Public record
Track record
Every scenario signal is generic research on a security, not a personal recommendation. Each one is logged with the share price at the moment it was published, then graded at windows fixed in advance — 1 day, 5 trading days and 30 calendar days after the event — against the realised move and the S&P 500 over the same period. Every rate below is shown with its sample size and a 95% confidence band, so you can see how much of it is skill and how much is still noise.
Pre-registered · written while this record was empty
What we would have to demonstrate before calling this predictive
We are not claiming this model is predictive. It ships as a clearly labelled second opinion, free, with its record published in full. If the 95% confidence band on win rate against the sector benchmark sits entirely above 50% over at least 200 sector-gradeable calls, we will have earned the right to describe it as predictive. Until then we won't.
Deciding this now, with no data, is the same discipline as fixing the grading windows before the outcome is known. Deciding it later, with subscribers attached, is when people start relaxing the test.
Every call is graded against both the S&P 500 and, where the company sits in a sector we track, a liquid sector ETF. The pre-registered stopping rule is tested against the sector benchmark, which is the harder of the two.
Coverage of this test
The sector-benchmark condition can currently be evaluated for 14.1% of tickers with a dated upcoming event (427 of 3,037 tickers; 2,610 have no sector benchmark) — the remainder classify as "other" and have no sector benchmark assigned. This is a known gap in our sector classification, not a change to the rule above. Calls on unclassified tickers are still graded and published against the S&P 500, but do not count toward the sector-benchmark condition.
What the threshold counts: the 200 is 200 sector-gradeable calls, not 200 graded calls in total. At today's coverage of 14.1%, reaching it needs roughly 1,418 graded calls overall. That is deliberately the slower reading: a band drawn on the thirty or so sector-gradeable calls this book would otherwise produce is far too wide to support any claim, so the claim waits rather than the sample being stretched to fit it.
Superseded on 2026-08-21 · kept for the audit trail
The stopping rule this replaced
After 200 graded calls, the 95% confidence band on directional hit rate must sit entirely above 50%, and the band on win rate against the sector benchmark must sit entirely above 50%. If either band includes 50%, we stop describing the model as predictive anywhere in the product or its marketing, remove it from every paid tier, and keep publishing it only as a clearly labelled secondary opinion beside this record.
Changed on 21 August 2026, while this record was still empty, from a rule that would have removed the claim if it failed. At the sector coverage we have today that test would have fired on insufficient data rather than demonstrated failure — so we withdrew the claim instead of deferring the test. The original rule and the reasoning are both kept here.
Catalyst is a research tool, not a regulated financial adviser. Nothing here is a recommendation to buy, sell or hold any security. Do your own research and consider your personal circumstances before investing.
Read the full disclaimerHit rate counts hits and misses only; a directional call whose move was smaller than 1% in either direction is reported separately as no clear move. A "mixed" call is the claim that the move stays inside that 1% band, so it is a hit inside the band and a miss outside it — it is never excluded. The 5-day window is five trading sessions, not seven calendar days. Confidence bands are Wilson 95% intervals for rates and normal 95% intervals for averages — they describe sampling uncertainty in this record alone, not the risk of any future trade. Windows are fixed before the outcome is known. Beta account and feature data may be reset or corrected as methods change. This never applies to published live calls or settled record outcomes: those are never re-graded, reset or deleted, and any correction is added as a dated note beside the original rather than replacing it. Past performance says nothing about future results. Scenario signal, not a personal recommendation. Catalyst does not know your circumstances, tax position or objectives. Research and information only, not financial advice.