Glossary

Stability Score

A 0-100 combined score that measures how closely a live trading strategy follows the edge from backtesting.

Last updated: 2026-06-07

Stability Score is Quantprove's Mode 2 (Validation) output: a 0-100 score that measures how closely your live trading results track your backtest. Quantprove compares your backtest trade log against your live trade log across four weighted buckets, PNL Distribution, Drawdown Profile, Edge Quality, and Annual Return, then scales the total by a confidence multiplier tied to your live sample size. A Stability Score of 60 or above means the two records still behave like the same strategy.

Why does Stability Score matter for traders?

A backtest answers one question: did this work in the past? Stability Score answers the harder one: is it still working now, and is there reason to believe it will keep working?

You cannot answer that from historical data alone. You need live trades, with real execution, real slippage, and real conditions, plus a way to test whether those trades still match the backtest.

Most traders skip this step. They see a profitable backtest, go live, and assume the edge carries over. Sometimes it does. Often it does not, and by the time the account is deep in drawdown the trader has lost months finding out.

Stability Score is how you check whether your edge is real or fitted for past data before it costs you. Upload your backtest and your live trades, and Quantprove returns a 0-100 score that tells you whether your live results still track what the backtest promised, or whether the edge has quietly decayed.

In simpler words: Stability Score tells you whether your live trading is still the same strategy you backtested.

How does Stability Score adjust for sample size?

Stability Score scales with how much live data you have. The raw score out of 100 is multiplied by a confidence multiplier sized to your live trade count.

A strategy with 100 live trades scores higher than one with 20, even when their averages are identical, because the evidence is stronger. Full confidence arrives near 500 live trades.

The distribution comparison uses the two sample Anderson and Darling test, which measures how likely your backtest and your live trades came from the same underlying distribution. It carries the single largest sub-weight, 18 of 100 points.

Mode 2 needs a floor of data to run at all: at least 30 backtest trades and 20 live trades.

What are the Stability Score tiers?

A Stability Score of 60 or above means your live results are tracking the backtest closely enough to trust. The four tiers:

  • 80+Strong
  • 60 to 79Solid
  • 40 to 59Weak
  • Below 40Critical

Where is Stability Score used in Quantprove?

Stability Score is the core output of Mode 2 (Validation), the second stage in Quantprove's workflow.

Edge Score grades a backtest on its own. Stability Score grades whether the live results live up to it. A strong Stability Score is the signal to graduate a strategy into Mode 3 (Monitor), where Health Score tracks every new trade against the backtest baseline to catch edge decay the moment it starts.

For the backtest mistakes that most often break live and tank a Stability Score, see common backtest mistakes.

Frequently asked questions

A Stability Score of 60 or above is Solid, and 80 or above is Strong: your live results are tracking the backtest closely. 40 to 59 is Weak, and below 40 is Critical, meaning live and backtest have meaningfully diverged.
Quantprove needs at least 20 live trades and 30 backtest trades to run Mode 2. Confidence then rises with sample size, reaching full weight near 500 live trades.
Edge Score grades a single backtest’s statistical edge. Stability Score compares a backtest against live trades and grades whether the live results still match. Edge Score asks “is there an edge?”; Stability Score asks “is the edge holding up live?”
A low Stability Score means your live trades have diverged from your backtest, in return shape, drawdowns, edge quality, or trade frequency. It is an early sign the edge is fitted for past data or has decayed, often before the equity curve makes it obvious.
Run Mode 2. If your live trades hold a Stability Score of 60 or above against the backtest across distribution, drawdown, edge quality, and return, the edge is behaving as designed. A large gap between live and backtest is the signature of overfitting.

References

See your strategy’s edge change over time.

Upload a live trade log and watch your Health Score across the record. Free to start.

Start for freeHow it works

No credit card required·Swiss Made