i Short answer
Across every large study of trading account records, women end up with better net results than men: by close to a percentage point a year in the original US brokerage data, and by between half a point and two points a year on the online platforms studied since. The gap is not analysis. Men in the same studies picked investments no worse than women did; they traded them more often, with bigger and more leveraged positions, and held losers longer. The costs of those three habits are the whole difference.
That makes this a page about behaviour rather than gender. The three habits have names, overtrading, overconfidence in sizing and the disposition effect, and each has a number on this site that shows what it costs in rand. The "women outperform" result is what trading looks like when those three costs are smaller.
📋 ON THIS PAGE
1. The finding, and how solid it is
The starting point is Barber and Odean's 2001 paper, which examined the accounts of more than 35,000 households at a US discount broker over six years. Men traded 45% more than women and earned net returns about 0.94 percentage points a year lower; single men traded 67% more than single women and earned 1.44 points less. The gross returns before costs were similar, so the gap was almost entirely the extra trading. Later work on online platforms in Britain and the United States, with samples running into the hundreds of thousands, found the same shape: women traded less, held longer and ended ahead, by amounts that depended on the period and the platform but kept the same sign.
These are account records, not surveys, which is why the finding is treated as settled in the behavioural finance literature. What the studies do not show is any difference in the quality of what was bought. The difference is in how much it was traded and how large the positions were.
2. Habit one: trading less
Every trade pays the spread, and on rand pairs the spread is large. A trader who places 40 round trips a month on USD/ZAR at a 150-pip spread pays R72,000 a year per lot; at 12 round trips the bill is R21,600. The spread cost calculator shows the figure for your own instrument and size. The extra trades in the studies did not produce extra gross return, so the cost came straight off the net result. Overtrading is the pattern in full, and the signal that you are in it is a trade count that rises after losses.
3. Habit two: smaller, less leveraged positions
Overconfidence shows up in position size before it shows up anywhere else. The men in the studies took larger positions relative to their accounts and, on the leveraged platforms, used more leverage. At 30x on USD/ZAR a 1.67% adverse day triggers a margin call, which the pair produces about nine days a year; at 10x the figure is 5%, which happens less than once a year. The leverage sleep test turns any leverage into that count of nights, and the risk of ruin calculator shows why the survival difference compounds into a return difference over time.
4. Habit three: closing losers
The disposition effect, selling winners early and holding losers in the hope they recover, is present in both groups in the studies but stronger in men, and it is the habit with the slowest, largest cost: a position that should have been closed at a 2% loss becomes a 10% loss that needs an 11% gain to recover. The drawdown recovery calculator puts a number on the recovery; the disposition effect article explains the bias and the rules that defeat it, chiefly a stop set before the trade and never moved further away.
5. What the gap looks like in rand
| Habit | Heavier version | Lighter version | Difference |
|---|---|---|---|
| Trades a month on USD/ZAR, 0.5 lot | 40 (spread R36,000) | 12 (spread R10,800) | R25,200 |
| Leverage and margin-call days | 30x, about 9 days | 10x, under 1 day | Several forced exits avoided |
| Average loss per losing trade | 2.5% of account | 1% of account | Half the drawdown to recover |
None of the three rows requires a better forecast. Together they are worth more than most strategies, which is the practical reading of the research: the gender gap is a measurement of how much three habits cost, taken on a very large sample.
6. What this does not mean
It does not mean women are better traders by nature, and the studies are careful on this point: the differences are averages across groups that overlap almost entirely, and the women who overtraded and over-leveraged lost like the men who did. It does not mean trading less is always right; a strategy with a genuine edge per trade earns more by trading it more, which the expectancy calculator makes explicit. It means that for the typical retail account, where the edge is small or unproven, the costs of overtrading and leverage dominate, and the group that incurred less of them finished ahead. The luck or skill calculator tests whether your own edge is real enough to justify the activity.
★ Why It Matters
The research is not a verdict on who should trade. It is a very large natural experiment showing what three habits cost, and the group that happened to have less of them finished ahead. Count your trades, size by risk and close losers on the rule, and you have copied the result whatever your gender.
✕ Common mistakes
- Reading the studies as "women pick better trades". Gross returns were similar; the gap is costs. The lesson is about activity and size, not selection.
- Trading less without checking the edge. If a tested strategy has a real edge per trade, fewer trades mean less profit. Trading less helps when the edge is small or unproven, which is most retail accounts.
- Fixing one habit and keeping the other two. Fewer trades at 30x leverage with no stop still ends in a margin call. The three habits compound; so does fixing them.
Key Takeaways
- Account studies consistently show women with better net results than men, by about 0.5 to 2 percentage points a year.
- The gap is costs: more trades, larger leveraged positions and held losers, not worse selection.
- Each habit has a rand cost you can compute on this site for your own account.
- The differences are group averages; individual behaviour is what matters.
- Trading less only helps when the edge per trade is small or unproven, which is the usual case.
See also: How Many Women Trade Forex in South Africa, and What Do the Numbers Say?, How Do I Avoid Overtrading as a Day Trader Specifically?, What Is the Disposition Effect and How Does It Affect Traders?.
Frequently asked follow-up questions
Is the "women outperform" result real or a media simplification?
Real, within its meaning. Several large account studies since 2001 find women's net returns ahead of men's by a modest margin, with the difference explained by trading frequency, position size and leverage rather than by what was bought.
How big is the difference?
About 0.94 percentage points a year in the original US brokerage study, and between roughly half a point and two points a year in later platform studies, depending on the period. Compounded over a decade, that is a meaningfully larger account.
Does it apply to South African CFD traders?
No South African account study has been published, but the mechanism is costs and leverage, and rand pairs have wider spreads and higher available leverage than the instruments in the studies. If anything, the cost of the three habits is larger here.
What is the single most useful change?
Count your trades. Most people who overtrade do not know their monthly count; a journal with that one number, next to the month's spread cost from the calculator, changes behaviour faster than any rule.
Are there studies where men did better?
In strong bull markets, more leverage and more activity can produce higher gross returns for a period, and some short-window platform figures show that. Over multi-year windows that include a downturn, the net result reverses, because the same habits produce the larger losses.
Is overconfidence really the cause, or is it risk tolerance?
The studies attribute the extra trading to overconfidence, measured by the gap between expected and realised results, and the position sizing partly to risk tolerance. Either way the fix is the same: a trade count and a risk-per-trade rule set in advance.
