Win Rate vs Profit Factor: Real Trading Data on a 70% Win Rate That Failed

Trader B won 70.3% of their trades and lost the account. Trader A won 65.6% and passed. The trader who was right more often is the one who blew up.

That is the whole lesson in two sentences, and most traders never learn it because they are watching the wrong number. Win rate feels like skill. It is mostly a vanity metric. What separated these two accounts was profit factor: whether the winners were bigger than the losers.

This study looks at two real two-step prop-firm evaluations from trading records we analysed. Every figure below comes from those records. Where we describe behaviour the data cannot directly show, we say so.

The two accounts at a glance

Metric	Trader A	Trader B
Evaluation size	$15,000	$25,000
Trades	192	353
Trading days	5	22
Win rate	65.6%	70.3%
Average win	$17.67	~$20
Average loss	$14.77	~$63
Profit factor	2.28	0.77
Result	+$1,203 – passed	−$1,720 – failed

Trader A cleared an 8% profit target in five trading days and stopped. Trader B traded for 22 days, spent most of them slowly bleeding, and then lost the account in a single afternoon and the morning that followed.

Win rate vs profit factor: the maths that decided it

Profit factor is gross profit divided by gross loss. Above 1.0, the account makes money. Below 1.0, it loses money, whatever the win rate says.

Trader A’s profit factor was 2.28. For every dollar lost, the account made $2.28. Trader B’s was 0.77. For every dollar lost, the account made 77 cents back.

The reason sits in the payoff. Trader A’s average winner ($17.67) was about 20% larger than the average loser ($14.77). Not a dramatic edge, but enough. Trader B’s average winner was around $20 against an average loser around $63. Every loss erased roughly three wins.

Run the break-even arithmetic and the outcome was settled long before the final day:

• Trader A needed to win about 45% of trades to break even with that payoff. They won 65.6%. That is a cushion of roughly 20 percentage points.
• Trader B needed to win about 76% of trades to break even with that payoff. They won 70.3%. A 70% hit rate was not enough.

Look at it from the other direction. At a 70% win rate, a trader’s average winner only has to be about 43% the size of the average loser to break even. Trader B’s winners were about 33%. That gap, ten percentage points of payoff, is the difference between a slow grind upward and a slow bleed.

The result was an expectancy of about −$4.21 per trade. Trader A, by contrast, netted about $6.27 per trade across 192 trades. One of those numbers compounds toward a funded account. The other compounds toward a breach.

How Trader B’s account actually died

The bleed alone might not have ended the challenge. The acute event did.

After weeks of gradual losses, Trader B opened seven short positions on gold inside a single hour one afternoon. Four of them went on within about two and a half minutes.

• None of the seven had a stop-loss set.
• All were in the same direction, added while price rose against the existing position.
• The positions were held overnight, still without stops.
• Gold ran roughly 60 points against the stack. The next morning the firm’s system force-closed everything when the daily loss limit tripped.

The single worst trade in the account lost roughly as much as 13 average winners combined. A 70% win rate cannot absorb that. No win rate can.

One more detail from the records: across the challenge, Trader B paid $169 in overnight financing costs. Holding positions through the close is not only a risk decision. It carries a running cost that quietly widens the gap between average win and average loss.

It was not a Method problem

It would be easy to read this as a story about a bad trader. The data does not support that. Seven wins in every ten trades is a strong market read by any standard. Trader B was right about direction more often than Trader A.

What failed was everything around the read: where the trade was exited, how large the losers were allowed to grow, and what happened when a position went wrong. In M·M·M terms, the Method leg held up. The Money and Mind legs gave way.

The wider dataset points the same way. Across roughly 26 accounts in the records we reviewed, failures were mechanical rather than analytical. Six of those accounts finished net profitable and still failed on risk-rule breaches. They made money and lost the evaluation anyway.

Every catastrophic loss in the set shared one feature: a single day where trade count spiked and positions stacked.

Trade count on its own was not the divider. Trader A averaged roughly 38 trades a day and passed. Trader B averaged about 16 and failed. Activity was not the problem. Payoff and the stopless stacking spike were.

The Money leg: stop first, target second

Two rules would have changed Trader B’s outcome.

1. A stop in every ticket before entry. Not after the fill, not when the trade starts to hurt. A stop-loss set at entry caps the loss before emotion gets a vote. Seven stopless positions held overnight is the scenario the stop exists to prevent.

2. A target that pays for the risk. At Trader B’s hit rate, winners needed to be at least 1.3 times the size of the risk just to survive with a margin. That is the floor, not the goal. The CTE standard is a minimum 1:2 risk-reward ratio, because it leaves room for the losing streaks every strategy produces.

Without the second rule, a high win rate is a slow bleed dressed up as almost-winning. The equity curve shows green trades all day and red weeks at the end of them.

Fixed position size belongs in the same ticket. Adding size to a losing idea turns one bad read into a stack of them, which is exactly how Trader B’s final afternoon unfolded.

The Mind leg: discipline is a container, not a feeling

The death of this account was behavioural. Seven same-direction shorts in an hour, four inside about two and a half minutes, each added as price moved further away, none protected. That is not an analysis error.

Inference, not data: the pattern is consistent with revenge trading, adding to a losing position to be proven right rather than accepting the loss. We cannot see what the trader was thinking. We can see that the behaviour matches the textbook cycle, and that it arrived after weeks of slow losses, which is when that cycle tends to start.

Most traders treat trading discipline as a mood they hope to be in on the day. It works better as a structure built in advance, so it holds on the days the mood is missing:

• A maximum number of open positions per instrument.
• No adding to a position that is in drawdown.
• No trade leaves the ticket screen without a stop.
• A hard stop to the session after a set number of losses or a set daily loss, well inside the firm’s limit.

Those rules do nothing on a good day. They exist for the one afternoon that ends a challenge. For more on the pressure specific to evaluations, see our guide to funded trader psychology.

Profit factor over win rate

The counterintuitive truth from these two accounts is simple. The trader who won less often, but kept winners bigger than losers, is the one who survived. The trader who was right seven times in ten lost the account because the three wrong trades were allowed to be enormous, and because one afternoon of stacking removed every limit at once.

If you track one number from your own journal this week, make it profit factor. Then check your average win against your average loss, and work out the win rate you would need to break even. If that number is higher than your actual win rate, no amount of being right will save the account.

Run your own numbers through the prop firm risk calculator and the trading calculators before your next challenge, and take the M·M·M Assessment to see which leg of your trading is carrying the weight, and which is quietly giving way.