the EMA, and you know, throw a band in there. You're you've moved to a level and those things are probably and are you know, the argument of course is well, if the trend line's at the level, maybe it's more
50:42
Unidentified speaker (speaker_1): meaningful. And you know, we we can test that and look at it. Um
50:46
Unidentified speaker (speaker_1): I
50:48
Unidentified speaker (speaker_1): in dis in the discretionary world.
50:51
Unidentified speaker (UNKNOWN): uh
50:51
Unidentified speaker (speaker_1): this is there's also this there's this argument against about should we have confluence and or confirmation. Those are kind of two separate things, right? Do you want to buy in anticipation of the move or sell an anticipation, or do you want to wait for the move to wait for the train to start moving
51:08
Unidentified speaker (UNKNOWN): and then
51:08
Unidentified speaker (speaker_1): you jump on board? Both of those things work, uh, but with the with with the confluence.
51:15
Unidentified speaker (speaker_1): people are many times just measuring the same things and they're using things that don't work. And is it possible, is it possible that something does not work at all? And when you put it with something else, that then it magically works.
51:31
Unidentified speaker (speaker_1): Uh probably not. It it might it might be possible that we have a signal that has no edge, and then we put something under it that filters out a bunch of the signals, and we have a I haven't really seen that work very reliably. It's theoretically possible, but the key to making this averaged thing work is that all
51:51
Unidentified speaker (speaker_1): of the each of the pieces has to have an edge separately and they need to be uncorrelated. Most of the things, like you know, the Fibonacci levels that we're talking about, uh, there's not an edge there in the first place.
52:03
Unidentified speaker (speaker_0): Can you remind me why uncorrelated was necessary? Because if I'm stacking this EMA trend line Fibonacci, da-da-da. Uh yes, they're all correlated, telling me the same thing, but that's the point. They are adding
52:12
Unidentified speaker (UNKNOWN): it.
52:13
Unidentified speaker (speaker_1): You're looking for the correlation, and you're assuming the correlation
52:15
Unidentified speaker (UNKNOWN): is assuming
52:16
Unidentified speaker (speaker_0): they are adding odds, stacking odds in my favor.
52:19
Unidentified speaker (speaker_1): You're just measuring the same thing a different way.
52:21
Unidentified speaker (speaker_0): Uh huh.
52:21
Unidentified speaker (speaker_1): You're you're the I that th that that's the argument against it. You because everything we look at in the market is, you know, if if if you really want to blow your mind, um going back to first principles, which is a phrase you said, everything we look at in the market is something that's happening in the market, but it's also the
52:41
Unidentified speaker (speaker_1): structure that we're imposing on the market. It's the way that we're measuring it. Even just a chart. You know, say, no, uh there's no structure, Adam. I'm just drawing a line of a like every trait, you know. a one-tick line chart. That's still a structure, right? Because you're still you still have an access. And certainly if we put it into bars or candles, we've put a structure around it. And then
53:01
Unidentified speaker (speaker_1): trend and you can see everything else, right? How however we measure that, uh RSI, trend line, Fibonacci, you're putting structure on it. So you're you're just measuring it. And is there that fundamental, you
53:14
Unidentified speaker (UNKNOWN): know,
53:14
Unidentified speaker (speaker_1): the the market may market made this move down? And then all of these things are going to speak. The uh, you know, the quantitative thing is see if you have a regression model, uh, and you know, people who have more math than I do understand this intuitively. And if those inputs into the regression model are highly correlated, they make your model look a lot better than it
53:34
Unidentified speaker (speaker_1): is. They make your confidence in the model, but it's not actually that good because your inputs, it's one of the
53:38
Unidentified speaker (UNKNOWN): things
53:38
Unidentified speaker (speaker_1): you're looking at specifically in most models, is you want to see those inputs are not correlated. And in reality, everything's kind of correlated. So you don't you never get to zero.
53:47
Unidentified speaker (UNKNOWN): Yes.
53:48
Unidentified speaker (speaker_0): I guess to visualize it, it's like all these confluences were like Venn diagrams where you're just adding more circles, but the crossover period is what's relevant. And as opposed to wrongs on a ladder, they're not climbing you higher. It's more like
54:00
Unidentified speaker (UNKNOWN): the
54:00
Unidentified speaker (speaker_0): Venn diagram. Okay. I'm also now curious on invalidations, because we've spoken about reasons price could do something and confluences. And then uh if they are too correlated, XYZ, uh you could you could do this process, but then I guess what would be more relevant is two things that appear optically similar or parameter similar, uh, but you would take one
54:20
Unidentified speaker (speaker_0): and you wouldn't take the other. What would certain invalidations look like?
54:24
Unidentified speaker (speaker_1): For me, um
54:29
Unidentified speaker (speaker_1): understand and we keep talking about buying, maybe we should talk about selling to make it symmetrical, but I'll keep it on buying. If if I'm looking to buy a consolidation, I don't want to see that it makes many probes to the upside and is rejected.
54:44
Unidentified speaker (speaker_1): So visually, if you're looking at a chart, I don't want that upper boundary to be really ragged. The market may makes a swing high and then comes down. I don't like if you're going above that swing high and getting rejected there because it's price re and we're back to Wyckoff, right? Uh on the other hand, I'm really happy to see that ragged bottom. You know, I'd like to see the
55:04
Unidentified speaker (speaker_1): the the contained and then it tries to sell off and they buy it right back up. So that kind of asymmetry. It's fine if the bottom's smooth too, but I certainly uh if I've been watching for an entry price and a pullback and it breaks out and then fails, in most cases I'm no longer interested.
55:24
Unidentified speaker (speaker_1): I, you know, I I have a very easy time killing setups. I I have a you know, I can be really excited. I don't get excited about anything, but I can be really excited about a setup and then it does something. I don't even want to talk, it's dead to me. I'm done with it. Um the sometimes it just goes on too long. You know, there should be my my thinking, and
55:44
Unidentified speaker (speaker_1): here we're back to stories. My thinking is that if I have identified one of these places where there's gonna be good continuation, there should be some urgency behind it. I I shouldn't see, you know, if there's if if we're talking about a commodity and there's a supply shortage, I should not see it make a big move up and then just kind of sit sideways for. months. That doesn't
56:04
Unidentified speaker (speaker_1): really say there's any kind of urgency there. Same thing if it's a one minute chart, you know, we just scale it differently. I want to see that there's urgency.
56:11
Unidentified speaker (speaker_0): Does that imply that you got in for a reason, whatever that reason may be, and volume or momentum is necessary, and therefore a sense of invalidation is lack of participation in your direction?
