V8 research track
Pair & Transition Intelligence
Historical relationship analysis across UK49 draw sequences — Research only
⚠️ RESEARCH LAYER — None of these signals are active in the production v3 candidate engine. Promotion requires walk-forward validation.
UK49s results are random lottery outcomes. Research models rank historical patterns and do not establish a reliable ability to predict future winning numbers. Experimental signals may disappear out of sample. 48 tests run. 0 validated edges found above baseline.
All lift calculations use: occurrences / pair_appearance_count ÷ baseline (6/49). Sample size shown alongside every lift figure.
Filters
Query the transition tables directly
Pipeline status — every relationship on this page
All +2 / +3 relationships sit at stage 2 until a walk-forward test exists
- 1
Historical
Observed in corpus
- 2
Research
Not validated
- 3
Walk-forward test
Not run
- 4
Prospective test
Not run
- 5
Promotion
Not eligible
N→N relationships — lag +2
Source number followed by target number 2 draws later (Lunchtime)
Lift is descriptive only and is not corrected for multiple comparisons. Positive lift here is not evidence of a predictive edge.
Relationship detail
Select a row to inspect its lag breakdown
Walk-Forward Results — research signals
Source: GET /api/v9/wf-results · Δ values are backend-reported deltas versus their own baseline
✅ Production v3 Frozen 2026-09-06 — BT: Trend only | LT: Pair→N lag=1 | DT: Pair→N lag=1 (provisional) | TT: N→N lag=2
Production v3 Architecture — Frozen 2026-09-06
Production signals versus the research signals shown on this page
| Draw | Signal | Source | Weight | Status |
|---|---|---|---|---|
| BT | Trend + Hot Balls | — | 0.00 | ✅ Production |
| LT | Pair→N lag=1 | BT same day | 0.40 | ✅ Production |
| DT | Pair→N lag=1 | LT same day | 0.35 | ⚠️ Provisional |
| TT | N→N lag=2 | LT same day | 0.35 | ✅ Production |
| Signal | Lag | Status |
|---|---|---|
| N→N | +2 | 🔬 Research |
| N→N | +3 | 🔬 Research |
| Pair→N | +2 | 🔬 Research |
| Pair→N | +3 | 🔬 Research |
These signals require walk-forward backtest validation before promotion to production.