Fixed Fractional vs Fixed Lot: Sizing That Really Compounds

You change nothing in the strategy. Same entries. Same exits. Same market. Yet two traders end up in completely different places because one uses fixed fractional and the other uses fixed lot.

This is a plain-English guide to fixed fractional vs fixed lot.

Same coin flip, different bet sizing.

Fixed fractional vs fixed lot sounds like a boring technicality until you see one equity curve bend upwards and the other just sort of… walk sideways.

Why everyone quietly loves fixed lots (at first)

You open a small account. Let’s say £1,000 for a concrete example.

You pick 0.01 lot on EURUSD or XAUUSD because the platform lets you, it feels small, and you’ve seen ten forum posts saying it’s “safe”.

That’s fixed lot position sizing: every trade is 0.01, win or lose, £1,000 or £1,500, doesn’t matter.

It feels wonderfully simple.

Risk per trade in pounds barely moves at the start, so your brain relaxes.

If your stop is 100 pips and each pip is, say, £0.10 on your 0.01 lot, the most you lose is about £10 per trade.

So in your head it’s, “Worst case, ten quid. I spend more than that on coffee and bad decisions.”

Then the balance changes.

At £1,000, that £10 loss is 1%.

At £500, the same 0.01 lot and same stop is 2%.

At £5,000, it’s 0.2%.

Same system. Same lot size. Completely different risk profile.

Fixed fractional vs fixed lot: the actual difference

Fixed fractional sizing says: risk a fixed percentage of current equity on each trade.

For example, risk 1% per trade.

If equity is £1,000, you risk £10. If it grows to £1,200, you risk £12. If it falls to £800, you risk £8. The position size is whatever lot size makes that pound risk match your stop distance.

Everything flexes with the account.

Fixed lot sizing says: trade 0.01 lots (or 0.10, or whatever) regardless of equity.

The pound risk then floats around as the account changes, and the percentage risk per trade drifts without you noticing.

So you think you run a “1% risk system” when really you run “0.3–4% depending on how badly it’s going”.

Which is a fun surprise to discover mid-drawdown.

Why fixed fractional actually compounds

Take a very simple, hypothetical system.

It wins 50% of the time. Average win is equal to average loss. So the raw expectancy is zero before costs.

This is a coin flip in disguise, just to isolate the sizing.

Now imagine two traders using this system for 100 trades.

• Trader A: fixed fractional, risks 2% of current equity per trade.
• Trader B: fixed lot, equivalent to 2% of the starting equity per trade.

Same entries. Same sequence of wins and losses.

Mathematically, Trader A’s equity process is multiplicative.

Each win multiplies equity by 1.02. Each loss multiplies equity by 0.98.

After 100 trades, with exactly 50 wins and 50 losses in any order, equity is:

Equity = 1,000 × 1.0250 × 0.9850

That cleans up to:

Equity = 1,000 × (1.02 × 0.98)50 = 1,000 × 0.999650 ≈ 981

A small loss (because spreads and costs drag it below zero expectancy).

Trader B keeps risking £20 per trade no matter what because the lot size is fixed.

With 50 wins and 50 losses, Trader B finishes almost exactly back at £1,000, minus costs.

So in a zero-edge system, fixed fractions actually do slightly worse than fixed lots because costs get to compound.

Which is the first uncomfortable truth.

Compounding only helps if there is an edge to compound.

Run the same test with a real edge

Now give the system a small edge.

Say win rate is 55% and average win equals average loss again, for simple numbers.

The expectancy per trade in percentage terms is 0.1 × risk (0.55 − 0.45 = 0.10).

For Trader A (fixed fractional at 2%):

Expected growth per trade ≈ 0.2% of equity (ignoring costs for the moment).

Over 200 trades, the expected multiplier is roughly (1.002)200 ≈ 1.49.

£1,000 drifts towards £1,490.

For Trader B (fixed lot):

The expected gain per trade in pounds is 0.10 × £20 = £2.

Over 200 trades, that’s roughly £400, taking you towards £1,400.

