There is a particular kind of trader who has been at this for three years and still cannot tell you, without guessing, what their average loss is.

They can tell you a great deal else. They know which order block they prefer. They know why the 9 EMA works better than the 20 on a 5-minute chart. They have opinions about liquidity sweeps and can spot a fair value gap at a glance. They have watched hundreds of hours of content on entry models.

Ask them what percentage of their losing trades broke a rule they had already written down, and the room goes quiet.

This is the gap. Not knowledge. Not entries. The gap is that almost nobody measures their own behaviour, and behaviour is where the money leaks out.

The entry obsession has a reason

It is worth understanding why entries absorb so much attention, because the reason is not stupidity.

Entries are the only part of trading that feels like a solvable problem. An entry has a clear right answer in hindsight. You can look at a chart, see where price turned, and know with certainty that the perfect entry existed at that exact candle. That certainty is addictive. It suggests that with enough study, you could have caught it.

Everything else in trading is probabilistic and slow. Position sizing does not produce a dopamine hit. Reviewing last month's trades does not feel like progress. Waiting for a setup that never appears feels like failure, even when it is discipline.

So traders optimise the one variable that offers clean feedback, and neglect the ones that actually compound.

There is also a supply problem. Entry content is easy to produce and easy to sell. A course promising a 90% win rate entry model will always outsell a course about record-keeping, even though the second one would help more people. The market gives you what you click on.

What the numbers actually say

Consider two traders with identical strategies.

Trader A has an entry model that wins 45% of the time with an average R-multiple of +2.0 on winners and −1.0 on losers. Over 100 trades: 45 winners at +2R and 55 losers at −1R gives +35R. Solid.

Trader B has a better entry model — 52% win rate, same R structure. Over 100 trades: 52 winners at +2R and 48 losers at −1R gives +56R. Better on paper by a wide margin.

Now add reality. Trader B takes 30 extra trades a month that do not match their plan — the impatient ones after a missed setup, the revenge trade after a stop-out, the oversized position taken to make back yesterday. Those trades run at a 25% win rate, and because size creeps up on them, the losers cost roughly 1.4R instead of 1.0R.

Thirty off-plan trades: 7 winners at +2R = +14R, 23 losers at −1.4R = −32.2R. Net −18.2R.

Trader B's superior entry model produced +56R. Their behaviour gave back 18.2R of it. Trader A, with the worse entry and the discipline to stop trading when the setup was absent, finishes ahead.

The numbers are illustrative, but the shape is not. The distance between a mediocre entry and a good one is smaller than the distance between following your plan and not following it. Almost every trader who has kept honest records for a year discovers the same thing: the strategy was never the problem.

The trades you don't remember are the expensive ones

Here is the mechanism that makes this invisible.

Human memory is not a neutral recorder. It preserves what is vivid and discards what is routine. The trade you remember from last month is the one that ran four R and made your week, or the one that gapped against you and stung. The eleven mediocre trades in between — the ones you took because you were bored at 11 a.m. and the chart looked sort of like your setup — leave no trace.

Those eleven are where the account bleeds. Individually small. Collectively decisive. And completely absent from your mental model of how you trade, because you never encoded them in the first place.

This is why "I know what my mistakes are" is almost always false. You know what your memorable mistakes are. That is a biased sample, and it is biased in exactly the direction that prevents you from fixing anything.

A journal is not a productivity ritual. It is a correction to a specific, well-documented failure of memory. It captures the boring trades — which is precisely the point, because the boring trades are the ones you cannot recall and therefore cannot address.

What a journal has to record to be worth keeping

Most abandoned journals were abandoned for a good reason: they recorded the wrong things.

A spreadsheet with entry price, exit price, and P&L tells you nothing you could not get from your broker statement. It takes ten minutes a day and returns no information. Of course you stopped.

A journal earns its time when it captures the decision, not just the outcome. That means three categories of data.

The mechanics

Symbol, direction, entry, stop, target, size, timestamps. This is the skeleton. It matters less for insight than people assume, but you need it to calculate anything — R-multiples, expectancy, profit factor, hold time distributions.

One detail that is routinely skipped and shouldn't be: the stop and target you set before entry, not where you actually got out. Without the planned levels, you cannot tell the difference between a trade that hit its stop and a trade you closed early out of nerves. Those are entirely different problems with entirely different fixes.

The reasoning

Why did you enter? Not the setup name — the actual reasoning. "4H bullish FVG respected, waited for retracement into 5M equilibrium, 1M BOS confirmed, SMT divergence against NQ" is useful. "Long setup" is not.

