Brokers offering leveraged products in regulated markets are required to publish what share of their retail customers lose money. The number comes back between 70% and 85% almost everywhere it's measured, across countries, across brokers, across bull markets and bear markets.
The comfortable explanation is that those people were lazy or stupid. It isn't true, and if you've traded you already know it isn't true. Plenty of them read more than you do. Some of them were right about the market and still lost money.
They lost on execution. They knew what to do and did something else instead. It happens reliably, in the same six ways, for reasons that are wired in rather than chosen.
The one thing to remember
Losing traders don't fail from a shortage of information. They fail at the moment of acting, when money is on the line and the brain does what brains do. That's why the fix isn't more knowledge. It's removing the moment of decision.
Losing hurts about twice as much as winning feels good
This is the foundation everything else sits on. Behavioral economists measured it decades ago: the pain of losing a given amount is roughly twice the pleasure of gaining the same amount. Not a metaphor, but a measurable asymmetry in how people value outcomes.
The consequences at a trading screen are immediate and awful. A position that's down $200 doesn't feel like a $200 problem. It feels like a $400 problem, and it demands to be solved now, which is why people close a losing trade by turning it into a bigger losing trade. Meanwhile a position that's up $200 generates only half as much feeling, so there's nothing pushing you to let it work.
You are not, at that moment, evaluating a trade. You are managing a sensation.
The six ways it actually goes wrong
| The failure | What it looks like | What it does to the account |
|---|---|---|
| Loss aversion | Holding a loser because closing it makes it real | Small losses become the ones that end accounts |
| The disposition effect | Taking a profit early, riding a loss late | Inverts your risk-reward ratio trade by trade |
| Revenge trading | Re-entering immediately, bigger, after a loss | Turns one bad trade into a bad week |
| Overtrading | Trading because you're at the screen, not because there's a setup | Commissions and slippage eat an edge that was thin anyway |
| Sizing by feel | More contracts when confident, fewer when scared | Biggest size on the least reliable trades |
| No written plan | Rules that live in your head and change with your mood | Nothing to be consistent with, so no consistency |
Take them one at a time, because each has its own logic.
Cutting winners and riding losers
Documented as the disposition effect, and it is the most expensive habit in retail trading. You take the $80 profit because a bird in the hand feels safe, and you hold the $200 loss because it's not a loss until you close it. Do that consistently and you have manufactured a system where your average winner is smaller than your average loser, which requires an unreasonably high win rate just to break even.
Compare it to what a working system looks like. The backtest published on this site has an average winner of $354 against an average loser of $193. That ratio, roughly 1.84 to 1, is what makes a 46.2% win rate profitable. A human doing the same trades on instinct would take the winners at $150 and let the losers run to $400, and the exact same entries would produce a losing account.
Revenge trading
You lose. The loss stings twice as much as it should. Getting back to flat becomes urgent in a way that has nothing to do with the market, so you take the next setup that looks vaguely acceptable, at double size, to make it back in one trade.
The market doesn't know you're behind and has no interest in helping you catch up. What actually happens is that you've now sized up on a trade you'd have skipped an hour ago, and if it goes wrong the urgency doubles again. This is how accounts blow up in an afternoon rather than a year.
Overtrading
An edge is thin. A good one might net you a fraction of a point per trade after costs. When you take fifteen trades a day because you're sitting there and the screen is on, you're not applying an edge fifteen times. You're applying it twice and paying costs thirteen times.
This is why trading one window a day is a feature and not a limitation. The alternative isn't more opportunities. It's the same opportunities plus a lot of expensive noise.
Sizing by feel
Ask someone how many contracts they trade and you'll often get "it depends." It depends on conviction. Which means the largest position goes on the trade they feel most strongly about, and feeling strongly is not correlated with being right. Often it's inversely correlated, because the setups that feel most obvious are the ones everybody else has already acted on.
Worse, size usually creeps up after wins and shrinks after losses, which is precisely backwards from a risk perspective. You end up largest right before the drawdown and smallest right through the recovery. Fixed sizing is dull and it is one of the few things in day trading that reliably helps.
No written plan
If your rules aren't written down, you don't have rules. You have preferences, and preferences bend. Every trader who has ever said "I don't normally do this, but" was about to break a rule they never actually wrote, which is what made it so easy to break.
A written plan can be tested. It can be judged. It can be found wanting and improved. A plan in your head can only be rationalized.
Most people don't lose because their plan was wrong. They lose because at the critical moment there wasn't one.
None of this is a character flaw
It's worth saying plainly, because the trading internet is full of people sneering at "weak hands." The behaviors above are not weakness. They're the standard output of a nervous system that evolved to treat loss as danger and to demand immediate corrective action when threatened. That machinery kept your ancestors alive. It's simply catastrophic when the threat is a number on a screen that will resolve itself in twenty minutes if you leave it alone.
Experience helps less than people expect. Professionals feel the same impulses; what they have is structure that makes acting on them difficult. Risk limits set by someone else. Position sizes fixed in advance. Rules enforced by a desk rather than by willpower. The discipline isn't internal. It's built into the environment.