56:21
Unidentified speaker (UNKNOWN): Yeah.
56:21
Unidentified speaker (speaker_0): Would you therefore act upon that in the sense of you're already in a position, you have your parameters of take profit entry. and stop loss. Price is just around your entry, or a little bit in profit. Will you say, okay, up upon a kind of time decay, this is taking too long, long. It could still play out. It's just not happening in the time frames I need it to, but it hasn't invalidated. It hasn't hit my stop. The
56:41
Unidentified speaker (speaker_0): idea is still there. Would you exit that position?
56:44
Unidentified speaker (speaker_1): Yeah, I I would.
56:45
Unidentified speaker (speaker_1): That is one of the
56:49
Unidentified speaker (UNKNOWN): So
56:49
Unidentified speaker (speaker_1): you know, when I told you that I basically traded the same way for decades, trading the basic structure, these kinds of questions, exactly how do you manage a trade? How do you get out when it's working? How do you get out when it's not working? How do you know when it's not working? What do you do if it's not working and then it's working again? You know, how do you make these are the kinds of decisions that do
57:09
Unidentified speaker (speaker_1): evolve and change? And there would have been a time uh when I was at day. a trader, the answer to that question was automatically, instantly, I'm out. I'm looking for any excuse as a day trader to say the trade is not working. I am done. And, you know, you you learn this by getting kicked in the head doing doing
57:29
Unidentified speaker (speaker_1): it the wrong way enough times. As soon as you have a little bit of a little bit of hope, you know, a little bit, let's just give it a little more time. Sometimes it'll work and you'll remember those times. but if you run your stats, far more often you need to get out.
57:42
Unidentified speaker (speaker_1): Uh but on on the daily time frame, it has paid to be a little bit stickier. And I think this is one of the ways the market has changed in the past three years. It's no longer paying to be sticky on the daily time frame. If something doesn't look good, better situation.
57:58
Unidentified speaker (speaker_0): I was I was thinking when to bring up this topic, but I I feel like you've there's many roads in this. conversation that have led to this question. You were talking about the trend line and having a trend line parallel to it, and there seems to be something there. I took note of it. You were mentioning how when you get to the range high and you're getting rejection as opposed to movement, that would be some sort of invalidation. I took note of that because what would that mean?
58:18
Unidentified speaker (speaker_0): And then when you're talking about okay, you got into an you got in for a certain reason, but it's not moving as quick as you would need to.
58:23
Unidentified speaker (speaker_1): Uh
58:23
Unidentified speaker (speaker_0): and you would you take off you take off the position even though nothing has changed apart from time has elapsed. I took note of that. These things all uh bring me to
58:33
Unidentified speaker (UNKNOWN): I
58:33
Unidentified speaker (speaker_0): don't know the term for it, but it might be a sense of market impact, uh kind of a reflection of order flow where there are moments in the market that are highly reactionary, I found. Let's say, just to draw a visual,
58:44
Unidentified speaker (UNKNOWN): uh the
58:44
Unidentified speaker (speaker_0): there's a grades to it that I have, but I high have of previous day. Uh we could call that some sort of resistance. I found if price sweeps it, pushes higher, naturally you invite in breakout traders. They're going to look to buy that. And then also you take the stops of the people that were holding their sales in that position. So this becomes a high impact or high inflection point. And I've noted that when you sweep it,
59:04
Unidentified speaker (speaker_0): you get a reaction down that might not be for the rest of the day, rest of the rest of the month, but for a duration it will.
59:11
Unidentified speaker (speaker_1): Yep.
59:11
Unidentified speaker (speaker_0): And that's basically the center of my edge. Now I might be not it it's not just high of previous previous day. It could be a high previous week. It could also be a structural point. Th there's many levels to it. But that's basically an extension of my edge. And then I've noticed like the bigger zones, like a daily high uh sorry uh let's say a daily impulse, a lower high to lower low. At the extreme of it, if we take that high, it gives a bit of a
59:31
Unidentified speaker (speaker_0): bigger reaction as opposed to just in New York session you take London's high, which will give a smaller reaction. What is going on here? What what am I describing?
59:38
Unidentified speaker (speaker_1): I think you've d like you laid out the dynamics perfectly. That's exactly what happens.
59:43
Unidentified speaker (UNKNOWN): I
59:43
Unidentified speaker (speaker_0): I mean it more of like why is it happening?
59:45
Unidentified speaker (speaker_1): Because you are the market is always probing into areas to see if there's real interest. And uh I couldn't have said it better than you did. In my first book, I laid out six basic price patterns, what you're describing, your edge, what's the was first. one that I put in the book. This idea of penetrating
1:00:05
Unidentified speaker (speaker_1): a level and then failing. So, you know, one of the things that has happened, markets always pro markets are looking for liquidity. Markets want to make trades. They the exchange exists to facilitate trade, to make trades happen. So the markets are always looking for places to make trades. And the
1:00:25
Unidentified speaker (speaker_1): market is, if we think of the market as a conscious bean, the market's going to probe above those highs and to say, is there really buying pressure here? Certainly there are orders. You know, I can I know I can go get those orders executed because people have shorts, they have stops, and there are the hopeful breakout traders. But how much buying pressure is how do I know? Oh, Well,
1:00:45
Unidentified speaker (speaker_1): let's go execute there and see. And if there wasn't much, then those that those people are going to sell. Uh, you know, the the hopeful breakout traders, when their hope is gone, they sell, and then you kind of get this cascade of selling. For me, there's also a reaction after that. So you have that penetration versus if you have acceptance where you where
1:01:05
Unidentified speaker (speaker_1): you go up above the high and turns out, oh, there's actually real buying pressure here. There is continuation. If you have that failure scenario, then there's potentially kind of a second move, right? You have the you have the failure, the market goes back below the high, everybody's like, oh, what happened, right? And you sit there for a minute. And then if
1:01:25
Unidentified speaker (speaker_1): more sellers come in, you may get another leg down. That's also quite a reliable, tradable move sometimes. And then of course, you know, it's uh back to your question about me. m What I'm doing and when I'm in a trade when I'm thinking about a trade, every tick that comes in is data. And
1:01:44
Unidentified speaker (speaker_1): I'm analyzing every piece of data that comes in. I'm probably analyzing most of it wrong, but I'm getting enough right to be able to be on the right side of the market more often than not. And so, you know, the market comes in and now it's processing that move. And now if the market sits near that high, you know, now maybe. we're sitting, you know,
1:02:04
Unidentified speaker (speaker_1): you you can establish a short there against that high, but the more time you sit there, the probabilities are not really favoring that drop. They're favoring another trip to those highs.
