Exact-RR vs Fixed-RR Backtesting: The Hidden Bug That Inflates Win Rates

This is a technical distinction that almost never gets disclosed publicly, and it's responsible for a large share of the "90% win rate" and "profit factor of 8" screenshots that circulate in trading communities.

How fixed-R scoring works

A backtesting harness has to decide how much R (risk-multiple) to credit a trade when it hits a target. The simplest approach: assume every winning trade is worth a flat number — +1R for the first target, +2R for the second, +3R for the third — regardless of where the strategy's actual price target sits relative to the entry and stop.

Why that's a real problem

If a strategy's real target is only 0.4R away from entry (a tight, nearby level), but the backtest credits every hit as a flat +1R, every single win is being overstated by 2.5x. Run that across a few thousand simulated trades and you get a profit factor that looks spectacular — and has nothing to do with what the strategy would actually produce live.

The reverse is just as real: a strategy whose actual target sits 1.5R away gets under-credited at a flat +1R, making a genuinely strong strategy look mediocre.

What exact-RR scoring does instead

Exact-RR (or "exact price-implied R") scoring credits each trade based on its real geometry — the actual distance from entry to stop, and the actual distance from entry to whatever price level the trade closed at. No flat assumptions. A win is worth exactly what it was worth.

How to tell if a backtest used the honest method

In practice, you usually can't tell from a summary screenshot alone — this is exactly the kind of detail that needs to be disclosed explicitly, or verified by looking at the raw trade-level data. A very high win rate paired with a very high profit factor, especially from a strategy with fixed round-number targets, is worth asking about directly.

Why this matters more than almost any other backtesting detail

Most retail trading education focuses on entry signals — what pattern to look for, what indicator to use. Very little focuses on how the backtest itself was scored, even though a bad scoring methodology can turn a losing strategy into an apparent winner without a single dishonest intention from whoever built it. It's a methodology bug, not necessarily fraud — but the effect on the numbers is the same either way.