How much to risk per trade, and how to compute it

Updated on September 11, 2026

The number, and why it comes second

One percent of capital per position. That is the answer you will find everywhere, and it is a sound one. Two percent shows up often among those who accept wider swings in their account. Below 0.5 %, costs and the spread between buying and selling prices start to weigh heavily against the gain you are aiming for; above 2 %, an ordinary losing streak becomes hard to absorb, which we come back to below.

Yet this argument about the number takes up most of the discussion while being the less important of the two subjects. Someone applying 2 % rigorously does far better than someone who believes they apply 1 % without ever having computed it.

Because the percentage is not a setting you choose in a platform. It is an outcome, the result of a division nobody enjoys doing at the moment of entering a position.

The calculation, in three numbers

You need three things, and only three. The amount you accept losing, meaning your capital multiplied by your percentage. The distance between your entry and your stop. And the value of one point of that distance for one lot of the instrument you are trading.

Your size is then the accepted amount, divided by the distance to the stop multiplied by the point value.

An example with round numbers. A 5,000 $ account, 1 % risk, so 50 $ accepted on this position. You buy the euro against the dollar at 1.0850 with a stop at 1.0820, which is 30 points of distance. On that pair a point is worth roughly 10 $ for a standard lot, so your stop would cost 300 $ if you took a full lot. Your size is 50 divided by 300, about 0.16 lot.

Change any one of those three values and the result changes. A stop twice as wide gives a size twice as small, for exactly the same risk in cash. That is the whole point of the method: it makes your positions comparable to one another, where a fixed size makes them incomparable.

One last point, and it is the one that discourages doing this by hand: the point value depends on the instrument, on your account currency and sometimes on the moment. It is not the same on gold, on an index or on a yen pair. So it has to be looked up every time, on the exact symbol your broker exposes to you.

Why almost everyone does the opposite

The common reflex is to open the order window, leave the usual size, often 0.10 lot because that was the last one, then place the stop wherever the chart suggests.

The order of operations is reversed, and the consequences are not trivial. On a position with a tight stop you may risk 15 $, on the next one with a wide stop 120 $. Your gains and losses are no longer comparable, and your monthly total depends more on where your stops happened to sit than on the quality of your decisions.

It also makes any review impossible to read. Six winners and four losers can perfectly well add up to a negative month if the losers carried three times the size. You then conclude that your method does not work, when it was your sizing that spoke louder than your method.

What 1 % means over a streak

The real argument for a small percentage is not caution, it is the arithmetic of recovery. A loss is not undone by a gain of the same percentage, because the gain applies to a capital that has shrunk.

• losing 10 % requires making 11 % to get back to par;
• losing 30 % requires 43 %;
• losing 50 % requires 100 %.

Apply that to a losing streak and the choice of percentage becomes concrete. Ten losing positions in a row at 1 % leave the account at minus 9.6 %, which is recovered without changing anything. The same ten positions at 5 % leave it at minus 40 %, and you then need 67 % to break even.

Ten consecutive losses are nothing exceptional across several hundred positions, including with a method that wins more often than it loses. A small percentage is therefore not a sign of timidity: it is what leaves you an account still standing by the time your method turns favourable again.

The traps that make the percentage lie

The first, and by far the most expensive, is correlation. Three positions at 1 % opened at the same time on the euro, the pound and the Australian dollar against the dollar are not three independent 1 % risks. It is one bet on the dollar, and it is worth close to 3 %. The calculation is right position by position, and wrong at the account level.

The second is the moved stop. A stop pulled back because price is approaching it retroactively turns your 1 % into whatever you like, and usually into a great deal more. The percentage only means something if the stop is treated as a decision taken before entry.

The third is that a stop is not a guarantee. On a price gap, at the Sunday open or on an economic release, price can jump straight past your level and fill you further away. Your 1 % can become 2 or 3 without anything having gone technically wrong.

The fourth is a stubborn confusion with leverage. Leverage does not set your risk, it sets the margin tied up. Two accounts on different leverage taking the same size with the same stop risk exactly the same amount. Distance to the stop and size decide, never leverage.

How AlphaGPT handles it

Since this is our subject, we may as well say what we do with it. You set your percentage once. From then on, every analysis reads the real equity of your account at your broker, takes the distance between entry and stop from the proposed plan, asks your broker for the point value of the exact symbol involved, and displays the size that follows.

The gain is not really time, it is elsewhere: the division always happens, including on the days you would not have felt like doing it. And it happens on your equity right now, not on the capital you had in mind.

This calculation also works in read-only mode, with an investor password, since it only needs to consult the account. If that distinction means nothing to you, it is explained in our guide on the MT5 investor password.

This page describes a method of calculation, not investment advice, and trading carries a risk of capital loss.