1:02:13
Unidentified speaker (UNKNOWN): I
1:02:13
Unidentified speaker (speaker_0): like what you mentioned before of confluence versus confirmation. And I I I've always toyed with which one is better. Around these scenarios, you have a liquidity point, and price shoots above it. Naturally this is a dangerous place to be because it is by definition an inflation inflection point.
1:02:27
Unidentified speaker (UNKNOWN): You
1:02:27
Unidentified speaker (speaker_0): could get buyers stepping in, it goes up, it could also be a rejection point, it goes down. How do we know which one? So there's other things you could do to I guess put the odds in your favor. But I I found it's a bit more dangerous to enter there. But when I see the rejection and the sellers are taking strength, whether it's defined by a M1 break of structure or XYZ parameters, then upon certain parameters I'll then enter the shots. shot. But the issue
1:02:47
Unidentified speaker (speaker_0): is it's hard to know if this is is it gonna going be to a sustained move, which is why I can't turn that into a swing trade, even
1:02:53
Unidentified speaker (UNKNOWN): though I'm in a
1:02:53
Unidentified speaker (speaker_0): bearish market and I found a sell at the top. I don't know if I can hold it because it could chop around and it invalidate the next day. So what I've kind of found a parameter for me that helps me is I just break even super fast. I'll just like
1:03:06
Unidentified speaker (speaker_0): as soon as I can, it might tap me out here and there. It does happen a lot and that is frustrating. But a lot of the times I can just take a shot, not knowing really what's going on. But it just if I'm protected, I'm risk free, and then it happens to play. And uh if I look back in my last year, around 30% of my profit was trades where I got the direction wrong. I was selling and the the week ended up
1:03:26
Unidentified speaker (speaker_0): bullish, the day
1:03:26
Unidentified speaker (UNKNOWN): ended
1:03:26
Unidentified speaker (speaker_0): up bullish, but I caught a reaction, broke even fast and let it run to targets.
1:03:31
Unidentified speaker (speaker_1): So
1:03:31
Unidentified speaker (speaker_0): I was wrong, but yet captured a profit. Seems weird, but that it was thirty
1:03:35
Unidentified speaker (UNKNOWN): percent of my profit for the last year.
1:03:36
Unidentified speaker (speaker_1): Those are my favorite trades to be to be wrong and make money. Yeah.
1:03:39
Unidentified speaker (speaker_0): Yes.
1:03:40
Unidentified speaker (UNKNOWN): Um
1:03:41
Unidentified speaker (speaker_0): I I'm also curious here because yes, I'm a I'm at a younger or a different stage in my life compared to yourself, but I also find that finding alpha enough or an edge enough is so difficult. Most people are not in a place of luxury or abundance to say, let me pick the edge that suits me the most.
1:03:55
Unidentified speaker (UNKNOWN): Right. How
1:03:55
Unidentified speaker (speaker_0): have you found yourself in a position where I don't know how I would turn my approach to a swing approach and and take it a bit more easy. How have you been able to reshape an edge that would be draining on your time and consuming you and you're let's say obsessed with personality that you have to now be like, I'm getting older now and I want to be less stressed, so I'm gonna now turn it into a longer time horizon hold. It feels
1:04:15
Unidentified speaker (speaker_0): like you're just gonna be looking for a whole new edge in a whole new system.
1:04:18
Unidentified speaker (speaker_1): I don't I don't think
1:04:21
Unidentified speaker (speaker_1): you get to do what you're suggesting. I d I don't think you get to take your specific edge. You'd
1:04:26
Unidentified speaker (UNKNOWN): have to change it. and make
1:04:27
Unidentified speaker (speaker_1): it make it a longer term hold. So, you know, back to kind of themes we run through this interview, we're talking about mean reversion and momentum. And there are traders who trade broadly speaking, mean reversion or and you're a mean reversion trader. In general, and when you say in general, you're gonna make a lot of bad generalizations, but in general, mean reversion
1:04:47
Unidentified speaker (speaker_1): traders need to take profits very quickly. As soon as there is money. on the table, you put it in your pocket and you're done. Momentum traders, generally speaking, can be a little bit stickier. You know, I'll I'll let it ride. I get I'll give it a little bit more time. Um
1:05:03
Unidentified speaker (speaker_1): you don't get to decide you want to play for a bigger move because the another way to think about it is every signal that we would have, and to talk broadly in signals, I think, and this may not be accurate, but I think they have an in a limited influence over the market,
1:05:23
Unidentified speaker (speaker_1): both in terms of time and magnitude. So I do not believe, yes, we can go back 20 years and we can find the intraday low from the British pound that held for years. And you can say that was the moment, that pattern. held. No, it didn't. You know, I I I don't believe that. And yeah, that's just I guess it's a
1:05:43
Unidentified speaker (speaker_1): belief. But your mean reverting patterns have a relatively short duration where they're valid because they're going to be absorbed. Something happened. Unless, of course, then it cascades into something else, but that's a different kind of trade. Now you're swing trading the momentum to the downside, which is a valid trade. You know, somebody can look. at the
1:06:03
Unidentified speaker (speaker_1): trade you take and say, you know, that guy, it's like, you know, the person who's first is the person who gets shot a lot of times, right? You don't want to be the first guy through the door, right? Uh so they'll they'll let you take that trade. I'm gonna let you take that trade. I'll tell you that's the kind of trader I am. And then when it breaks to the downside, again, I'm
1:06:23
Unidentified speaker (speaker_1): interested in that move. So you've taken what I would consider to be a higher risk trade. It's actually probably a higher probability trade. You probably make money on more of those, but the way you make money on more of those is by taking smaller profits, which means you're not playing for a big move. You're not playing for a long move. So you, I don't think you get to say, I just want to hold these trades longer.
1:06:43
Unidentified speaker (UNKNOWN): How
1:06:43
Unidentified speaker (speaker_1): can I
1:06:43
Unidentified speaker (UNKNOWN): do that?
1:06:44
Unidentified speaker (speaker_1): Yeah. So I and I you figured
1:06:46
Unidentified speaker (UNKNOWN): that out.
1:06:47
Unidentified speaker (speaker_0): You've kind of spoken in labels of the the principles of the market could be mean reverse. reversion or trending or it could be um momentum these orientations.
