A side by side runs two of your own versions and puts their figures in two columns. Both runs use the same dates, the same prices and the same fill rules. The screen says so, because that is what makes the comparison fair.
The rows
- Trades. How many trades the version took in those dates.
- Wins. The count beside the percentage, so a three-trade week does not look like a big sample.
- Made or lost, if every trade risked $100. See below.
- Biggest drop from a high. Drawdown, worked out from the restated trades.
- Average R. Profit for each $1 risked.
- Green days and red days. Days that ended up and days that ended down.
- Trades it skipped. Open it for the reasons in plain English, with a count for each. Zero trades is an answer, not a blank.
- Fees over the window. In dollars, and dollars per contract round turn (one buy and one sell).
- Stop-fill cost over the window. In dollars, and dollars per contract.
- Made or lost after fees.
Why the money is restated
Raw profit and loss depends on how much each trade risked. Two versions sized differently can show very different numbers while making the same decisions. So the top money rows are restated as if every trade risked $100, worked out trade by trade. They say before fees, because a commission does not shrink when you trade smaller.
A trade with no usable risk figure is left out and counted, never quietly treated as zero.
The three cost rows
These three rows are different, and their labels say so. They use the risk each run really used, because a fee is a fixed charge per contract. Restating it would show a fee nobody paid.
A stop becomes a market order when it is hit, so it fills wherever the market is. The stop-fill row charges each stop exit an average amount for that instrument. It has its own row and is never part of the fee, because a fee is billed and this is an allowance. Where that amount is an estimate, the result says so. An instrument with no figure is charged nothing on this row, and the result says so, so it never reads as a free stop.
The chart
One chart, two lines: the running total for each version at $100 of risk per trade, on one scale, over the same dates. Nothing is shaded to mean good. Tap either line to read that point.
Two engine versions are never one table
Every result shows the engine version (the version of Wildflower) that measured it. If your two runs were measured on different engine versions, the Lab says so and does not put them in one table. Numbers from two engine versions do not compare.
What it needs
Both runs are backtests on our servers, so a side by side needs the Backtest or Bot plan.