Does AI Stock Picking Work? The ClaritX Forward-Tested Record
ClaritX pre-registers every AI stock-selection strategy with frozen rules and judges it only on forward data — wins and losses stay public. Strategies are simulated with next-close execution and 10bps/side trading costs against a total-return S&P 500 benchmark. Nothing is backfilled: each arm has a registered inception date and its curve starts there.
As of , ClaritX tracks 16 registered strategy arms. The strongest live arm is AI band + valuation room at +4.2pp vs the S&P 500 over its window (total return +9.1% vs SPY +4.9%). The weakest arm, Top 10 by AI score (control), is -21.3pp — published, not hidden: negative results stay on this page permanently.
| Strategy arm | Live since | Days | Return | S&P 500 | Alpha | IR | Status |
|---|---|---|---|---|---|---|---|
| AI band + valuation room | 106 | +9.1% | +4.9% | +4.2pp | 0.90 | Passed Gate 1 on 2026-08-21 | |
| Live Ranking (dynamic) | 92 | +7.7% | +5.0% | +2.8pp | 0.49 | Testing | |
| Lab: Live Ranking, momentum + value only | 17 | +2.5% | +0.5% | +1.9pp | — | Testing | |
| Lab: Live Ranking, exact top 30 daily | 17 | +1.7% | +0.5% | +1.1pp | — | Testing | |
| Lab: Live Ranking, value ranked within sector | 17 | +0.2% | +0.5% | -0.3pp | — | Testing | |
| Momentum 12-1/vol Top-30 (SPMO recipe) | 59 | +4.0% | +4.5% | -0.5pp | 0.16 | Testing | |
| AI band + hysteresis | 106 | +3.6% | +4.9% | -1.3pp | -0.37 | Retired 2026-09-23: failed its pre-registered kill rule | |
| AI band × momentum Top-30 | 57 | +2.9% | +4.2% | -1.3pp | -0.20 | Testing | |
| AI score band 70–89 | 106 | +3.4% | +4.9% | -1.5pp | -0.44 | Retired 2026-09-23: failed its pre-registered kill rule | |
| Lab: Live Ranking, max 6 per sector | 17 | -1.0% | +0.5% | -1.6pp | — | Testing | |
| Site Ranking (composite v3, quarterly) | 105 | +2.4% | +5.0% | -2.6pp | -0.48 | Testing | |
| Live Ranking (PMHM, forward test) | 29 | -3.4% | +0.2% | -3.6pp | -2.63 | Testing | |
| AI band percentile | 106 | +0.6% | +4.9% | -4.3pp | -1.01 | Retired 2026-10-12: failed its pre-registered kill rule | |
| Composite v2 (control) | 105 | -0.6% | +5.0% | -5.6pp | -1.00 | Testing | |
| Top 30 by AI score (control) | 105 | -7.4% | +5.0% | -12.4pp | -2.05 | Retired 2026-09-23: failed its pre-registered kill rule | |
| Top 10 by AI score (control) | 105 | -16.3% | +5.0% | -21.3pp | -3.13 | Retired 2026-09-23: failed its pre-registered kill rule |
Do higher AI scores do better?
As of , across 1,087 forward-tested analyses: stocks scored 90–100 ran -5.3pp against the S&P 500 with a 43.9% win rate; stocks scored 0–49 ran -11.0pp with a 31.1% win rate. No band beat the S&P 500 on this snapshot: bands scored 70 and above ran -5.6pp to -2.8pp against it. The score still separated losers from the rest — bands below 70 ran -11.0pp to -10.5pp — 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 | 43.9% | -0.4% | -5.3pp |
| 85–89 | 203 | 57.1% | +2.0% | -3.0pp |
| 80–84 | 192 | 45.8% | -1.0% | -5.6pp |
| 70–79 | 301 | 52.8% | +1.5% | -2.8pp |
| 50–69 | 212 | 36.8% | -6.3% | -10.5pp |
| 0–49 | 122 | 31.1% | -6.9% | -11.0pp |
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.
How to read this
Every arm follows pre-registered rules (entry, exit, rebalancing) frozen before its clock starts. Changing a rule restarts the clock. Kill criteria and graduation gates (8-week, 6-month, 12-month) are defined in advance. Past performance does not guarantee future results — this is educational research, not investment advice.
Methodology & evidence — Research Hub · Monthly AI Rank Report · Free AI Stock Rankings · Disclaimer