1:06:56
Unidentified speaker (speaker_0): Uh how do I know which one I am? Because mean reverting and trend uh it depends on the time frame you're looking at. I could be trending daily bullish, but h h one bearish, M fifteen bullish, M one bearish. Which which time
1:07:09
Unidentified speaker (UNKNOWN): frame
1:07:09
Unidentified speaker (speaker_0): is the mean reverting and trend?
1:07:11
Unidentified speaker (speaker_1): Well the so you this is one very common mistake. Uh
1:07:17
Unidentified speaker (UNKNOWN): you
1:07:18
Unidentified speaker (speaker_1): have a time frame that you trade. That's your trading time frame. And let's say you get in let's just go back to our example. Let's say you get short on your one minute timeframe, and then it turns into a loss and you're holding it. And I'm I'm guessing viscerally you're like you wouldn't do that. So like, you know, you've you've learned that, that's but let's say
1:07:38
Unidentified speaker (speaker_1): you do that. because you notice the 15 minute is still kind of bearish.
1:07:42
Unidentified speaker (UNKNOWN): And
1:07:42
Unidentified speaker (speaker_1): then that
1:07:43
Unidentified speaker (UNKNOWN): goes
1:07:43
Unidentified speaker (speaker_1): away on the 15 minute. And you look at the, oh, the hourly. I have known more than one trader who
1:07:51
Unidentified speaker (speaker_1): have entered trades on very low time frames and ended up holding losing trades for weeks. The equivalent of risking a few hundred dollars and losing tens of thousands of dollars. This is, I call it the time frame justification system. And the way you avoid it is you say, This is the time frame I trade. I
1:08:11
Unidentified speaker (speaker_1): am in this market because of this pattern, because of this condition, and when that no longer exists, I mean business being in that market. So you you could look to develop a separate edge on the daily timeframe, but you really need to stick. with your time frame. It also has to do,
1:08:30
Unidentified speaker (UNKNOWN): you want
1:08:30
Unidentified speaker (speaker_1): me to let
1:08:31
Unidentified speaker (UNKNOWN): the thing up.
1:08:32
Unidentified speaker (speaker_1): It also has to do with how you manage the trade. Because uh we could be in a situation where, you know, let's say the the the euro goes up and then comes down and you sh you short it and I buy it at the exact same time. We both can make money on that trade. You know, you you might be out and I might buy more from you
1:08:52
Unidentified speaker (speaker_1): and then it goes up later and you would have had a losing trade, but you had a profitable trade because you have your you took your shorter profits as a mean reverting trader, again, very broadly speaking, needs to do. And because I was trading momentum, I needed to be stickier and I gave it more room to run. So it really is it's not just what the market does, it's you defining what you're going to do in
1:09:12
Unidentified speaker (speaker_1): response to what the market does.
1:09:13
Unidentified speaker (UNKNOWN): You
1:09:13
Unidentified speaker (speaker_0): mentioned you should have a time for. frame you select to trade upon. Uh by trade upon, you mean a a time frame I execute upon. Yeah. Because I can also do a top-down analysis on multiple timeframes to ideate something. And then I will always execute on M1. Uh should I separate those two things?
1:09:30
Unidentified speaker (speaker_1): I mean, you you you already do. Where where you get in trouble is if you're using and when you have one common frame is you can use several higher time frames or you can use a higher and a lower time frame. But and there are certainly things you can do, like you have your trading time frame and you execute on the lower timeframe and then you manage it
1:09:50
Unidentified speaker (speaker_1): on your trading time. Like those those but you you're consistent. You know you're not it's not like you're you're trading 15 five in one minute and then all of a sudden I call you up and you're trading on the two hour chart today. That's a break of discipline and that that will get you in trouble. So yeah, you just and you might even, you know, I could imagine that you tell
1:10:09
Unidentified speaker (speaker_1): me, so most of my trades are on the one minute chart, but I have this one setup that happens a few times a month on the hourly chart, and then I trade that. But the trade is consistent on the timeframe that you're where you get in trouble is bouncing.
1:10:21
Unidentified speaker (speaker_0): That makes sense to me. I I I've have with time I have found with the style of trading I have a top down analysis doesn't serve me because what relevance does a monthly trend or a weekly or even a daily trend have unless I'm in the inf like the extremes. Yeah.
1:10:35
Unidentified speaker (speaker_1): And
1:10:35
Unidentified speaker (speaker_0): most of the time I'm in no man's land. How's that going to affect my London session tomorrow? Uh so I've avoided it and and I'm basically between M fifteen and M one. Is there any merit in thinking in this way?
1:10:44
Unidentified speaker (speaker_1): That's where I've evolved over the years. I most of my trading
1:10:48
Unidentified speaker (UNKNOWN): So I
1:10:48
Unidentified speaker (speaker_1): gotta be careful here because I tell you most of my trading is on one time frame. But what I'm doing when I look at that one time frame chart, I'm visualizing the higher time frame. So I don't think there's anything wrong with you seeing the structure on the higher time frame. And I suppose, you know, you're aware of levels on the higher time frame that I may miss. And if you're coming into an important monthly level that I might not see on my
1:11:08
Unidentified speaker (speaker_1): one minute chart, it might be good to know that. Like, you know, I understand that argument, but I think in general, uh people add pieces of complexity and it doesn't necessarily add to better results. I I almost always have if I'm trading on the daily time frame, I'll have the weekly time frame there. But I'm uh my intention's on the daily
1:11:28
Unidentified speaker (speaker_1): I'm I I'm trading one chart, one time frame.
1:11:31
Unidentified speaker (UNKNOWN): Back
1:11:31
Unidentified speaker (speaker_0): to that scenario we said where we we probe high and we're looking for sellers to get in to then to then see a rejection.
1:11:37
Unidentified speaker (UNKNOWN): The
1:11:37
Unidentified speaker (speaker_0): word we use was liquid. liquidity. Do we
1:11:39
Unidentified speaker (UNKNOWN): see
1:11:39
Unidentified speaker (speaker_0): sellers' liquidity and lack of buyers liquidity? I I've had this conversation with a few people and they say to me, Well, there's liquidity everywhere in the markets. Literally at every point, every pip, every tick, there is liquidity. Will there be zones of heightened liquidity, such as a previous day's high? And does heighten liquidity equal better opportunity, or does it kind of also cancel
1:11:58
Unidentified speaker (UNKNOWN): out?
