Fixed Fractional vs Fixed Dollar Position Sizing

Fixed fractional risks a constant percentage of current equity on every trade. Fixed dollar risks the same cash amount every time regardless of what the account is worth. The difference only becomes visible in a drawdown, and then it decides everything.

For almost everyone the answer is fixed fractional, because of one property: it cannot run the account to zero through arithmetic alone.

How each behaves as the account moves

Start with 50,000 and a 1% risk, which is 500 a trade.

Fixed fractional. After a 20% drawdown the account holds 40,000, and 1% is now 400. The cash at risk fell automatically. After a run up to 60,000 it is 600. The proportion exposed never changed.

Fixed dollar. After the same 20% drawdown the account holds 40,000 and the risk is still 500, which is now 1.25% of what remains. After a 50% drawdown it is 2% of what remains. The percentage exposed climbs exactly as the account’s ability to absorb it falls.

That is the whole argument. Fixed dollar sizing increases proportional risk during drawdowns, which is when increasing it is least affordable.

The arithmetic that makes it matter

Fixed fractional has a property worth stating plainly: a sequence of percentage losses never reaches zero. Lose 1% of what remains, repeatedly, and the account becomes very small without ever becoming nothing. In practice you stop long before that, but it means the method cannot destroy the account by itself.

Fixed dollar has no such floor. A fixed 500 taken from a declining account is an ever-larger share of it, and the sequence terminates. The last few losses before that point are each a substantial fraction of the remaining equity.

This is why the two methods diverge most sharply in exactly the scenario that matters, and why risk of ruin produces very different numbers for the same strategy depending on which one is used.

What fixed fractional costs

It is not free, and the cost is real.

Slower recovery. Risking 1% of a reduced account means smaller positions through the whole climb back, so recovering a drawdown takes longer than it would at a flat cash risk. The compensation is that there is an account to recover.

Shrinking position sizes compound downward. A losing run produces progressively smaller positions, so each subsequent win contributes less. This is the same mechanism that protects you, seen from the other side.

Awkward quantities. A percentage of equity divided by a stop distance rarely produces a round number of shares, and on small accounts it can fall below one share.

Where fixed dollar genuinely fits

Two cases, both specific.

Very small accounts. Below a certain size, 1% of equity divided by a sensible stop distance produces a position smaller than the minimum tradable quantity, or so small that commission dominates. A flat cash risk, set conservatively against the account size, is a workable bridge until the account can support percentage sizing.

Accounts with external cash flows. If you are drawing income from the account or adding to it regularly, equity moves for reasons unrelated to trading performance. Percentage sizing then reads a deposit as permission to trade larger, which it is not. A fixed amount, reviewed periodically against account size, separates the two.

Neither of these is an argument for fixed dollar in general. Both are reasons it can be the practical choice in a particular situation.

The hybrid most people actually run

A common and sensible arrangement: percentage sizing with a cash ceiling.

Risk 1% of equity, but never more than a stated maximum regardless of what the percentage produces. As the account grows, the percentage would eventually put a large absolute sum at risk on a single trade, and the cap keeps that within what you can watch lose without reacting badly.

That ceiling is a behavioural control rather than a mathematical one. It exists because the size at which a trader stops following their own rules is usually a cash figure rather than a percentage, and it is worth deciding where that is before meeting it.

Which equity figure the percentage uses

A detail that changes results more than it looks like it should: fixed fractional needs a definition of “equity”, and there are three candidates.

Closed equity counts only settled, closed trades. Open positions do not affect the figure until they are shut. It is the steadiest of the three and the slowest to react, so size does not change because an open position moved today.

Open equity marks everything to market, so unrealised profit immediately raises the size of the next position. This compounds fastest in a trend and is the most volatile, because a pullback in open positions shrinks the next trade at the same time.

Core equity counts the account minus the capital currently committed to open positions. It is the most conservative, and it has a useful property: it reduces position size automatically as more positions are opened, which pushes back against taking on too many at once.

Most traders should use closed equity. It avoids the feedback loop where an open winner funds a larger new position which is then exposed to the same market conditions, and it keeps the sizing input stable enough that the rule is easy to follow without recalculating constantly.

Whichever you pick, pick one and record it alongside the percentage. Two traders running “1% fixed fractional” on different equity definitions are running different systems, and a result compared across the two is not comparing like with like.

Choosing

Use fixed fractional. Set the percentage from your own drawdown tolerance rather than inheriting 1% as a default — the method for that is in set risk per trade from your drawdown limit — and add a cash ceiling once the account is large enough for the percentage to produce sums you would find difficult to watch.

Use fixed dollar only while the account is too small for percentages to work, and treat it as temporary.

Either way, the decision that matters more than the method is keeping the figure constant through a drawdown. The position size calculator handles the arithmetic, and expectancy and sample size covers why holding the number steady matters more to long run results than selecting entries well does.