Successful traders have some unusual quirks in how they think. For my entire career, I've felt these "quirks" are valuable, and the world would probably be a better place if more people adopted them. However, I've always struggled to answer the enthusiasm of a young person who wants to get into trading, thinking the skills developed will translate to another career path if trading does not bloom for them.

I find that much of what we do as traders is actually pretty specialized and focused, and might not translate to success in other domains. However, I've recently had a personal experience where my trading mindset might just end up saving my life, or, at the very least, making it better.

We're all on a health journey

Physically, my story is the same as most other men in the developed world who've been around about half a century, and maybe made a bit worse by my sweet tooth, love of fine dining, and fairly extensive wine cellar--all of this is such a common story, especially in America today, that it's pretty boring.

We also know how consistently people fail when they try to "improve themselves." Of course, broad goals like that are no goals at all, but even specific goals bring challenges that are difficult to overcome. "Eating better" when that dessert looks so appealing puts a mental ideal against biological operating system code, and we know which one wins.

Furthermore, we tend to be creatures of habit, and for good reason: habits have generally served us well and kept us alive. Whatever operating system we're running is justifiably suspicious of attempts to change.

And yet, I've built entirely new habits and completely transformed my routine in three months' time.

So what changed? Well, I thought about what has worked for me in financial markets. The key to everything I've figured out about markets or achieved in trading depends on data--on looking at what is, trying to understand it, and figuring out what to do with it. As one of my mentors used to say, "let's ask the data, shall we?"

In trading, this is a combination of quantitative analysis of market patterns, subjective evaluation of patterns, and an overarching awareness of what other market participants might be trying to accomplish. That sentence hides a lot, and contains the seeds of many blog posts for another day.

I wondered if there was some way to apply those skills to solving my health problems. You can't manage what you can't measure, so the first task was getting the data. For me, this started with a complete look at what can be measured through blood work. Yes, this means a few dozen vials of blood drawn, but you'll have a baseline and know what you need to address. (Your mileage may vary, but my doctor was not much help here. Doctors are resistant to testing anything outside their specialty. I would suggest you look at a service like functionhealth.com.)

You might also consider a full body MRI. This is controversial because such a scan will almost certainly show areas of concern that are, most likely, nothing at all to worry about. These so-called incidentalomas can lead to procedures with very real risk. Often, the tiny risk of these biopsies dwarfs the even tinier risk that the finding is meaningful. This will be far from intuitive in the moment, and here's where the lesson of calculating actual risk from the web of probabilities and payoffs starts to have real life impact. My personal answer to this dilemma was to pre-commit that this first scan would be treated as a baseline for future comparisons. There, it has real value, but the first scan itself can be highly misleading.

DEXA body composition scans are an underused resource. For a relatively small fee, you can get measured insight into how your body is managing lean mass vs. fat. For anyone on a GLP-1, I predict this is going to become the next epidemic: losing weight usually means losing fat and muscle together. If you don't want that to happen (and you really don't--lean mass preservation is key to longer-term health) you're going to have to do some unpleasant work lifting some heavy weights. I wonder what fraction of people seeing real weight loss with these relatively new drugs are holding on to enough muscle mass to make the next decades of their life comfortable. A DEXA scan, at intervals, puts the data in front of your face in a way that can't be denied.

This is an ongoing project, and it's a data-hungry one. Wearables are good, but consider your options. Anything I write here is likely to be outdated in a few months, as these devices evolve quickly. I've worn an Oura ring for years, primarily for sleep tracking. If you eat a standard American diet, you probably should wear a continuous glucose monitor for a few weeks to see what's going on under the hood. Morning fingersticks and home blood pressure cuffs are cheap; the typical once-a-year datapoint from the doctor's office is not that useful.

As I said, it's hard to make changes and it's even harder to make changes stick. What worked for me was collecting these datapoints and correlating changes in the data to changes I was making in daily habits: eating, alcohol, exercise, sleep, meditation, and many other aspects had to be tracked carefully. It was also necessary to re-run some blood work periodically, on a much faster cadence than my PCP would have approved.

All of this was intense, to say the least. Creating the framework within which to analyze the data required use of both AI and python. I found I could not trust AI's assessment of the research, so I had to actually read studies to untangle what's current and real from misconception. I did not log the time I spent on this, but I would conservatively say probably a hundred hours all-in.

Once done, I had a feedback loop that was self-sustaining, and here regular AI oversight became useful, with one meaningful caveat coming up. Regular data logging became minutes each day, and I created a summary "score card" that put all of the relevant metrics on a single page, drawing from my logs and using APIs to grab data from my wearables. AI was a big help in coding this; it would have taken me weeks and AI one-shot the framework and then iterated the details in a few hours.

What Did I Learn?

So what were the actual lessons I took from this work?

I respond very well to fluctuations in the data. It has been useful to both look closely at details of changes and also to keep an oversight on the big picture. This takes time, intent, focus, and constant attention. It's not possible to do this work--or, I believe, any meaningful work in any field--without some degree of obsession. Tracking small changes in things I could measure allowed me to make behavioral changes.

It all goes back to collecting the right data. Without that data, I'd be acting on faith and hope and there's no way I could have sustained any focus. An abstract idea like "getting healthier" or "living longer" is simply not going to hold up against a croissant and some cassis jam. Measuring and recording data became its own subdiscipline that drove everything else.

