The weekday filter lets you ask "what if this strategy only traded on certain days?" — by enabling or disabling days of the week per strategy.
For each strategy you choose which weekdays (Monday–Sunday) are enabled. Disabled days are removed from that strategy's series before aggregation — as if those trades never happened. The portfolio curve, metrics and downstream analysis all reflect the filtered data.
A weekday is an entry rule. "No Mondays" means don't open a position on Monday — the decision a trader actually makes. So a trade is judged by the day it was opened: one opened on Friday and held into Monday stays in with Mondays off, and one opened on Monday is out even if it closes on Tuesday. Its P/L still lands on the closing day, as everywhere else in the app — the filter chooses which trades count, not when their result is realised.
This only makes a difference for positions held across days. On an intraday book (0DTE and the like) open and close are the same day, so nothing changes.
Files with only a daily equity curve
Some exports carry one value per day and no trade list (a daily equity export, for instance). There is no way to know when a position was opened, so for those strategies the filter can only act on the calendar day of the series. Everything with trade-level data uses the entry day.
It changes the whole analysis
The weekday filter isn't just a view — it's applied at the source. Correlations and Monte Carlo both respect the active filters, so the numbers they produce are for the filtered portfolio.
The Weekday page breaks each strategy down by day of the week, showing for every weekday:
positive / (positive + negative) — so flat P/L = 0 days never dilute it);Heads up
Max time under water is counted in calendar days on the single-weekday curve. Each weekday trades at most once a week, so a peak that isn't reclaimed for, say, 20 trades already spans ~20 weeks of calendar time. On lightly-traded days (a handful of Mondays across years) this easily reads hundreds of days. The number is real — "had you traded only this day, you'd have sat below the prior high for N calendar days" — but it runs higher than the full portfolio's TUW and is not comparable 1:1 with it.
This is where you spot that, say, Mondays carry most of the risk while Wednesdays do the work.
Each trade is counted under the day it was opened, the same rule the filter itself uses — so what a row promises is what switching that day off actually does. The totals still add up to the strategy's real P/L: a trade held across days is simply attributed to its entry day rather than its exit day, never recomputed.
Like the Metrics ledger and the Monthly P&L grid, this module follows the workspace's reading calendar. Turn on align starting capital in the workspace and pick a range: the table and the per-weekday curves then cover the framed window only, with a band above them naming the active period. The cumulative curves restart from zero on its first day, so what you read is each weekday's contribution inside that stretch.
Without that checkbox the range selector is a pure zoom and this module stays on the full history — the same rule as everywhere else.
No rebase here
There is nothing to rebase on this page. Weekday statistics are computed on the strategy's one-contract series — P/L in dollars with no capital underneath it — so following the period means clipping it to the window, and that is all. The day-by-day rows keep their real dollar amounts.
Days are enabled or disabled in one place: the Workspace, using the M T W T F toggles on each
strategy row. This keeps a single source of truth — there's no separate switch to fall out of sync with.
The Weekday page is analysis-only: it shows the per-day breakdown for the selected strategy and reflects whichever days are currently active, but you don't turn days on or off there.
Filters are part of the session, so they're captured when you freeze a variant — letting you compare "all days" against "Mondays off" side by side.