Exact path depends on sequence, but the shape is the point.

Fixed fractional growth curve bends.

The pound profit per trade grows as equity grows, so later wins are worth more.

Fixed lot growth curve is closer to a straight line because you grind the same pound risk over and over.

In other words: fixed fractional lets the mathematics of compounding actually show up in your statement rather than just on a quote image on social media.

Fixed fractional vs fixed lot: what happens in a drawdown

This is where the risk argument wins, not the profit argument.

Take the same two traders, both starting at £1,000.

Now give them an absolutely brutal, hypothetical run: 15 straight losses.

No one enjoys this part, but it’s the one that matters.

Trader A, fixed fractional at 2%:

After 15 losses, equity is 1,000 × 0.9815 ≈ 739.

That’s a 26.1% drawdown.

Trader B, fixed lot at £20 risk per trade:

After 15 losses, equity is 1,000 − 15 × 20 = 700.

That’s a 30% drawdown.

Close, but fixed fractional is slightly better.

Now imagine the same horror show happens later.

Say equity has reached £2,000 before this 15-loss streak.

Trader A now loses 2% of 2,000, then 2% of 1,960, and so on. Trader B is still losing £20 a time.

Trader A’s percentage drawdown is still roughly 26%.

Trader B’s is roughly 15% (300 off 2,000).

This is the catch: fixed fractional keeps your percentage risk stable, so pound drawdowns grow as you grow.

That’s both the benefit and the psychological trap.

Fixed fractional protects your account early by shrinking bet sizes when it hurts.

Fixed lot protects your emotions later by keeping pound swings familiar even when the account is bigger.

Neither is a free lunch; the question is which risk you want to manage: percentage or emotion.

Risk of ruin doesn’t care about your feelings

Risk of ruin is the probability you eventually blow up to some unacceptable level if you keep trading a given edge with a given sizing.

Lower average risk per trade usually lowers risk of ruin, all else equal.

Fixed fractional has a natural safety valve: when equity drops, pound risk per trade drops too.

This is why a 50% drawdown is twice as hard to recover from in money, but four times as annoying in psychology.

Your lot size under fixed fractional has halved, so the comeback is slower.

The maths is protecting you while your brain is screaming that it’s broken — we talk about that conflict more in /blog/is-my-trading-system-broken-or-just-unlucky.

Fixed lot has no such protection.

If you keep trading 0.10 lot on XAUUSD on a shrinking account, the percentage risk per trade quietly ratchets up.

At some point, what started as “sensible 1% risk” quietly becomes “this single trade is 10% of my remaining equity”.

That’s how systems with a real statistical edge still get wiped out.

Small accounts, minimum lots, and why gold hurts more

On pairs with low pip value and tight stops, you can approximate fixed fractional even with small accounts.

On instruments like XAUUSD (gold), you hit floor constraints quickly.

Say the minimum lot size available is 0.01.

On a £5,000 account, that might be peanuts.

On a £500 account, a 0.01 lot gold trade with a sensible stop can be 5–10% risk without you meaning it to be.

So your “fixed fractional” plan gets broken by the platform minimum and silently turns into “oversized fixed lot”.

It’s one reason why a lot of retail traders think gold is “wild”.

It isn’t magic; the tick value is just punching far above their equity.

Automation doesn’t fix this either — whether you or a robot presses the button, 0.01 is still 0.01.

If you’re running or mirroring systems on gold on a small balance, position sizing is not a detail.

It’s the whole game.

Sometimes the rational answer is: this instrument is not appropriate for this account size if you insist on sensible percentage risk.

What automation changes (and what it doesn’t)

Automation is very good at one thing: doing the boring maths consistently.

Calculate 1.37% of current equity, convert it into risk-per-pip at current price, then into lots, then double-check margin, then send the order.

A machine will do that all day without getting bored and “just rounding it” like a human does after three coffees.

But automation doesn’t decide the model.

If you tell a system to use a fixed lot, it will happily march you into an oversized risk of ruin faster than you can spell “backtest”.

