Most traders who struggle are not struggling because their read on the market is wrong. They struggle because trading discipline breaks down in four predictable places, and those four failures cost more, year after year, than any single bad forecast does. These failures are documented in academic data covering hundreds of thousands of real brokerage accounts, which means they are not personal flaws. They are the default setting.
This post walks through each failure the same way: the behavior, what it measurably costs, and the specific fix. No motivational language, no promises. Just the four places the wheels come off.
Failure 1: Trading more than your edge justifies
The behavior. You take the setup you planned. Then you take the one that looked “close enough.” Then a third, because the first two are working and you feel sharp. Nothing about this feels reckless in the moment — it feels like being engaged.
The cost. Brad Barber and Terrance Odean studied 66,465 household brokerage accounts from 1991 to 1996 in a paper bluntly titled Trading Is Hazardous to Your Wealth. The average household turned over 75% of its portfolio each year and earned 16.4% annually. The most active fifth earned 11.4% while the market returned 17.9%. Their gross returns were roughly the same as everyone else’s. The entire gap was activity — costs and timing, compounded.
The fix. Cap the number of positions you are permitted to open in a week, and write the number down before the week starts. A cap is unpleasant precisely because it binds on the days you feel most confident, which is the point. If you are consistently hitting the cap by Wednesday, that is data about your filtering, not an argument for a bigger cap.
Failure 2: Selling what is working, holding what is not
The behavior. A position is up. You close it, because a small realized gain feels like proof you were right. Another is down. You hold it, because closing it converts a paper loss into an admission. This is the disposition effect, and it is one of the most robust findings in behavioral finance.
The cost. Odean’s 1998 study of 10,000 discount brokerage accounts found investors were 1.5 to 2 times more likely to sell a winner than a loser, even after controlling for taxes and rebalancing. Worse, the winners they sold went on to outperform the losers they kept by roughly 3.4% over the following year. They were systematically keeping the wrong half of the book.
The disposition effect is not an information problem. Every one of those traders had the same screen in front of them. It is a bookkeeping problem in the head — an unrealized loss is treated as not-yet-real.
The fix. Define the exit before the entry, in writing, and make it price-based rather than feeling-based. For overnight positions this matters more than it does intraday, because your exit is not actually available while you sleep — a topic worth understanding in detail if you have not read why overnight stop-losses do not protect you the way you assume.
Failure 3: Sizing up to get it back
The behavior. Revenge trading rarely looks like rage. It looks like a reasonable-sounding adjustment: the last two were losers, the setup now is clearly better, so this one gets 1.5x the usual size. The size increase is justified by conviction, and conviction is elevated precisely because you are behind.
The cost. The recovery arithmetic is unforgiving, and escalating size after losses is the fastest route down it. The evidence that traders correct this on their own is also weak: Barber, Lee, Liu and Odean studied the full population of Taiwanese day traders from 1992 to 2006 and found only about 20% turned a profit net of fees, with a follow-up study showing most kept trading anyway rather than updating on the results.
| Drawdown | Gain required to break even |
|---|---|
| 10% | 11.1% |
| 20% | 25.0% |
| 30% | 42.9% |
| 40% | 66.7% |
| 50% | 100.0% |
The fix. Fix your size as a percentage of account equity and change it only on a calendar, never on a result. If you want a stronger guardrail, add a stand-down rule: after two consecutive losing overnights, the next session is review-only. The rule has to be written before the losses happen, because that is the only time you are qualified to write it.
Failure 4: Having no record, and therefore no diagnosis
The behavior. You remember your trades as a narrative rather than a dataset. The narrative is heavily edited: the big win is vivid, the four small losses that funded it are a blur, and the trade you talked yourself into at 3:55 p.m. has quietly disappeared.
The cost. Harder to quantify than the first three, and it would be dishonest to attach a percentage to it. But the cost is structural: without a log you cannot tell the difference between a losing streak that is normal variance and one that is a repeated error. Both feel identical. One requires patience, the other requires a change, and guessing wrong is expensive either way.
The minimum viable trading journal
Elaborate journals get abandoned in a fortnight. Six fields, logged before the position is opened, are enough to catch all four failures above:
- Date and instrument — so streaks are countable rather than remembered.
- Position size as a percentage of equity — this alone exposes revenge sizing.
- Written reason for entry — one sentence, before entry, not after.
- Pre-committed exit — the price, written down while you are calm.
- Was this a planned setup or an improvised one? — a yes/no that quietly measures overtrading.
- Outcome, and whether you followed the plan — kept as two separate columns, because a profitable trade that broke your rules is a problem, not a win.
That last distinction is the one most traders skip and the one that does the most work. Reviewing plan adherence separately from profit and loss is also the spirit behind our own published audit of 24 pressure readings, including the sessions the model got wrong — a scorecard is only useful if it records the misses.
Where structure helps and where it does not
Every fix above is a form of pre-commitment: deciding while calm what you will do while stressed. That is also the honest way to use any external input, including SpyTomorrow’s overnight pressure readings — as one filter applied to a plan you already wrote, not as a reason to override the plan or to size up. A reading is decision support. It is not a forecast, and treating it as one simply relocates the discipline problem rather than solving it.
None of this makes the overnight session safer. Prices still move for reasons that have nothing to do with you, and knowing what actually sets SPY’s opening price while you are asleep is a separate education from managing your own behavior. But of the two problems, behavior is the one fully inside your control — and by the evidence, it is where most of the money goes.
Start with one trading discipline rule
Do not try to install four rules at once; that is its own kind of overconfidence. Pick the failure you recognized most uncomfortably while reading, write the single rule that addresses it, and log adherence for thirty sessions. Trading discipline is not a personality trait you either have or lack. It is a set of constraints written down in advance, at a moment when nothing is at stake.
This article is for educational and informational purposes only. It does not constitute financial advice, investment recommendations, or a solicitation to buy or sell any securities. Past performance and historical data do not guarantee future results. All trading involves risk, including the potential loss of principal. Always do your own research and consult with a qualified financial advisor before making trading decisions.
