Every figure in the Lab carries one label that says where it came from. The labels are plain grey text, all the same weight. None of the three is more or less trustworthy than the others. They answer different questions.
The three labels
Backtest. Measured by playing past prices back, under the fill rules printed under the result.
Rehearsal. Measured on live prices as they arrived, with no order ever placed.
Live. Real orders and real fills, taken from your broker's own reports.
They are never mixed
A figure with one label is never added to a figure with another. They never share a table column or a chart line. A total made from a backtest trade and a live trade would describe nothing that really happened.
The fill rules are part of the result
Every backtest prints the assumptions it used: spread, slippage on entry, how far a stop fills past its price, and delay, all in ticks. It also states that a waiting order only fills when price trades through its level. These sit under the figures on every result, never hidden in a pop-up. A profit figure means little if you cannot see the rules behind it.
Where a value was Wildflower's own assumption rather than one you set, it says so beside the number.
Money is shown at one risk size
Money figures in the Lab are restated as if each trade risked $100, before fees. That is how two runs with different trade sizes can be compared. Fees are shown separately, because a fee per contract does not shrink when you halve your size.
Two versions are never one table
Every result says which version of Wildflower's testing produced it. Results from different versions are shown under separate headings, never in one table. A log that spans a version change splits at that point.
The line at the bottom
Under every side-by-side and every weekly log it says: these are results from past prices, and past results are not a promise of future results. The rehearsal diary has a different line, because a rehearsal happens in the present. It shows what the idea did on live prices, not what your account would have been paid.
What can go wrong
Sometimes a saved trade list is shorter than the run's own trade count. Then the running-total chart and the day counts are left out, rather than drawn from part of the record. A partial chart would cover a shorter time than its label says.