Loading…

Trading model research: what passed validation and what failed

Of the 30 models in the Model Hub as of 18 Nov 2026, 14 passed validation in some form (1 rated fit for purpose and 13 fit only with conditions or for limited use), 10 failed (5 rated not reliable, 5 not fit) and 6 are not yet validated. Failures stay on the record with their evidence.

Data as of .

Every model and its verdict

Sorted by result, then model number. The verdict column is the hub's wording, unedited.
ModelWhat it tries to answerResultVerdict as recordedStageReviewed
Option pricing models — Option Pricer M-009How accurately does the Option Pricer turn option prices into implied volatility, Greeks and dividends?Passed: fit for purposeFIT FOR PURPOSEProduction30 Aug 2026
Moonshot Odds Model M-001Can a model find stocks likely to gain 150% or more in 12 months?Passed with conditions or for limited useFIT FOR LIMITED USE — RELIABLE WITH CONDITIONSProduction1 Sep 2026
Capital reverse-DCF & ROIC engine M-002What growth and return on capital does a stock's price already require?Passed with conditions or for limited useFIT FOR LIMITED USE — RELIABLE WITH CONDITIONSProduction1 Sep 2026
Valuation Lens Suite M-003What does a stock's price imply under eight valuation lenses, and do those reads predict returns?Passed with conditions or for limited useFIT FOR LIMITED USE — CONTEXT ONLYProduction2 Sep 2026
Options strategy tags (US + NSE) M-004Is a stock's option implied volatility rich, cheap or event-distorted right now?Passed with conditions or for limited useFIT FOR LIMITED USE — US vol branches only; NOT VALIDATED on NSEProduction3 Sep 2026
Option Lens Suite V/E/S/T/F/X M-006Are a stock's options pricing more volatility than the stock is forecast to deliver?Passed with conditions or for limited useFIT FOR LIMITED USE — V and T as rankers; NOT VALIDATED for E/S/F and for threshold levelsProduction3 Sep 2026
Wing pricing — SHAPE × VOL M-007Do stocks make 2-standard-deviation moves more often than far out-of-the-money options price in?Passed with conditions or for limited useFIT FOR LIMITED USEProduction13 Aug 2026
Daily Picks — Two-Track Screen M-011Can a daily screen pick stocks that touch +10% within 10 sessions more often than chance?Passed with conditions or for limited useFIT FOR LIMITED USEProduction3 Sep 2026
High/Low Retest Strength M-012When a stock retests a prior high or low, how often does the level break within 10 sessions?Passed with conditions or for limited useFIT FOR LIMITED USEProduction3 Sep 2026
Pre-open Gap Board M-013Do stocks that gap down 3–8% in an uptrend recover, and do big gap-ups fade?Passed with conditions or for limited useFIT FOR LIMITED USEProduction3 Sep 2026
Realized-Volatility Forecaster M-019Which method best forecasts a stock's realised volatility over 5, 21 and 63 trading days?Passed with conditions or for limited useFIT WITH CONDITIONSProduction4 Sep 2026
Swing Excess-Return Ranker M-022Can a model rank which liquid US stocks beat the rest over the next 21 sessions?Passed with conditions or for limited useFIT WITH CONDITIONS (US ranked candidate list; unchanged from v1)Research17 Sep 2026
Regime Model — P(risk-off, 21 sessions) M-025What are the odds the US market turns risk-off within 21 sessions?Passed with conditions or for limited useFIT WITH CONDITIONSContext only16 Oct 2026
Earnings-Move and IV-Crush Model (US) M-026How big will a US stock's earnings-day move be, compared with what options imply?Passed with conditions or for limited useFIT WITH CONDITIONSResearch19 Oct 2026
Swing Setup Score M-014Does a 0–100 swing setup score predict which stocks rise over the next 10 sessions?FailedNOT RELIABLERetired3 Sep 2026