If you tell it to use fixed fractional, it will also happily shrink your stake after a drawdown and make the recovery feel painfully slow.

That frustration is what pushes people to start tinkering.

“I’ll just bump risk up after a good run and down after a bad one.”

Congratulations, you’ve just invented discretionary sizing on top of an automated strategy, with all the usual recency bias problems from /blog/recency-bias-trading-why-last-few-trades-feel-like-truth.

The machine follows the rule.

Your job is to make sure the rule is mathematically sane before you switch it on.

That includes having the adult conversation with yourself about drawdowns, not just the fun bit about compounding.

So which sizing should you actually use?

“Fixed fractional vs fixed lot” is a false binary if you think about it as right vs wrong.

The better question is: what’s the goal, over what horizon, on what size, with what edge?

Then pick the sizing that actually lines up with that.

Some practical anchors:

• Edge present, long horizon, want compounding: fixed fractional (or a close variant) makes sense.
• Edge uncertain, testing phase, don’t want capital swings yet: tiny fixed lots can be fine while you’re just proving execution.
• Very small account, high pip value instrument (e.g. gold at 0.01 minimum): fixed fractional may not be possible without absurdly wide stops; question whether you should trade it at all.
• Large account, psychological ceiling on daily P&L swings: you might use a hybrid, like fixed fractional up to a cap and then flatten into a pseudo-fixed-lot zone.

The consistent mistake is pretending a fixed lot system is a fixed percentage system.

It isn’t.

Unless you keep manually recalculating and changing the lot size as equity moves, which is exactly the kind of manual faff automation was supposed to remove.

Position sizing, portfolios, and simple beats clever

Once you run multiple strategies across FX, indices and gold, the sizing question gets entwined with correlation and portfolio risk.

Risking 1% fixed fractional per trade on three highly correlated strategies is not 3% total risk; it’s often closer to one big 3% punt.

We covered why correlated bets mislead you in /blog/what-is-correlation-and-why-it-wrecks-trading-portfolios.

The temptation is to get clever.

Dynamic sizing, volatility scaling, intraday equity curves, time-of-day filters, a spreadsheet that looks like mission control.

Then six months later you realise that half your “edge” was just overfitting the position size to the backtest.

Simple, robust sizing wins far more often than it loses.

A steady fixed fractional percentage that you’ve actually stress-tested across multiple systems and market regimes beats the bespoke scheme you can’t even explain without opening Excel.

Which is the theme behind /blog/simple-vs-complex-trading-system-why-simple-wins.

Pick a small number.

Make sure it keeps risk of ruin low and drawdowns tolerable in percentage terms.

Then automate it and leave it alone long enough for the edge to express itself.

The one question to ask before you switch anything on

Forget features, indicators, and some bloke on YouTube whose equity curve only goes up.

Stare at your sizing rule and ask a boring question:

“What happens to this rule if my balance halves, or if it doubles?”

If the answer is “risk per trade doubles when balance halves”, you’ve quietly wired in a self-destruct.

If the answer is “risk per trade in pounds doubles when balance doubles and I’m fine with the swings”, then fixed fractional at that percentage might be acceptable.

If the answer is “I’ve no idea, I just typed 0.10 in the ticket”, then you’ve pushed execution ahead of design.

Systems don’t blow up because they were automated.

They blow up because the underlying risk maths was wishful thinking wearing a suit.

Position sizing is where you either correct that or pretend not to see it.

Trading always carries a risk of loss; no position sizing method removes that risk, it only changes its shape.

Related reading

• What Is Correlation (And Why It Wrecks Trading Portfolios)?
• Why Trade Multiple Strategies Instead of Just One?
• How I Built a Fully Automated AI Trading Platform | ARCIS 2.0

Start your free 14-day ArcisTrade demo →

Watch fixed fractional vs fixed lot play out on real automated strategies without touching the order ticket yourself.

Multiple systems across FX, indices and gold, winners and losers visible, no martingale, no grid.

P.S. Run it like a lab: focus on how the equity curve breathes with different sizing, not on chasing overnight profits.