The tell is always the same. Someone shows me a position they are proud of, the thesis is genuinely good, and then somewhere in the conversation they mention that they woke up at 3am to check the price. Not once. Most nights. That detail tells me more about whether the trade will work out than the thesis does, because a position you babysit at 3am is one that gets closed at the worst possible moment, and it will feel like a decision when it is really just your body tapping out.
I want to talk about the gap between the size the math says you can carry and the size you can actually carry as a human being with a stress response. Those are two different numbers, and almost everyone sizes to the first one and then quietly suffers.
The math is not wrong, it is just answering a different question
If you have run the numbers on a trade, you probably have some version of a Kelly fraction floating around in your head. Edge over odds, the fraction of your bankroll that maximizes long run growth. It is a beautiful piece of math and it is correct. The problem is what it assumes about you. Kelly assumes you will hold the position exactly as planned regardless of what the equity curve does in the middle, that you will not flinch during the drawdown that full Kelly all but guarantees, and that a 40 or 50 percent paper loss on the way to being right is emotionally identical to a rounding error.
That assumption is false for basically everyone, which is why serious practitioners size at half Kelly or a quarter Kelly and treat the full number as a ceiling they will never touch. The math does not know that you have a mortgage, or that this is the position you told your partner about, or that you have been burned before and the scar tissue changes how you read a red candle. It optimizes growth for a version of you that does not exist. The emotional oversizing problem is not that people ignore the math. It is that they follow it too literally and forget the math never priced in their nervous system.
What oversizing actually feels like from the inside
The insidious part is that oversizing does not announce itself as fear. It shows up as a subtle loss of objectivity that you will rationalize the whole way down. Here is what it looks like in practice, and if two or three of these are true for a position you hold right now, you are probably too big.
- You check the price at times you would never check it for a position of half the size. The 3am wake up, the phone glance during dinner, the tab you refresh out of habit.
- Your mood tracks the tick. A green hour makes the whole day feel good, a red one sits in your chest. The position is renting space in your emotional life that it has no business renting.
- You have stopped reading news that might be bad for the thesis. Or you read it and immediately find reasons it does not count. That is not analysis, that is defense.
- You catch yourself doing math on what it would be worth if it doubled, more often than you think about where your invalidation actually sits.
- You would be embarrassed to tell another trader the real size, and you know exactly why.
Compare all of that to how you think about an asset you have researched but do not own. You can hold two contradictory scenarios in your head at once, you weigh the bear case fairly because nothing you own is threatened by it, and you sleep fine. That calm, detached quality is what good analysis feels like. The moment your own position corrupts it, your edge on that trade is quietly gone, because you can no longer see it straight.
The calibration is a slider, not a switch
The practical fix is almost stupidly simple, which is probably why people resist it. You size the position down until your analysis of it becomes as objective as your analysis of the assets you do not own. That is the whole test. You are not solving for maximum growth, you are solving for the largest size at which you can still think clearly.
Here is how I actually run it. Put the position on at the size the math suggested. Then, over the next few days, watch your own behavior the way you would watch a stranger's. Are you checking it more than you would a neutral holding? Does a normal 10 or 15 percent wiggle, the kind every real position has, change how you feel about the thesis? If the answer is yes, cut the size. Not by a token amount, cut it in half. Then observe again. Keep halving until the position goes quiet in your head, until you can look at a bad day and shrug because the dollar move is small enough that your judgment stays intact.
Two things usually happen when people try this. First, the size that lets them sleep is dramatically smaller than they expected, often a quarter or less of what they started with, and that number is uncomfortable to accept because it feels like admitting weakness. It is not weakness, it is just the honest capacity of your particular temperament, and it is worth knowing. Second, their returns on those positions tend to improve. Not because smaller is magic, but because a position you can think about clearly is one you exit on your plan instead of your panic. The whale traders whose exits look so clean from the outside are, more than anything, sizing so they never lose the plot.
A few rules of thumb that survive contact with a bad week
Set your size assuming a drawdown roughly twice as deep as the one you are picturing, because the one you are picturing is always too shallow. If a position halved from here and you would be forced to sell to sleep, you are already too big regardless of what any model says.
Do the sizing when the market is closed and boring, never in the middle of a move. Conviction feels cheap and infinite when the tape is green, and your calm self is a better risk manager than your excited one. Write the size down before you enter, and if your live self wants to add because it is running, that urge is data about your emotional state, not about the trade.
And treat the 3am wake up as a hard signal, not a personality quirk to power through. When I built the market scorecards and whale tracking inside Blockcircle, the goal was to let people replace the anxious refresh with a system that watches for them, so the check is a decision on a schedule rather than a compulsion at 3am. But the tool only helps if the size is right first. No dashboard fixes a position that is too big for the person holding it.
The number you can actually live with is knowable. You just have to be willing to size down until you find it, and to accept that it is smaller than your ego wants it to be. A position you can sleep through is one you can also think about, and thinking clearly is the only edge that survives a drawdown.