Paks vs Ujpest
NB I., Hungary · 2026-08-21 · 17:30 CET
WWLDL
LLWWDThe score model makes a Paks win the most likely outcome at 59%, leaning to goals (Over 2.5 at 73%). The data's biggest gap to the market: Paks win — 58.6% vs 48.6% implied (+10.0 pts). Generated from the data — the market usually knows why it disagrees.
Scoreline probabilities
Priced from the same score distribution as every market below.
First goal · head-to-head
Full Market Baseline (Dixon-Coles)
One statistical model of the final score — every probability below derives from the same score distribution, so they can never contradict each other. Expected goals: Paks 2.39 – 1.40 Ujpest.
| Market | Data % | Implied | Divergence |
|---|---|---|---|
| Paks win | 58.6% | 48.6% | +10.0% |
| Draw | 20.8% | 24.1% | -3.3% |
| Ujpest win | 20.6% | 27.3% | -6.7% |
| Over 1.5 goals | 90.0% | — | — |
| Over 2.5 goals | 72.9% | 65.8% | +7.1% |
| Over 3.5 goals | 52.3% | 46.0% | +6.4% |
| Both teams to score | 69.2% | 65.9% | +3.3% |
| HT over 0.5 goals | 77.2% | — | — |
| HT over 1.5 goals | 47.1% | 45.5% | +1.6% |
Model dc_v2.0 · rating sample: Paks 43.8 / Ujpest 43.7 weighted matches · positive divergence means the data leans above the market — the market usually knows why.
Track your pick
My record →Flat 1u, average market odds, settled either way. Not betting advice — a record, like ours.
How to read this page: probabilities are what this league's historical first-goal-time and scoring data alone imply — they know nothing about injuries, motivation or weather, and the market usually does. We publish every output, including when the data and the market agree. Statistics, not betting advice — never bet more than you can afford to lose.