Do Higher AI Stock Scores Beat the S&P 500?
No band beat the S&P 500 on this snapshot: bands scored 70 and above ran -4.7pp to -1.8pp against it. The score still separated losers from the rest — bands below 70 ran -11.6pp to -10.0pp — so on this data the signal works as a loser filter, not as a source of alpha. Published as measured.
Performance by AI score band
As of , across 1,084 forward-tested analyses: stocks scored 90–100 ran -4.0pp against the S&P 500 with a 50.9% win rate; stocks scored 0–49 ran -11.6pp with a 29.8% win rate. No band beat the S&P 500 on this snapshot: bands scored 70 and above ran -4.7pp to -1.8pp against it. The score still separated losers from the rest — bands below 70 ran -11.6pp to -10.0pp — so on this data the signal works as a loser filter, not as a source of alpha. Published as measured.
| AI score band | Analyses | Win rate | Mean return | Alpha vs S&P 500 |
|---|---|---|---|---|
| 90–100 | 57 | 50.9% | -0.3% | -4.0pp |
| 85–89 | 203 | 59.1% | +2.0% | -1.8pp |
| 80–84 | 192 | 45.8% | -1.3% | -4.7pp |
| 70–79 | 300 | 52.0% | +1.3% | -1.9pp |
| 50–69 | 211 | 38.4% | -7.0% | -10.0pp |
| 0–49 | 121 | 29.8% | -8.6% | -11.6pp |
Each analysis is scored 0–100 when published, then judged only on what the stock did afterwards, against a total-return S&P 500 over the same window. No backfilling, no survivorship pruning; the losing bands are in the table.
What this does — and does not — mean
Even the top band lost 49% of the time on this snapshot. A score shifts the odds across many names; it is not a forecast for any one of them. The honest use of a signal like this is as a screening and prioritisation layer — which names deserve research time — not as a buy list. One snapshot reflects one market regime; the page is rebuilt from each new snapshot and earlier ones are preserved in the track-record history.
How the scores are built
Every score comes from the ClaritX Research Engine — nine research passes per ticker (news sentiment, technicals, social buzz, financials, analyst ratings, peer comparison, insider activity, dividend health, and a synthesis verdict) condensed into a 0–100 score. Rules are frozen before measurement; results are judged only on forward data.
See also: the live strategy track record · the free Top-30 AI ranking · Research Hub · editorial policy. Educational research only — not investment advice.