You can size every single trade flawlessly and still blow up. Risk 1% on trade one, 1% on trade two, up through trade six — and now 6% of the account is exposed at once. Each decision was disciplined; the sum is reckless.
Portfolio heat is the number that catches this. It’s the one your per-trade rule can’t see.
Defining heat as the sum of open-trade risk
Portfolio heat is the total risk you have live in the market right now: the sum of the potential loss on every open position if each one hit its stop. Hold three trades risking 1%, 0.5% and 1% and your heat is 2.5%. Easy to define, easy to ignore.
Measured in R-multiples, heat is just the count of open positions weighted by each one’s risk. Thinking in R keeps it clean and size-independent — a position risking 1R is risking 1R whether it’s two contracts or twenty. Your heat is the total R currently exposed across the whole book.
Why per-trade sizing alone misses the aggregate
Per-trade sizing answers one question — “how much do I risk on this trade?” — and answers it well. What it structurally cannot see is how many other trades are already open when you place this one. It has no memory and no total.
That blind spot is where accounts break. A trader with a solid 1% rule takes a valid signal, then another, then another as setups line up in a trending session. Each is individually correct. But by the fifth open position the book carries a loss potential that would look obviously insane as a single trade. The danger isn’t any one position — it’s the aggregate no per-trade rule is watching. Heat is the missing scoreboard.
Setting a max heat ceiling
The fix is a hard ceiling on total open risk, decided in advance. A commonly cited range is a maximum heat of roughly 3–6% of the account, though the right number for you depends on your win rate, how correlated your trades tend to be, and how much drawdown room you have.
| Max heat setting | Who it suits | Trade-off |
|---|---|---|
| Lower (toward 3%) | Correlated book, thin drawdown budget, prop evaluation | Fewer concurrent trades; slower to compound |
| Middle (~4–5%) | Mixed, mostly uncorrelated setups | Balanced concurrency and safety |
| Higher (toward 6%) | Genuinely uncorrelated, high sample confidence | More exposure; less margin for a cluster of stops |
The exact figure matters less than the fact that it’s fixed and enforced. Whatever ceiling you pick, once total open risk reaches it the account is full — no new position until an existing one closes or trails to break-even and frees up heat.
How correlation makes stated heat understate real risk
Here’s the catch that makes naive heat dangerous: it assumes your open trades are independent. They usually aren’t. If three of six open positions share a driver — all long the same index, all short the dollar — their risks don’t diversify, they stack, and on a bad day those three stop out together.
So your stated heat can read a comfortable 4% while your effective heat on a correlated move is far higher, because one event resolves several positions the same direction at once. Two defenses:
- Count a correlated cluster as one position when totting up heat, not as several small independent ones.
- Keep a lower ceiling if your style concentrates in a few correlated instruments — the more your trades move together, the more real exposure exceeds the number on screen.
Run candidate trades through a position size calculator and a risk-reward calculator so each addition’s risk and reward are exact, then check whether adding it keeps the corrected total under your ceiling.
Refusing new entries once heat is maxed
The rule only works if “full” actually means no. When heat is at the ceiling and a beautiful setup appears, the temptation is to squeeze it in — “just this one.” That’s the exact moment the ceiling exists to override.
Two ways to make room without breaking the rule:
- Wait for an open position to close or trail to break-even, which retires its risk and frees heat for the new trade.
- Rotate — if the new setup is genuinely better than something you already hold, close the weaker trade first, then enter. You swap exposure rather than stacking it.
What you don’t do is exceed the ceiling because this one feels different. Every over-heat blow-up started with one exception.
Tracking heat in real time across accounts
Heat is only useful if it’s live. A figure you compute by hand at the start of the session is stale by the second trade, and it gets far worse when you run multiple prop accounts — the same driver held on three accounts is one big correlated exposure no single platform totals for you.
This is where continuous tracking replaces mental math. Shibiki auto-journals every fill and sums your open risk into a live heat figure across all connected accounts, so you always know whether there’s room before you click. Its live edge health wraps each strategy’s win rate in a Wilson confidence interval, so a thin, uncertain edge argues for a lower ceiling rather than a hopeful one. And because a max-heat rule can be enforced as a hard limit at the broker, the account itself can refuse the entry that would push total open risk over the line — turning your ceiling from a good intention into a wall.
Related: R-multiple system · Position size calculator · Risk-reward calculator