1:12:01
Unidentified speaker (speaker_1): Giving uh caveat of
1:12:02
Unidentified speaker (UNKNOWN): the
1:12:02
Unidentified speaker (speaker_1): to this answer that's not my specialization. I don't know for sure. Uh yeah, there's liquidity at every point, but there's a varying amount of liquidity. What is liquidity? It's not just resting orders, it's people being willing to execute there. And you certainly could revisit a level very quickly based on what price does and liquidity is complete. It's constantly changing, right? Uh I think that
1:12:22
Unidentified speaker (speaker_1): ultimately all of our edge comes from these shifts in liquidity. That's That what it is. Um, and you know, then maybe we're back to your order book question. Maybe people are able to access that and understand that in a way that I'm not really looking at. Uh, but I I think it all does come down. It comes, I mean, we're using liquidity, which sounds like a big fancy word, but
1:12:42
Unidentified speaker (speaker_1): really what it means is just people being willing to transact at a Better price and people, you know, there's always a buy order for every sell order, but you yeah, know, what what urgency. urgency is or what what motivation is behind those? That's really the question.
1:12:54
Unidentified speaker (speaker_0): As you mentioned to yourself of your evolve ever evolving, uh, I feel like I've reached a bit of a ceiling. And I know what the next step would be. I just have a skill gap, which is uh I do try to trade mechanical around parameters and it is objective. I don't trade intuition or gut feelings or anything like that.
1:13:11
Unidentified speaker (UNKNOWN): Uh
1:13:11
Unidentified speaker (speaker_0): so therefore it would serve me to take it algorithmic and and quantitative and uh make it systematize and let's say even eradicate human error by just having it automated. I just don't really know how to uh
1:13:22
Unidentified speaker (UNKNOWN): use
1:13:22
Unidentified speaker (speaker_0): self taught. So maybe you could shed some light again, especially in the AI era, how I myself and many others listening could take that step.
1:13:30
Unidentified speaker (UNKNOWN): Well
1:13:30
Unidentified speaker (speaker_1): let me tell you a bad thing to do with AI first. So this and I'm seeing a lot of this the bad thing you do with AI is you go to your AI of choice and you say, what is the edge in this pattern? Or find a uh find a trading system that works for you know what concept should I I don't think
1:13:50
Unidentified speaker (speaker_1): at this point, maybe this will change. I don't think AI is really good at that ideation level. Um I also would not trust an AI to say, uh, you know, we're talking look the stats behind your trade. Uh, you know, if the Euro takes out the one minute at in this time window during the day against these sessions, you know, what are the stats after that? I
1:14:10
Unidentified speaker (speaker_1): would, if you asked AI, I would have zero confidence in that answer. What I have done, so I have a quantitative framework that I've driven for years that was written in And MATLAB and Python and R. And I am now driving it with AI. So I'm having conversations with AI. I'm asking
1:14:30
Unidentified speaker (speaker_1): these kinds of questions, but the AI is not answering the questions. The AI is driving my code. But what it means is because I'm a I'm a pretty shitty programmer, to be honest, I I don't have to actually program the system, uh, figure out how to you know program the levels.
1:14:49
Unidentified speaker (UNKNOWN): I
1:14:49
Unidentified speaker (speaker_1): can tell the AI what I want. It can kick back visual confirmation. I can look at a few hundred charts and say, yeah, that's right. You know, let's tweak this. And then I can run the test. But it's coming from my code driven by AI.
1:15:02
Unidentified speaker (UNKNOWN): I
1:15:02
Unidentified speaker (speaker_1): think anybody can do something like that, but I think you have to kind of and it sounds you know, I know it sounds daunting, but you can develop your own framework. That's gonna be much more valuable than asking AI for the answer.
1:15:13
Unidentified speaker (speaker_0): I've spoken to a variety of guests on the show, and a unanimous common denominator between all of them is the emphasis they put on data and actually knowing the inner workings and the insights of your edge and your performance. That's why I'm proud to bring a partner of the show, TradeZella, the number one journaling, back testing, and all in one insights experience created by traders for traders. What TradeZella really gives you is
1:15:33
Unidentified speaker (speaker_0): deep insights about your trading that would ordinarily not be visible, whether it's through understanding your trade types and playbooks or even insights powered by artificial intelligence through Zella AI. Whether you trade forex, futures, cryptos, the stock market, it all seamlessly connects to TradeZella. So there is no additional work. You've seen me reference it dozens of times and all of the benefits I've had in my trading from the insights I found from my
1:15:53
Unidentified speaker (speaker_0): TradeZella. So join myself and thousands of other viewers of the show. You'll get the best discount using the link in the description or code T TOT O T for Titans of Tomorrow. So the reasons I would want to go. on this path and and I could hire someone or whatever the answer would be. But anyway, the reason I would do it is to seek a certain advantage. And the advantage would be I guess an element of buying back a bit of time. Uh the
1:16:13
Unidentified speaker (speaker_0): advantage would be when human error comes in that could be eradicated, even to this day it does happen from time to time.
1:16:18
Unidentified speaker (UNKNOWN): Uh
1:16:19
Unidentified speaker (speaker_0): and then also um p performance related of like if I'm traveling travelling right now, I just don't trade because I'm in different time zones
1:16:25
Unidentified speaker (UNKNOWN): and
1:16:25
Unidentified speaker (speaker_0): I can't really do it. But I could in theory have it still running and check up on upon it's uh so these would be the bear benefits I would pursue.
1:16:32
Unidentified speaker (UNKNOWN): But
1:16:32
Unidentified speaker (speaker_0): now you are also algorithmic and mechanical and objective, but also discretionary. And it feels like then you would remove the said benefits that I'm looking for by being discretionary in in the mix. So
1:16:43
Unidentified speaker (speaker_1): there's there's several ways to do it. One of the classic ways to do it would be to have and I think this could be done pretty easily for what you're looking to do, have your entry be automatic and then you manage the exit.
1:16:59
Unidentified speaker (speaker_1): So
1:17:00
Unidentified speaker (speaker_1): the code takes the entry and then you have a position. Now the problem is if you're not at the screen, if you're not paying attention, then
1:17:07
Unidentified speaker (UNKNOWN): you know
1:17:08
Unidentified speaker (speaker_1): then you have an issue. Um the
1:17:12
Unidentified speaker (speaker_1): there there are many ways to
1:17:15
Unidentified speaker (speaker_1): intertwine discretion and systematic, discretionary trading, systematic trading. And I know most of the you've talked to systematic traders who tell you they have no discretion. That's not true because you still you decided what to test, you decided, this is the word related to discretion, right? You decided what to trade, you decided what kind of
1:17:35
Unidentified speaker (speaker_1): risk to trade, and you're going to decide when the thing stops working. when to turn it off. So there is absolutely an element of discretion. Uh we can't have we we can't be purely systematic. You can't be pure uh maybe someone can. I've never seen it. A purely discretionary trader would have none of what we're discussing. They would have no plan. They would look at the
1:17:55
Unidentified speaker (speaker_1): chart, they say, I'm gonna sell it there, buy it there, and then you know, they're just making uh that person over time, they're going to accumulate.
