Your worst losing streak has not happened yet

A run of losses feels like bad luck. It is closer to arithmetic. At a 45% win rate, six losses in a row is the normal outcome of a hundred trades, not the unlucky one, and you can work out how much it costs before it happens.

Most traders size for the average trade. You know roughly what you win and roughly what you lose, you pick a percent that feels survivable, and you get on with it. Then the run arrives. Not a catastrophe, not a black swan, just six or seven ordinary losses with nothing between them, and suddenly the percent that felt survivable has taken a fifth of the account.

The run was always coming. That is what a win rate under 50% means. The only question is whether you had worked out its size while you were calm or found out during it.

How the chance is worked out

Take a 45% win rate over a hundred trades. The chance of hitting a run of at least six losses somewhere in there is 72.9%. A run of seven is 49.7%. A run of ten is about one in ten.

Those numbers are computed exactly, by walking every way the wins and losses can fall and adding up the arrangements that contain a run. Two shortcuts were available and both were rejected.

The first is simulation: deal ten thousand random hands and count. It is easy and it is wrong here, because it returns a slightly different answer every time you load the page, and a number that moves on refresh is not a number you can size on.

The second is a textbook formula for the expected longest run. It is one line and it is an approximation that only settles over very long runs of trades. Every trading record is small, which is exactly where that approximation is worst. It also returns a fraction, and telling someone their expected worst streak is 5.8 losses is precision about a thing that only comes in whole numbers.

A number that changes when you refresh the page is not a number you can size on.

What a run costs

Losses compound against a shrinking account. Ten losses at 2% each is not 20%, it is 18.3%, because the second 2% is 2% of what the first one left. That works in your favour, and it is the optimistic reading.

The pessimistic reading is the one that matches how people actually trade. If you risk a fixed amount of money per trade rather than re-sizing to a percent as the account falls, ten losses costs the full 20%. The calculator shows the compounding figure and treats it as the floor, because most traders do not shrink their position after every loss, and the ones who do it by accident are usually the ones who cannot face looking.

Reading it backwards

The same arithmetic runs the other way. Name the worst drawdown you are willing to sit through, and it says what risk per trade keeps a normal run inside it. At a 45% win rate over a hundred trades, holding that run to a 10% drawdown means risking about 1.74% a trade. Holding it to 20% means about 3.65%.

The tool shows that as a table with no row marked, and the omission is deliberate. Which drawdown you can live with depends on your account, your income, and what else that money is for. Travise does not know any of those and will not pretend to. It says what each choice costs and leaves the choice where it belongs.

What it assumes, and where that breaks

Every calculation here treats each trade as independent of the last one. Real trading is not. One bad regime, one tilted afternoon, one revenge entry straight after a loss, and the losses arrive in a clump rather than spread out.

That error only runs one way, which is the important part. Clustering makes real runs longer than independent arithmetic predicts, never shorter. So read every number here as a floor. Your true worst run is at least this long, and if you are the kind of trader who sizes up after a loss, it is meaningfully longer.

There is one number that beats all of this, and no calculator can give it to you: the longest losing run already sitting in your own history. That one is not a model. It happened. If it is longer than the run this page called normal, the page is the thing that is wrong.

Where the idea comes from

Benjamin Graham called this a margin of safety, and put it at the centre of everything he wrote about investing. His own description of what it is for is the useful part: the function of the margin of safety is that of rendering unnecessary an accurate estimate of the future.

Graham was talking about paying less for a business than it was worth, and he would not have recognised most of what the people reading this page do for a living. The transfer is a borrowing, not a lineage. What carries over is the shape of the idea: you do not need to forecast correctly, you need to be positioned so that forecasting incorrectly does not end you. For a trader that is not a discount to intrinsic value. It is a position size that survives the run you are going to get.