Why did you exit? Target hit, stop hit, closed early, moved to break-even — and what you were thinking when you did it.

This is the section that feels tedious and is worth the most. Written entry reasoning is a lie detector. It is very hard to write "I entered because price was moving and I did not want to miss it" and still believe you were following a plan. The act of writing it down is itself a filter — a surprising number of traders report that keeping a journal reduced their bad trades before they had reviewed a single entry, simply because they knew they would have to write down the reason.

The rule check

This is the part almost nobody does, and it is the part that changes outcomes.

For each trade, against your own written plan: which rules did you follow, and which did you break? Not a vague grade. A specific list.

Answer those five questions on every trade for two months and you will have something no indicator can give you: a measured, ranked list of exactly which of your own rules costs you the most money.

For a fuller breakdown of the fields worth capturing, see what to log in a trading journal.

The metric that matters most is one you have never seen

Every trading platform shows you win rate, P&L, and maybe profit factor. All outcome metrics. All backward-looking. All heavily influenced by variance over any sample you are likely to have.

The metric that actually predicts your future results is rule compliance — the percentage of your trades that followed your own plan.

It has a property none of the outcome metrics share: it is entirely within your control. You cannot decide to win more trades. You can decide to stop taking setups that do not meet your criteria. Compliance is the only number on your dashboard that responds directly to a decision you make.

It also stabilises far faster than P&L. Twenty trades tells you almost nothing about whether your edge is real — the confidence interval on a win rate at that sample size is enormous. But twenty trades tells you a great deal about whether you are disciplined, because compliance is measured per trade rather than inferred from a distribution.

And critically, it separates two things that outcome metrics fuse together: a good trade that lost, and a bad trade that won.

A trade that followed every rule and hit its stop is a good trade. The market did not cooperate. Nothing needs fixing. If you treat it as a failure and start adjusting your strategy, you will destroy an edge that was working.

A trade that broke three rules and made money is a bad trade. It rewarded a behaviour that will cost you badly over the next fifty repetitions. If you treat it as a success, you have just trained yourself to do the wrong thing.

Outcome metrics get both of these exactly backwards. Only compliance gets them right — which is why it belongs at the centre of any journal that intends to change how you trade rather than just record what happened.

What changes when you actually do this

Traders who keep a compliance-focused journal for a few months tend to report the same sequence.

First, the volume drops. Knowing you will have to write down the reason kills a meaningful share of impulsive entries at the moment of temptation. Nothing else is required — the friction alone does the work.

Then the pattern surfaces. Somewhere around trade forty or fifty, a rule violation starts appearing repeatedly. Almost always it is one specific thing: entering before the retracement completes, or sizing up after a loss, or trading a session you had already decided to avoid. It is rarely what the trader expected.

Then the cost gets quantified. Once violations are tagged consistently, the arithmetic becomes trivial. Not "I should be more patient" but "not waiting for the retracement cost me $2,140 across nine trades this quarter." That number changes behaviour in a way that generic advice never has.

Then the strategy questions become answerable. With a clean separation between disciplined trades and undisciplined ones, you can finally evaluate your edge honestly. Filter to compliant trades only. Is the expectancy positive? If yes, you do not have a strategy problem — you have an execution problem, and you now know precisely which execution problem. If no, you have learned something genuinely valuable: your plan needs work, and no amount of discipline will save it.

That last distinction is the one that ends years of circular searching. Most traders cannot answer it because their records mix both categories together, so every drawdown looks like evidence the strategy is broken. It usually isn't.

The honest version of the argument

Entries do matter. A plan with no edge cannot be rescued by discipline, and anyone who tells you otherwise is selling something. If your compliant trades lose money over a large sample, the problem is the strategy and you should work on it.

But that is a testable claim, and testing it requires — records. You cannot know whether your edge is real until you can isolate the trades where you actually applied it. Which brings you back to journaling, from the other direction.

The realistic picture is this: most traders searching for a better entry already have an entry good enough to be profitable. What they lack is evidence, because they never measured. So they keep changing the variable they can see instead of the one that is costing them, and every strategy change resets the sample size to zero.

That loop can run for years. Journaling is what breaks it — not because writing things down is virtuous, but because you cannot fix what you have never measured.

Ready to see where your own rules are costing you? TradeProof AI grades every trade against your written strategy, tracks your compliance over time, and shows you exactly which rule you break most — and what it costs. Log your first trade free.