1:18:03
Unidentified speaker (speaker_1): experience and they're gonna accumulate a system around their if they're successful, they're gonna have some systematic elements to their to their discretion. I
1:18:14
Unidentified speaker (speaker_1): would suspect for you the i if it does not remove the part of you being at the screen, but I also don't think you can trade the time frame you're gonna going to trade without you being at the screen. Um but that idea about having the entry and then you managing the exits. And of course when we program the entry we'll also program some
1:18:34
Unidentified speaker (speaker_1): exits, right? So one of the things that we need to do whenever we start meddling with a systematic approach, we need to analyze our discretionary inputs and see is it adding anything. And I've seen this, you know, I mean I've I've traded some systematic approaches in the past. and I found on balance my discretionary
1:18:53
Unidentified speaker (speaker_1): input is actually helpful, but I'm monitoring it and I know and I can I can quantify the alpha contribution of my discretion over just letting the system run. Also has to do with the kinds of systems. If we're trading a system that's based around the kinds of patterns that we're talking about, that would be somewhat similar to
1:19:13
Unidentified speaker (speaker_1): discretionary trader. I think I have valid input as a discretionary trader, but the aver the weird averaged alpha approach, I probably don't have any valid discretionary input there, I would assume.
1:19:25
Unidentified speaker (speaker_0): Is time of day of execution relevant to you?
1:19:28
Unidentified speaker (speaker_1): Yeah, I mean it this goes back to liquidity, right? Because uh the most liquid market can be completely dull and dead for hours and hours out of the day.
1:19:39
Unidentified speaker (speaker_1): There certainly are intraday edges that are very time frame specific.
1:19:45
Unidentified speaker (speaker_1): You can, if people want to start doing quantitative research, you can
1:19:50
Unidentified speaker (speaker_1): start chopping the day up into hours or half hours and asking, you know, hey, what if we make a new high in this half hour? Is that different than making it? And the answer is yes. So it it m matters quite a lot. And this is why I'm sure you've talked to day traders who only trade a few hours a day. They've carved their edge. And yeah, I mean maybe this is you too. You've carved your edge out of a
1:20:10
Unidentified speaker (speaker_1): specific window when specific things are happening. And absolutely that matters.
1:20:14
Unidentified speaker (speaker_0): Yeah, in the end, uh trading 24 hours a day would be chaotic. So it ends up being two hours a day for me. But I don't know if I found anything specific that would benefit me or because there are times that uh I I'm seeing the exact same pattern, if if I'm to call it signatures signature of surprise that I'm interested in happening at a London Open, I take it. Happening three hours later in my no man's uh no man's land time
1:20:34
Unidentified speaker (speaker_0): time frame, uh I won't take it. And I don't know if I improve the odds because I just I'm susceptible to less opportunities, therefore less losses, or is there actually a heightened statistical skew in my one up two hour window that I have?
1:20:46
Unidentified speaker (speaker_1): I would think that could be tested. I mean if you want
1:20:48
Unidentified speaker (UNKNOWN): to
1:20:49
Unidentified speaker (speaker_1): if you want to send me some I I can I can throw it into my research process and take it
1:20:52
Unidentified speaker (UNKNOWN): up. I'll take you up on that
1:20:53
Unidentified speaker (speaker_0): offer. That sounds fun. In fact, what would be the best way for me to approach all of this if I wanted to become a little bit more quantitative? Would it be upload price action to something and then just start conversating through AI? Or would it be here's my last hundred trades, feed it in and then ask questions?
1:21:09
Unidentified speaker (speaker_1): I think b I think both of those are valid approaches. I think also there's something to be said for
1:21:15
Unidentified speaker (speaker_1): constructing what I would consider to be kind of
1:21:19
Unidentified speaker (speaker_1): a
1:21:21
Unidentified speaker (speaker_1): I I don't have a word for it.
1:21:24
Unidentified speaker (UNKNOWN): So
1:21:24
Unidentified speaker (speaker_1): we're talking the kind of quantitative analysis we're talking about is one thing. There's also a way which we haven't really discussed of coming up with a specific rule set, right? So um, you know, I'm looking for a price to be below the 200-day 200 day moving average and then to cross a above the the five, the three crosses above the five day moving average, and I take a buy with a stop. You know,
1:21:44
Unidentified speaker (speaker_1): so you and you've talked to a lot of people probably who do this. That's certainly a valid way to to solve the puzzle. I think there's some value in doing that because you will learn how so I would try to, I would try to structure a system like that. Um and you know I I've I gave one example, but
1:22:04
Unidentified speaker (speaker_1): for you, it might be the rule might be in this time window. Uh if the market goes above this level and is back below it, I don't know if you ex back below it in one minute or you know, whatever. That's that's a variable you can look at. Uh, you know, I'll get short here, my stop will be at or beyond that previous high, my profit target is here. If I don't
1:22:24
Unidentified speaker (speaker_1): have a profit in fifteen 15 minutes, I get out of the you know, and those are all variables and things. There would be a lot of value for you, I think, in taking what you're doing on a discretionary basis and putting it into a clean rule set like that, and then seeing it run. And then going back and, you know, so one of the things it's very easy to push the
1:22:44
Unidentified speaker (speaker_1): button and let the system run. What you also need to do, I think this this is how at least this is how I learned a lot. is go back and look at the specific executions. And you gotta look at a lot of them. You, you know, you need to you don't go and look at three trades and say, well, those looked stupid. I wouldn't have done those. You know, we're talking dozens of trades.
1:23:04
Unidentified speaker (speaker_1): And you know, if you've done a back test, hundreds, this takes a lot of time. You have to be a not normal person to to to want to do that. And if you are normal, it will make you a not normal. But you know, you're you're going through and you're looking at hundreds of these and you s and you can take some notes, but you're also building some. This is how how we build intuition is seeing patterns,
1:23:24
Unidentified speaker (speaker_1): right? And seeing occurrences. And then you'll think about making them. I've had cases where I've tried to systemize approaches, exactly what I'm explaining here, and I was not successful. I was doing something on a discretionary basis, and I went through iterations of this or systematic approach with it. Couldn't could not
1:23:44
Unidentified speaker (speaker_1): systemize, but it made my discretionary trading better because it made me look at a lot of patterns, it made me look at a lot of outcomes, it made me think about what exactly was contributing to the edge. And you know, so the system development was a failure, but I made money from it. And that's that's like having a making money on a losing trade. Exactly.
