04 · Metrics
Risk Consistency
Risk Consistency scores whether every trade risks about the same as your typical trade, weighted by what each risked. How it works, gates, an example, fixes.
Whether every trade risks about the same, against your own typical risk. Weighted by what each trade risked, so one blow-up counts for what it cost. Risking less than usual is never penalised. Risk consistency is one of the two heaviest parts of the Limbix Risk Score, because one oversized trade can decide a month.
How it's calculated#
Risk Consistency compares each trade's risk with your own typical risk, in four steps:
- Take the trades with a risk amount. Limbix takes every trade in the last 30 days that has a risk amount: what it put at risk between entry and stop, in the trade's own currency.
- Find your typical risk. It groups the trades by currency and finds the median risk in each group. That median is your typical trade.
- Score each trade against it. A trade near your typical risk keeps full points. A trade well above it costs points, and the bigger it is, the more it costs. A blow-up far beyond your usual size scores nothing.
- Weight by what each trade risked. A trade that risked more has more say in the final score. A trade that risked three times your usual amount counts for far more than a normal one.
Three rules shape the score:
- Risking less is never penalised. A trade smaller than your typical one scores full points. Sizing down doesn't threaten survival. Whether it's a conviction problem is a separate question, which the Under-Conviction Size detector asks.
- The median, not the mean, sets "typical". A mean would be pulled up by the very outlier it is meant to catch.
- Currencies are never mixed. A median across euros and dollars would be an exchange-rate artefact. Each currency group needs 8 trades of its own to contribute. Each trade is scored as a multiple of its own group's median, so the groups can then be pooled.
The exact weights and tolerances are part of Limbix's method version provisional-2026-09.4; they'll be published here once the method is validated.
What moves your score#
| Action | Effect on Risk Consistency |
|---|---|
| Size every trade from the same risk amount | Keeps the score at or near 100 |
| Take one trade well above your usual size | Costs points, and that trade carries extra weight |
| Take a much bigger trade | Costs more, and can score nothing on its own |
| Risk less than usual | No effect. Smaller trades are never penalised |
| Record a stop on a winning trade | Adds it to the score. Without one it has no risk amount |
| Step up your size for good | Early bigger trades cost points until the new size becomes your median |
| Trade a second currency | Scored in its own group once it has enough trades |
Worked example#
You placed 20 trades in the last 30 days, all in euros. Sixteen risked €200, which makes €200 your typical risk. Two risked double that. Two more risked three and a half times it.
The sixteen normal trades keep full points. The four bigger ones lose points, and the two biggest lose the most. Because the score is weighted by risk, those four trades count for far more than four twentieths of it: they carried a large share of the money you put on the line.
A plain count would say 16 of 20 trades were fine, which sounds strong. Risk Consistency lands well below the strong range, and it becomes the component most likely to hold your Risk Score down. The breakdown card counts the trades that ran over your typical risk.
When it shows "—"#
| Cause | Threshold |
|---|---|
| Too few trades with a recorded risk amount in the last 30 days | Fewer than 8 |
| Enough trades overall, but spread across currencies so no single currency has enough | Each currency group needs 8 |
A trade has a risk amount when it has a stop, or when it lost without one (the exit stands in for the stop). A winning trade with no stop has no risk amount and is left out. See the R-multiple.
When Risk Consistency shows "—", its weight passes to the other components. It is never scored as 0.
What's a good risk consistency score?#
A score of 100 means no trade ran meaningfully above your typical risk. There is no separate band scale for components. The breakdown card colours each one with the discipline cutoffs: 85 and above reads as strong, below 50 as breaking.
Risk Consistency is one of the two components that carry most of the Risk Score's weight.
How to improve it#
- Size from a fixed risk, not a fixed contract count. Decide the amount first, then set quantity from your stop distance. See playbook risk parameters.
- Set a max risk per trade. It gives you a line to size under, and it feeds Risk Rule Adherence. See set your risk limits.
- Read the breakdown. Once you have enough trades, the card counts how many ran over your typical risk. See find your weakest part.
- Record stops on winners too. Without a stop, a winner has no risk amount and can't prove your sizing was consistent. See add a stop and target.
Where you'll see it#
- Discipline & Risk (
/discipline): the Risk consistency row in What the risk score is made of, with a sentence naming how many trades ran over your typical risk. See read the risk ring. - Indirectly everywhere the Risk Score and Discipline Score appear.
Frequently asked questions#
What is consistent risk per trade?#
It means each trade puts roughly the same amount at risk, measured from entry to stop. Limbix compares every trade to your own median risk, not to a percentage of your account. Trades close to your typical risk count as normal variation. The score then falls as trades grow beyond it, and a trade far above your usual size scores nothing.
Why isn't risking less than usual penalised?#
Because this is a survival score. A smaller trade can't damage your account, and sizing down after losses is often exactly right. It may be a loss of conviction, which is a different problem. The Under-Conviction Size detector flags that separately, and Limbix keeps the two in different groups so one screen never praises and blames the same behaviour.
Does it compare my risk to my account size?#
No. It compares each trade with your own typical risk. That means it works even if you never entered an account size, and it doesn't punish you for scaling up correctly as your account grows. If you double your risk per trade, the first bigger trades lose points until the new amount becomes your median.
How are trades in different currencies handled?#
Each currency is its own group with its own median, and amounts are never converted. A €22 index trade and a $757 FX trade become comparable only as multiples of their own group's typical risk. A currency with fewer than 8 trades is left out, so one lonely trade can't score a free 100 against itself.
Why doesn't Limbix publish the exact weights?#
Because the method is still being validated on real traders. The point where oversizing starts to cost you, and how fast it costs, may change as Limbix calibrates them, and publishing numbers that may change would mean public corrections. An exact line also invites people to size to the line rather than trade well. This page explains everything that decides your score.