Retail traders get none of that for free. They have to build it themselves, out of the same brain that's trying to break it.
What a machine does differently
An automated system doesn't have more discipline than you. It has no capacity for indiscipline, which is a different and much more reliable thing.
- It doesn't feel the loss. The stop gets hit, the trade closes, the next evaluation starts from zero. There's no residue.
- It can't cut a winner early. The target was placed in the market at the same instant as the entry, at twice the risk. There is no mechanism by which it changes its mind at 60% of the way there. The stop and target go in with the entry, as described in how it works.
- It can't revenge trade. The next trade takes exactly the same size as the last one, because size is a fixed parameter, not a mood.
- It doesn't trade out of boredom. When the conditions aren't there, it does nothing, for as long as nothing is the right answer.
- Its plan is written by construction. A strategy is literally a written plan. That's what the code is. It can't drift, because drifting requires a decision.
And it sits through the thing humans cannot sit through: a system that is wrong more often than it's right. A 46.2% win rate means losing streaks of four and five are ordinary arithmetic, not malfunctions. Almost nobody can watch that live without intervening. A machine doesn't experience a streak at all, only a sequence of independent evaluations. The full argument is in what a trading bot actually is.
The uncomfortable numbers are on the same page as the good ones
Win rate, average loser, maximum drawdown, worst months. If a system only shows you the equity curve, ask what it left out.
The mistakes don't vanish. They move
This is the part most automation pitches skip. Buying a system doesn't delete the psychology. It relocates it. The same six impulses come back wearing different clothes:
- Switching it off in the drawdown. Loss aversion, exactly as before, applied to the system instead of a trade. Three red days and the finger goes to the switch, right before the recovery.
- Tweaking the parameters. Cutting winners, in a new outfit. Tightening the stop because yesterday it got clipped turns the thing you bought into something with no track record at all.
- Adding contracts after a good run. Sizing by feel. The drawdowns scale with the size and your tolerance for them doesn't.
- Running a second system to fill the quiet hours. Overtrading. The boredom was never about the market.
The uncomfortable truth is that the machine is the easy part. Leaving it alone is the job. That is exactly the territory covered in the mistakes people make with automation, and it's worth reading before you own a system rather than after.
What honest expectations look like
A system that works is not a system that wins every day. The published backtest here runs 88 months on a $50,000 funded account, one micro contract: $274,406 gross, ≈$260,700 net after commissions, an average of roughly $2,900 a month, with a maximum drawdown of $4,379 along the way and plenty of losing months inside that record. Those are simulated results over historical data. No real money was at risk, and past performance, real or simulated, does not guarantee future results.
If your expectation is a smooth line upward, you'll switch off the system at the first valley and end up in the losing majority. If your expectation is a positive edge that arrives unevenly, you have a chance. The difference is entirely in what you decided before you started, which is why deciding how much you can afford to lose matters more than anything you'll read about entries. Put your own figures into the calculator and look at the bad scenarios first.
Hypothetical performance. The figures above come from a backtest over historical data, not from a live account. Simulated results are prepared with hindsight, carry no financial risk, and cannot fully reflect real execution, slippage or liquidity. Past performance is not indicative of future results.
Frequently asked questions
Is the 70% to 85% figure reliable?
It comes from the mandatory broker disclosures on leveraged retail products in Europe, the UK and Australia, which makes it more trustworthy than most trading statistics: the brokers publishing it have no incentive to make it look bad. It measures the share of retail accounts that lost money over a given period, so it counts people who lost a dollar the same as people who lost everything. The consistency of the range across very different firms and markets is what makes it worth taking seriously.
Can't I just learn discipline?
Some people do, and it takes years and a lot of tuition paid to the market. The more practical route is to change the environment rather than the person: write the rules down, fix the position size in advance, place the stop and target at entry, and remove your ability to intervene mid-trade. Automation is the strongest version of that approach, not a different one.
Does a trading bot guarantee I'll stop making these mistakes?
No. It removes the mistakes inside the trade and hands you a new set outside it: switching the system off during a drawdown, changing its settings, or increasing size after a good month. Those decisions are still yours, and they're where most automated accounts actually go wrong.
Why does a system with a 46.2% win rate make money?
Because the target is set at twice the risk. When winners are worth about 1.84 times losers on average, you only need to be right roughly a third of the time to break even, so 46.2% leaves a comfortable margin. It also means the system spends a lot of time losing, which is precisely why it's easier for a machine to run than for a person.
If most people lose, why trade at all?
Because the majority losing does not mean the activity is unwinnable. It means the default way of doing it fails. The honest answer is that trading is worth doing only with money you can afford to lose, a plan you wrote down, and expectations set by evidence rather than hope. If any of those three is missing, the statistics are describing you.
In short: the 70% to 85% who lose aren't uninformed. They're human, and human wiring is close to perfectly wrong for this task: hold losers, cut winners, size by feeling, trade to feel busy. A machine removes those failures from inside the trade, at the price of handing you new ones outside it. Knowing which ones are still yours is most of the work.
