Position Sizing for a Daily Loss Limit: The Exact Math
A daily loss limit does not care about your strategy, your win rate, or your conviction. It counts dollars, and it counts them mid-trade. Position sizing is the only lever that decides how many mistakes fit between you and that line, and the right size is not a feeling — it is one division, done before the session starts. Here is the formula, then the lot arithmetic for the three markets evaluation traders use most.
The formula
Two inputs: the daily loss limit in dollars, and the number of consecutive losing trades you must be able to survive in a single day. One deduction: a buffer for spread and slippage, because limits are enforced on equity to the dollar.
Risk per trade = (daily limit − buffer) ÷ losses to survive.
On a $100,000 account with a 5% limit, the line sits at $5,000. Reserve 10% of it — $500 — as buffer, leaving $4,500 of usable room. Decide to survive four straight losses:
$4,500 ÷ 4 = $1,125 per trade, or about 1.1% of the account.
| Losses to survive | Risk per trade | % of $100,000 |
|---|---|---|
| 2 | $2,250 | 2.25% |
| 3 | $1,500 | 1.50% |
| 4 | $1,125 | 1.13% |
| 5 | $900 | 0.90% |
| 6 | $750 | 0.75% |
The table is the whole argument against 2% risk. At $2,000 per trade, two ordinary losses put you $4,000 down with an open third position able to end the account mid-trade. Nothing about that requires bad trading — just a normal cold streak arriving on the wrong day.
Why four is the honest minimum
Streaks are arithmetic, not bad luck. With a 50% win rate, any given run of four trades has a 1-in-16 chance of being four straight losses. At a 60% win rate it is roughly 1 in 39; at 40%, about 1 in 8. Take five trades a day for a month and you generate dozens of overlapping four-trade windows — a four-loss run is not a tail event, it is scheduled. Sizing that only survives two or three losses converts a routine streak into a breach.
Better still, pick the survival number from your own records: the longest losing streak in your last 100 trades, plus one. For most intraday strategies that lands on four to six.
From dollars to lots
Risk per trade becomes a position size through one more division:
Size = risk per trade ÷ (stop distance × value of one unit of stop distance per lot).
EURUSD
One standard lot moves about $10 per pip. With a 30-pip stop, one lot risks $300. At $1,125 of allowed risk: $1,125 ÷ $300 = 3.75 lots. Round down, always — 3.7 lots risks $1,110 and keeps the arithmetic honest. The same trade with a 50-pip stop supports only 2.2 lots. The stop distance sets the size; the size never sets the stop.
Gold
A standard gold lot is 100 ounces, so a $1.00 move in the price is $100 per lot. A $9.00 stop risks $900 per lot: $1,125 ÷ $900 = 1.25, so trade 1.2 lots. Gold's habit of covering $9 in a few minutes is exactly why the division matters more here, not less.
Indices
Index contracts have no universal spec — one broker's lot pays $5 per point, another's $20. Suppose $5 per point and a 40-point stop: $200 per lot, so $1,125 ÷ $200 = 5.6 lots. At $20 per point the same stop supports only 1.4 lots. Read the contract specification before the first trade; guessing the per-point value is how a planned $1,125 loss arrives as $4,500.
| Instrument | Value per unit | Stop | Risk per lot | Lots for $1,125 |
|---|---|---|---|---|
| EURUSD | $10 per pip | 30 pips | $300 | 3.7 |
| Gold | $100 per $1.00 | $9.00 | $900 | 1.2 |
| Index CFD (check the spec) | $5 per point | 40 points | $200 | 5.6 |
Open positions count against the same limit
Daily limits are measured on equity, so floating losses spend the allowance in real time. Two positions open at once, each risking $1,125, is $2,250 committed to a single moment — half the day's room gone in one correlated move. The fix is to divide by concurrent exposure: with the four-loss plan above, two simultaneous positions get $562 each, not $1,125 each.
The buffer earns its keep here too. A 30-pip stop on 3.7 lots filled two pips late costs an extra $74, and news-hour spreads can triple that. The $500 reserve is what stops "filled slightly late" from becoming "breached by $19".
The number the formula cannot supply
The formula assumes you actually stop at the chosen count. Four losses at $1,125 leaves you at −$4,500 with $500 of buffer — mathematically alive, practically done for the day. The fifth trade after four straight losses is rarely the strategy; it is the tilt the limit was designed to catch, and it is the most common ending catalogued in mistakes that blow evaluation accounts.
So write both numbers down before the session: the risk per trade, and the loss count that ends the day. The first is arithmetic. The second is a promise.
Rehearse the division under enforcement
The math takes five minutes to learn and one loose afternoon to abandon, which is why it should be rehearsed somewhere the limit is enforced rather than imagined. Run the formula on virtual funds under a live daily limit — FundedLot enforces the same lines an evaluation does — until stopping at four losses feels procedural rather than heroic. The wider defensive system this slots into is covered in risk management for funded traders.