1:24:01
Unidentified speaker (UNKNOWN): Feels
1:24:01
Unidentified speaker (speaker_1): good. I'm
1:24:02
Unidentified speaker (speaker_0): curious if overfit. overfitting, which is
1:24:04
Unidentified speaker (UNKNOWN): I mean
1:24:04
Unidentified speaker (speaker_0): yesterday's conversation I had, it seemed like a a big topic.
1:24:07
Unidentified speaker (speaker_1): It's a
1:24:07
Unidentified speaker (UNKNOWN): it's a
1:24:07
Unidentified speaker (speaker_1): big topic.
1:24:08
Unidentified speaker (speaker_0): But I I don't know if it's like a problem that most people won't encounter or is it something actually problematic.
1:24:13
Unidentified speaker (speaker_1): It's a problem in that kind of in that kind of development very specific. Because uh
1:24:20
Unidentified speaker (UNKNOWN): what
1:24:20
Unidentified speaker (speaker_1): are what are the variables? Okay, so we're talking about a time window. We're talking about a certain penetration through ah sorry. sorry, certain penetration through the high. We're talking about where you put your profit target. And then and there's others. But
1:24:32
Unidentified speaker (UNKNOWN): what
1:24:33
Unidentified speaker (speaker_1): you could do is test every combination of those. And
1:24:37
Unidentified speaker (speaker_1): some of those will work. You know, if we're uh testing random levels back to our Fibonacci discussion, some of those random levels are going to work. There's going to be non non-random random action. You may know, you know, statistically if you do a bunch of tests, you can't just look at the p-value. You have to adjust for the fact I did a bunch of tests. I did 20 tests. Therefore,
1:24:57
Unidentified speaker (speaker_1): there's a probability at a 95% that one of those is going to look good, right? Uh, so you you have to, I think this is where a lot of those kinds of system developers get in trouble, is they'll do all of these tests and then it is massively overfitted. And there are ways to overcome that, right? Like if I see, if you tell me you tested this grid of
1:25:16
Unidentified speaker (speaker_1): five variables and you found some magical combination that worked and all the others look like garbage and one works, I don't think you have anything. But if there's kind of a plateau where a lot of things work, you might have something. The gold standard is to do your development and hold out an out out-of-sample of sample and then test that.
1:25:36
Unidentified speaker (speaker_1): Most people don't do that right because the problem with out of sample is how many times can you do the out of sample?
1:25:44
Unidentified speaker (speaker_1): So if you're doing your system development and then you're like, let me just look at this on my you've completely invalidated your test. Most people, the outside of you know real quant firms don't have the discipline to do all of the system development, which may take weeks and weeks and weeks, and then to run it on the out-of-sample, out of sample, which answers. to some extent, what if I had
1:26:04
Unidentified speaker (speaker_1): traded this over the past year, six months, or whatever, if you see persistence there. You can do a cheap out-of-sample. Uh, all of my tests, I also look at like uh I call them temporal quantos. So, you know, I'm I'm slicing whatever data that I'm looking at into periods, and I want to see some persistence.
1:26:24
Unidentified speaker (speaker_1): If it only looks good, if I'm testing 10 years. and only looks good in two and a half of those, it's probably not great.
1:26:31
Unidentified speaker (speaker_0): We also spoke about at the beginning of alpha decay. So if I'm
1:26:33
Unidentified speaker (UNKNOWN): going
1:26:33
Unidentified speaker (speaker_0): through all this in sample out sample testing
1:26:36
Unidentified speaker (UNKNOWN): and
1:26:36
Unidentified speaker (speaker_0): then I start to deploy it, is there any degradation rate you found or of
1:26:41
Unidentified speaker (UNKNOWN): of
1:26:41
Unidentified speaker (speaker_0): alpha decaying and therefore an edge an edge has a certain li shelf life?
1:26:45
Unidentified speaker (speaker_1): I think so. Yeah.
1:26:46
Unidentified speaker (speaker_0): What what is that uh for yourself
1:26:48
Unidentified speaker (speaker_1): it varies, but it's re it's varies. It's real and it seems to go in one direction. And it's not it's not the direction you want. It's it's not the answer you want. Um, I think you can greatly accelerate alpha decay by publishing. So tell people about it. Uh one of the
1:27:04
Unidentified speaker (speaker_1): absolute standards is, you know, there's tons of academic papers that have real alphas published. Uh there are books that have real and there's also lots of books that don't have alphas that they say they don't. do, but they don't. But what you will find is that those alphas that were good at one point in time, invariably they've degraded since publication.
1:27:25
Unidentified speaker (UNKNOWN): And
1:27:25
Unidentified speaker (speaker_1): this goes back to our discussion about this doesn't mean that you can't that I can't trade the pullback, that you can't trade the tests through that level. But how we trade it today may degrade, and we may need to modify. that. You know, you may need to change your change what the penetration looks like, change what the action after, you know,
1:27:45
Unidentified speaker (speaker_1): any of these variables may need some adjustment. And I think as a discretionary trader, when you're in touch with the data, you're going to do some of those adjustments automatically. And that goes back to where I said, you know, I tell you, I've traded the same way for decades. It's a lie because I've modified the subtle things around it,
1:28:05
Unidentified speaker (speaker_1): sometimes without even knowing. You know, I I don't even know I'm doing something different until I go back and look at a different period. I'm like, oh, my stops are in quite a different place now. And, you know, I evolved with the market. It's the advantage of discretionary trading.
1:28:17
Unidentified speaker (speaker_0): Right at the beginning, on your own admission, you said your intuition sucks. And in fact, I should do the opposite of your intuition
1:28:22
Unidentified speaker (UNKNOWN): and how
1:28:23
Unidentified speaker (speaker_0): to make money. Um
1:28:24
Unidentified speaker (UNKNOWN): you
1:28:24
Unidentified speaker (speaker_0): you mentioned intuition is kind of like your subconscious. picking up on patterns.
1:28:28
Unidentified speaker (speaker_1): Maybe.
1:28:28
Unidentified speaker (speaker_0): Uh so that feels like some sort of advantage. But anyhow, the question is if you are optimizing through quant uh through probabilities and you have a quantitative approach and you say your intuition sucks, what's the purpose of discretion?
