How the research pages are computed

Each page tests one calendar rule, stated in advance: when to enter, how many sessions to hold, and on which symbol. A trade enters at the open of its entry session and exits at the open of its exit session, in New York exchange sessions. The rule is the same in every year; nothing is tuned to the years it is measured on.

Each trade is compared with the same-holding return: what holding the same symbol, in the same direction, for the same number of sessions made on average, over every such hold before the research cutoff. That is what the trade would have earned without any calendar idea.

The difference is the trade’s excess. A rule can make money in a year and still do worse than holding, because the market rose anyway. The pages count a year as a good year for the rule only when its average excess was above zero, not merely when it made money.

Each page lists every year on record with its average return, its same-holding return and its excess, and names the years the rule did no better than holding. An average can hide a few large years; the year table shows whether the result came from many years or from a handful. The worst year is shown beside the average.

Every return is gross: before any trading cost or tax. A rule held for a few sessions and traded many times a year pays those costs many times, so a small gross excess can be smaller than what it costs to trade.

The figures above the year table use prices through the research cutoff. Trades that closed after it are listed separately as out of sample and never enter an average, so the reader can see how the rule did on prices the research had not seen.

A calendar rule describes what happened in past years. It is not a forecast of what a symbol will do, the figures are not a statistical test, and nothing here is investment advice. Calendar patterns can weaken or stop, and several already have. Read the notice on hypothetical performance before relying on any figure.

Every figure on a research page is computed by the same evaluator as the Scan and Check in the app, from daily opening prices. Each page shows the date of the research cutoff and of its latest prices. The glossary defines every figure.