AI is a useful partner, with caveats. There are things that AI does much better than humans. (I still see apparently intelligent people claiming that AI is nothing more than a prediction and engagement machine or a "really good autocomplete". People cling to ignorance if they can take comfort in it.) But there are also places where AI misfires. There have been a dozen cases where my intuition about the way a datastream would bend was proven correct while AI was contradicting. (And the jury is still out on one single case where it looks like my intuition might have lost--the scorecard is literally 11 to 1.) I've spent a lifetime looking at data and changes so I don't know if this intuition would carry to other people without this experience, but I suspect it might. Trust your intuition enough to test it. It's the test that really matters.

Experts are unreliable. I don't want to bash doctors, but I'll say your doctor deserves suspicion. I found my doctor completely unaware of or unwilling to engage with many aspects of this work. Consider diabetes--the progression of this disease is well known and is clearly driven by the modern American diet which we've exported to much of the rest of the world. More than half of adults linger in a pre-diabetes stage or have active diabetes. There are appropriate tests that could be done, but, they usually aren't done and insurance usually won't cover them. The plan is to give patients some vague dietary and lifestyle advice until they reach the point when they need to be put on drugs for the rest of their life to treat diabetes. You've probably seen this pattern in people you know, and it's a great path for patients to become life-long clients of big pharma.

Progress is not a straight line. This is the one that kills many traders' hopes and dreams--accomplishments and successes seem to be almost immediately reversed the next week. What's really happening here is the system (you, in this case) finding its new level with some back-and-forth noise. I mentioned exquisite sensitivity to short-term signal before--this is what allowed me to see shifts in the data before normal statistical measures flagged a change and before AI would accept it was real. However, this short-term intuition has to be baked into longer-term perspective and oversight. If it seems like I'm arguing for and against myself, you're right, and it's precisely this tension that's necessary to understand this work.

I don't believe the specifics of what I have done or what I'm dealing with are relevant to anyone else. I'm not an expert and I'm not your guru. You'll need to figure out what's going on in your "meat suit" and then figure out what needs to be done. I've given you the framework for the right tests, and asking further questions of a good AI can help move you along. Share this blog post with that AI, as well. (Also, and of course, you'll need to make adaptations for your own body. An old injury, for instance, will change what you can and should do.)

So how does this connect to trading?

For me, this work has already had meaningful benefits to my health and perceived well-being--I feel much better--but it has also reinforced some trading lessons for me. A few key takeaways:

You really can't change what you aren't measuring. This is a famous quote (of disputed provenance), but the idea is solid. A corollary might be something like "if you want to change something, figure out how to measure it." More data is not always better. I discarded more potential measures and datastreams than I kept, but a big part of the journey is figuring out what, exactly, you want to change and how you will know when it is changing.

For traders, the bottom line is not always the best measure. Results are confounded in the short term by random fluctuations, market regime, and other inputs. Measuring what you can, upstream of final results, just might be the key to success. Figure out how you can measure process as well, because this is the key to managing process.

Baselines matter. You must understand what you can normally expect to see, both in terms of trend and variation, so you can understand what impact your inputs might be having. Don't measure useless stuff, or, to say it more constructively, make sure what you're measuring is connected to what you want to change.

Feedback loops drive everything. Figure out how to harness them or become a victim.

Noise is not just an irritation; it's a critical part of the equation. You probably can't directly measure the thing you want to measure. Your measurements will always be obscured by random fluctuations on top of and around the real thing. A single datapoint is probably nothing--unless it is something.

The right response to noise is not to ignore it; it's to ignore it selectively and correctly. Finding that tuning where you can still pay attention correctly without being blown hither and thither is the entire goal. This is where much of the art and maybe most of the results live. Have a plan and modify it accordingly, without responding too much to meaningless blips. A really good system will somehow figure out how to separate the noise and make it visible as its own thing. This is not always possible, but it's a good goal to have in mind.

Last, you're going to fuck up, probably often. I think the element of trading psychology is misunderstood and somewhat overemphasized by people who want to sell you solutions to problems that are much simpler than they seem. But a key is finding a balance in your own psychology where you face your failings realistically but you don't beat yourself up so much you want to quit. Again, it comes back to finding that balance between flexibility and keeping your eye on the ball. You won't always get it right, but with the right guidance system you'll be able to stay on the path and get where you want to be, despite missteps and mistakes along the way.

Your process should be capable of proving you wrong. Otherwise, all of your work is nothing more than justification for your beliefs. The scientific method belongs to all of us. Use it. Frame your ideas as hypotheses and actively look for evidence that proves you are wrong. "Active" is the key word--devote a meaningful percentage of your efforts and attention to proving yourself wrong, to breaking the thing you are building. This framework is the key to trading success, and, just maybe, to being better in everything we do.

These are the answers I could never give so clearly. The value of the trading mindset is not in skills that will transfer to another career, but it's in a way of thinking that seeks truth and modifies our behavior to align with that truth.

I am a writer. The ideas, structure, writing, and mistakes in this post are my own. I use AI as I would use a capable human editor--mainly for line editing and fact checking. I remain deeply committed to human creative work and am equally committed to using the tools that support this work well. Images accompanying these posts may be generated by AI.