1:28:43
Unidentified speaker (speaker_1): I
1:28:45
Unidentified speaker (UNKNOWN): there
1:28:45
Unidentified speaker (speaker_1): are patterns that I can trade. And I don't I don't think it's the same thing as intuition. I think if I identify a pattern like the pattern you've described to me, if I can identify that pattern on a chart, I might not be able to quantify
1:28:59
Unidentified speaker (UNKNOWN): it.
1:28:59
Unidentified speaker (speaker_1): I might not be able to explain it to code in a way that I could execute. But I can see it on a chart. I can trade it on a chart. In fact, it might be quite easy for me, but if I describe it in code, what I'm gonna find is I'll describe that, but I'll also describe four examples that aren't actually
1:29:17
Unidentified speaker (UNKNOWN): the
1:29:17
Unidentified speaker (speaker_1): thing. So there is some degree of filtering. Maybe if I were a better programmer, maybe uh, you know, some of the modeling tools now. Uh I did quite a bit of work many, many years ago with uh was working with a group of tr of uh engineers who were looking at the algorithms that radars use to a
1:29:37
Unidentified speaker (speaker_1): radar from an airplane. looking down at another airplane, everything's just reflected back to the radar. You may know this, or many of your listeners do. You get a lot of ground clutter. So it's not the simplest thing to pull. So there are algorithms, of course, that process that. So they were looking at those kinds of algorithms, which
1:29:57
Unidentified speaker (speaker_1): also had found use in speech recognition, and they were trying to apply those kinds of algorithms. to my discretionary pattern recognition.
1:30:05
Unidentified speaker (UNKNOWN): It's
1:30:05
Unidentified speaker (speaker_1): an interesting project. We spent a lot of time on
1:30:07
Unidentified speaker (UNKNOWN): it.
1:30:08
Unidentified speaker (speaker_1): Zero. Z zero result from it. Uh I think, you know, maybe some of the modeling tools are a little bit more sophisticated now. Maybe you could show uh the right kind of training environment. This is a pattern, this is not a pattern, and then it will see things. It could maybe tell us things about our perception. We don't know. I
1:30:25
Unidentified speaker (UNKNOWN): don't
1:30:25
Unidentified speaker (speaker_0): know. I feel like we could have an another like 10 hour
1:30:27
Unidentified speaker (UNKNOWN): conversation.
1:30:28
Unidentified speaker (speaker_1): Yeah.
1:30:28
Unidentified speaker (speaker_0): So I think the best way for me to end is asking you,
1:30:32
Unidentified speaker (UNKNOWN): is
1:30:32
Unidentified speaker (speaker_0): there anything I should have asked you today?
1:30:36
Unidentified speaker (UNKNOWN): I
1:30:36
Unidentified speaker (speaker_1): think we did a really good job to pat ourselves on the back, uh, staying focused. Because there were a lot of things that we could have talked about. And there are I think that one of the things that is occupying a lot of my thought at this point in my life is thinking about life outside of markets and thinking about, you know, how do you
1:30:57
Unidentified speaker (speaker_1): how do you interact with the markets responsibly with the degree, let's say focus because it's a nice word, but we really mean single minded, crazy obsession, while also being a good human and also enjoying the time that we're alive. Yeah. And I think that crosses into many other questions. And we can talk about
1:31:16
Unidentified speaker (speaker_1): meditation and spiritual practice and art and the meaning and the meaning of everything. And um
1:31:22
Unidentified speaker (UNKNOWN): I
1:31:22
Unidentified speaker (speaker_1): mean we could talk about knowledge and you know what we know and how we know it and how sure we are with I think that's actually a the
1:31:29
Unidentified speaker (speaker_1): maybe that's a separate interview sometime.
1:31:31
Unidentified speaker (UNKNOWN): But
1:31:31
Unidentified speaker (speaker_1): I think we did a good job staying on track today. Yeah
1:31:33
Unidentified speaker (speaker_0): it was a very colorful colourful conversation with a lot of wisdom and lessons. I thoroughly enjoyed it. Thank
1:31:37
Unidentified speaker (UNKNOWN): you for the opportunity.
1:31:37
Unidentified speaker (speaker_1): I enjoyed it too Thank you.
1:31:38
Unidentified speaker (UNKNOWN): well.
This page was produced automatically: the audio was transcribed with nvidia/parakeet-tdt-0.6b-v3+sortformer-v2.1 and interpreted by Qwen3.6-35B-A3B-UD-IQ3_S.gguf. Transcription and interpretation can contain errors; always confirm against the original episode. Nothing here is investment advice.

More from Titans Of Tomorrow

Titans Of Tomorrow

The Woman Who Beat 99.6% Of Hedge Funds: OrderFlow Is Noise! THIS Is 90% Of The Game!

Episode discussion: Linda Raschke argues order flow is noise, advocates for price action, and discusses trading psychology and AI.

Speakers’ claims: Linda Raschke argues that order flow is largely noise generated by algorithms and that technical indicators based on price derivatives are lagging. She claims top funds like Renaissance Technologies use sub-24-hour holding periods and that large institutions lack the liquidity flexibility of retail traders. Raschke advises discretionary traders to cut size during drawdowns but maintain size for systematic strategies. She distinguishes between using AI for operational grunt work versus alpha generation, warning against relying on LLMs for direct signals. The discussion covers trading psychology, flow state, and the importance of physical health for longevity in trading.

Market MicrostructureTrading PsychologyAlgorithmic Trading
1h 22m5 assets25 claims to verify
Titans Of Tomorrow

33,554% USIC Champion: The Reason Most Traders Blow Up Is Embarrassingly Simple!

Episode discussion: Mark Minervini discusses risk management, VCP strategy, and verified trading returns in episode 33,554% USIC Champion.

Speakers’ claims: Mark Minervini explains that his trading success relies on strict risk management, specifically using Volatility Contraction Patterns (VCP) and high relative strength stocks. He argues against the efficient market hypothesis, citing price action over fundamentals, and emphasizes that losses work geometrically against portfolios. Minervini shares verified records of his trading performance, including a 33,500% return on a $32,338 start, audited by an accounting firm. He discusses the psychological challenges of trading, including the 'cardinal sin' of not cutting losses and the importance of humility. The episode also covers his views on trading competitions, leverage, and the evolution of trading technology.

Risk ManagementVCP StrategyTrading Psychology
1h 13m2 assets25 claims to verify
All insights