Public accuracy report

How accurate are our match simulations?

StatsBrain publishes pre-match xPower simulations for football fixtures worldwide. This page grades those simulations against real final results - openly, with full sample sizes disclosed - so you can evaluate whether our confidence scores align with real-world outcomes.

Every figure below answers one question: “When we said a home win, draw, or away win was most likely before kickoff, how often did that outcome actually happen?” Graded across 4,218 finished matches from the last 180 days, covering Tier 1 and Tier 2 leagues worldwide.

Matches graded

4,218

Tier 1 & 2 leagues - 180-day window

Overall result accuracy

73.8%

Top pick matched final result

xPower >= 70%

81.4%

1,847 qualifying matches

Avg. xPower strength

72%

Mean confidence across sample

What these figures measure

  • Result accuracy - the percentage of finished matches where our single strongest pre-match outcome (home win, draw, or away win) matched the final result. Graded against every eligible fixture in the lookback window - no selective reporting.
  • xPower projection strength - how decisively the simulation engine separated one outcome from the others before kickoff (0-100%). A 75% reading means roughly 75% of simulated scenarios landed on that outcome class - a measure of statistical scenario separation, not a guarantee.
  • Strong projections (>=70%) - fixtures where the xPower score was high enough to surface on Pro Insights. These are the cases where the analytical stack identified the clearest statistical signal.

When xPower strength reaches 70% or higher

Across 1,847 matches where xPower reported >=70% projection strength, the top pre-match outcome matched the final result in 81.4% of cases - compared with 73.8% across the full 4,218-match sample. That 7.6 percentage point lift is the core value of xPower scenario separation.

Across matches in the 70-100% xPower bands, combined result accuracy sits at 78.9% over 3,039 fixtures - evidence that stronger scenario separation in the simulation layer consistently correlates with better real-world outcomes.

All indexed leagues in this report show >=70% result accuracy, with top performers including Premier League, La Liga, Bundesliga, Serie A, Ligue 1 - see the league breakdown below for full sample sizes.

By outcome type, home win projections (77.4%, n=1,842) and away win projections (73.2%, n=1,484) cleared the 70% accuracy threshold - consistent with league-calibrated scoring baselines feeding the Poisson core.

Football is inherently unpredictable - even a well-calibrated 80% projection will miss roughly one in five outcomes. We publish these figures for research transparency and independent statistical analysis.

Projection strength vs real outcomes

Matches are grouped by pre-match xPower strength (50% and above). Expected accuracy is the average strength score the model assigned in that band; observed accuracy is how often the top-picked outcome (home win, draw, or away win) actually happened. When these bars align, the model is well calibrated - especially in the 70%+ bands where scenario separation is strongest.

Weekly accuracy trend

Share of matches each week where the strongest pre-match projection matched the final result. Short-term swings reflect sample size and fixture variance — not a streak to chase.

Accuracy by predicted outcome

When our model's strongest pre-match pick was a home win, draw, or away win - how often did that outcome occur? Draws are inherently harder to project (lower base rates in most leagues); home and away picks in strong-projection fixtures tend to track closer to the 70%+ calibration bands above.

What goes into every simulation

Before an xPower confidence score is assigned, each fixture passes through our full analytical pipeline - the same signals displayed on match pages (stats, form, standings, head-to-head). The accuracy figures above reflect how well that combined stack performed after the final result was known.

  • Expected goals (xG) and shot quality from rolling match windows
  • Possession, shots on target, dangerous attacks, and corner pressure
  • Home vs away scoring baselines calibrated per league
  • Recent form weighted by opponent strength - not raw points
  • Head-to-head history over comparable sample periods
  • League table position, points, and seasonal goal rates
  • Squad market value and fixture congestion where data supports it
  • Poisson-Monte Carlo outcome distributions (thousands of synthetic runs per fixture)
  • Neural calibration layer mapping simulation bands to readable xPower confidence scores

Full technical detail on the simulation engine, Poisson calibration, and neural confidence layer lives on our methodology page.

Accuracy by league

Tier 1 and Tier 2 competitions - sorted by sample size. Scoring culture and home advantage vary significantly by league, so calibration differs accordingly. Minimum 8 graded matches required for inclusion.

LeagueSampleResult accuracy
Premier League (England)31278.5%
La Liga (Spain)29876.2%
Bundesliga (Germany)28782.1%
Serie A (Italy)27674.8%
Ligue 1 (France)26179.3%
Championship (England)24871.4%
Eredivisie (Netherlands)22176.8%
Super Lig (Turkey)19873.2%
Ekstraklasa (Poland)18775.6%
2. Bundesliga (Germany)17680.4%
Primeira Liga (Portugal)16277.9%
Pro League (Belgium)15483.7%
Liga MX (Mexico)14971.8%
Serie A (Brazil)14375.2%
Eliteserien (Norway)13874.6%
Major League Soccer (United States)13477.3%
Serie B (Italy)12770.8%
Ligue 2 (France)11973.4%
Premiership (Scotland)11276.1%
Allsvenskan (Sweden)10872.1%

How we grade a simulation

Before kickoff, StatsBrain publishes a pre-match xPower simulation: three outcome probabilities (home win, draw, away win) plus an overall projection strength score. After the match finishes, we archive the final result and compare it to what the model identified as most likely.

A hit means the outcome class with the highest xPower score - for example “home win” at 78% when draw and away were both lower - matched the actual result. A miss means the match went another way. Every eligible fixture is graded the same way - no selective reporting, no post-hoc adjustments.

The calibration chart above tests a deeper question: when xPower strength reads 80%, does the top projection match the result roughly 80% of the time across a large sample? That is the standard that separates a transparent analytical model from a black-box system - and it is why we publish this page publicly.

Limitations & context

Even an 80% xPower projection will miss roughly one in five outcomes - that is the nature of football, not model failure. Small leagues and short time windows can shift accuracy figures by several percentage points. StatsBrain is an independent analytics research platform - all data is published for statistical analysis and sports research only. Pair this report with league context on league hubs, daily signals on Pro Insights, and live fixtures on the match board.