The Gap the Plain Backtest Ignores
The standard In-Out backtest simulates being fully invested during a "trade" and fully in cash between trades. But in reality, cash sitting between trades isn't worthless — it can be parked somewhere low-risk and still earn something, rather than earning nothing at all.
What Arbitrage Adds
In-Out + Arbitrage keeps the identical entry and exit rules as the plain In-Out backtest. The only change: for every gap in time when the simulation is "out" of a position, that idle capital is assumed to grow at a flat, modest annualised rate — modelling money parked in a low-risk instrument rather than left doing nothing.
Why This Gives a Fairer Picture
Two strategies with identical trade signals can look very different once you account for what happens to cash between trades. A strategy that's "out" of the market often, for long stretches, benefits meaningfully from this idle-return assumption — while a strategy that's almost always invested barely notices the difference.
Reading the Result
The result screen breaks down each idle period separately — how many days the capital sat out, what it earned during that stretch, and how it folds into the running total alongside the actual trades. Compare the "In-Out" and "In-Out + Arbitrage" totals side by side to see exactly how much of the difference comes from the trades themselves versus the idle-cash assumption.