Chart Pattern Scanner M-015Do classic chart patterns such as flags, triangles and cups beat matched controls?FailedNOT RELIABLEContext only3 Sep 2026
Candlestick Pattern Scanner M-016Do candlestick patterns such as the hammer predict returns against matched controls?FailedNOT RELIABLEContext only3 Sep 2026
Moving-Average Support/Resistance Episodes M-017Do moving-average stacks and support or resistance tests predict trend continuation?FailedNOT RELIABLERetired3 Sep 2026
Crash-Odds Model M-020Can price and volume data predict a 40% crash within 12 months?FailedNOT RELIABLERetired4 Sep 2026
Swing Target Odds Model M-021Can any stock ranking make a +5% target, 1.5×ATR stop swing plan profitable?FailedNOT FITRetired8 Sep 2026
Next-Day Upside Ranker M-023Can a model pick the US stocks that gain most overnight, from close to next open?FailedNOT FIT (Use A: closing-auction buy / next-open sell; Use B: '+3–10% tomorrow' picker)Research6 Oct 2026
US 12-month Sector-Routed Ranker M-027Can ranking US stocks within their sector beat 12-1 momentum over 12 months?FailedNOT FITResearch19 Oct 2026
Country Relative-Return Ranker (22 markets) M-028Can any rule predict which of 22 stock markets will beat the world next quarter?FailedNOT FITResearch19 Oct 2026
Intraday Trade Ranker M-030Which liquid US stocks to day-trade after the first hour: ranked at 10:30 ET, long/short, exit at the close. Shadow only.FailedNOT FIT (D1h/D1/D0 direction rankers)Research4 Nov 2026
Options Picks structure selector M-005Which option structure, expiry and strikes express a volatility view, and what does it cost?Not yet validatedNOT VALIDATED (structure, direction, strikes) — FIT FOR LIMITED USE (vol-regime legs only)Production3 Sep 2026
Strategy Board — 24 structures × 4 expiry rungs M-008Which multi-leg option structures are priced away from what the stock's own moves deliver?Not yet validatedNOT VALIDATEDResearch30 Aug 2026
Dealer gamma / OI positioning (GEX) M-010Does dealer gamma positioning show whether market makers will dampen or amplify a move?Not yet validatedNOT VALIDATEDProduction15 Aug 2026
Volume Climax Base Rates M-018What happens after a stock prints record trading volume?Not yet validatedNOT VALIDATEDContext only3 Sep 2026
Bottleneck Beneficiaries (AI build-out) M-024Which US companies are positioned to be paid by the AI build-out's bottlenecks?Not yet validatedNOT VALIDATED — pendingShadow8 Oct 2026
FX Trend chip (SMA 50/200 state) M-029Is each FX pair in an up, down or mixed trend — and does an 'up' chip carry the disclosed fade reading (strengthened currencies give back?Not yet validatedNOT VALIDATED — fade claim unreproduced; pre-registered DXY re-test NULLContext onlyn/a

How the verdicts read

The hub also reviews 10 tools (data pipelines, calculators and reports) that make no forecast and are not rated; they are listed in the live Model Hub.

Questions

How many trading models passed validation?
14 of 30: 1 rated fit for purpose and 13 fit with conditions or for limited use. 10 failed (5 not reliable, 5 not fit) and 6 are not yet validated, per the Model Hub as of 18 Nov 2026.
Which models failed validation?
Swing Setup Score (NOT RELIABLE); Chart Pattern Scanner (NOT RELIABLE); Candlestick Pattern Scanner (NOT RELIABLE); Moving-Average Support/Resistance Episodes (NOT RELIABLE); Crash-Odds Model (NOT RELIABLE); Swing Target Odds Model (NOT FIT); Next-Day Upside Ranker (NOT FIT); US 12-month Sector-Routed Ranker (NOT FIT); Country Relative-Return Ranker (22 markets) (NOT FIT); Intraday Trade Ranker (NOT FIT).
What rating scale does the validation review use?
The hub's policy names the ratings FIT FOR PURPOSE · FIT FOR LIMITED USE / RELIABLE WITH CONDITIONS · NOT RELIABLE. The recorded verdicts also use FIT WITH CONDITIONS, NOT FIT, NOT VALIDATED.