Free statistical lab · Local and deterministic

Football prediction sample size calculator

Test how uncertain a reported hit-rate advantage really is, then plan how many future matches you need to detect a meaningful improvement in accuracy or paired Brier score.

Canonical publication record

Abstract

A transparent planning and inference calculator for football prediction hit-rate intervals, independent and paired comparisons, minimum detectable effects and paired Brier-score sample requirements.

Author and publisher
Football Proof AI
Technical report
football-sample-size/1.0.0
Published
Last modified
Release status
Current release
Review status
Editorial technical note; not externally peer reviewed

Interactive · football-sample-size/1.0.0

Separate signal from a flattering sample.

Estimate uncertainty, compare hit rates and plan the evidence needed before evaluation. Runs locally in this browser and makes no network request. Equal inputs produce the same downloadable JSON.

Browser-only · deterministic report
1 · Observed evidence design

Paired and independent comparisons are not interchangeable. If both models predicted the same fixtures, enter the four paired correctness cells; two marginal hit rates do not preserve the within-match dependence.

2 · Plan hit-rate evidence

33.3% is not assumed. A 1X2 uniform guess, historical most-common outcome, bookmaker-implied top pick or prior model can imply very different reference rates. Declare a time-safe reference before evaluating the target model.

3 · Plan paired Brier evidence

Supply the paired-difference standard deviation from a separate pilot or earlier time window using the same Brier convention. This calculator does not invent it from marginal scores. The paired plan uses the alpha and power selected above.

Approximate statistical report
Model A hit rate54.0%Wilson 95%: 49.6%58.3%
Model B hit rate48.0%Wilson 95%: 43.7%–52.4%
Difference · A minus comparison+6.0 pp95%: +2.3 pp to +9.7 pp · paired per-match difference normal interval
Two-sided comparisonp 0.0016z = 3.162 · uncorrected McNemar normal approximation for paired proportions
Matched fixtures
500
Discordant
90
Discordance rate
18.0%

Design warnings

  • This mode uses each match as one pair. The four cells must describe the same matches for both models.

Hit-rate sample plan

Required matched pairs
479
Expected discordance
30.0%
MDE at current n
6.8 pp
  • Two-sided McNemar-style normal planning approximation on matched fixtures.
  • The two marginal hit rates and expected discordant-pair rate define a feasible paired four-cell table.
  • The expected discordant-pair rate is pre-specified and remains stable over the evaluation window.

Paired Brier-difference plan

Required matched pairs
503
MDE at current n
0.0100
Supplied difference SD
0.0800
Convention maximum
0.6667
Feasible SD ceiling at target
0.6666
Max feasible mean at supplied SD
0.6618
  • Two-sided paired-mean normal approximation using one A-minus-B Brier difference per match.
  • The class-averaged 1X2 convention fixes the maximum absolute per-match Brier difference at 0.6666666666666666.
  • At planned mean-difference magnitude 0.01, bounded differences require SD no greater than sqrt(0.6666666666666666^2 - 0.01^2) = 0.666591662447442.
  • At supplied SD 0.08, the largest feasible mean-difference magnitude is sqrt(0.6666666666666666^2 - 0.08^2) = 0.6618492611195124.
  • The supplied paired-difference standard deviation comes from a separate pilot or earlier time-safe sample and uses the same Brier convention.
  • Match-pair differences are treated as independent with stable variance; team, league and time clustering are not modelled.
Deterministic local report

Formula football-sample-size/1.0.0. No generated timestamp or remote value is added, so the same valid inputs produce identical JSON bytes.

First decision

Use the comparison your match history actually supports

Statistical significance is not a property of two percentages alone. The calculation changes when predictions share the same matches because their wins and misses are correlated.

  1. 01

    One model vs reference

    Use one accepted prediction record and a reference rate declared before examining that record.

  2. 02

    Independent samples

    Use when A and B were evaluated on separate fixtures or genuinely independent match samples.

  3. 03

    Paired same-match test

    Use four correctness cells when both models forecast every one of the same settled fixtures.

Two paired models can have identical marginal hit rates but very different uncertainty depending on how often they disagree. Keep the row-level match pairing; totals alone cannot reconstruct it. A feasible discordant rate q must satisfy |pA − pB| ≤ q ≤ min(pA + pB, 2 − pA − pB); the calculator rejects combinations that would imply a negative four-cell count.

Reference choice

A football baseline is not automatically 33.3%

Uniform
A one-third reference describes random equal 1X2 picks, not the real outcome mix.
Empirical
A historical most-common-outcome strategy can be much stronger, but must be estimated only from earlier matches.
Operational
A bookmaker-implied top pick or frozen prior model can be the relevant hurdle if specified before evaluation.

The calculator therefore keeps the reference editable. Document its source, time cutoff, league scope and decision rule next to every published result.

Formula contract

What each result estimates

Confidence intervals, planning equations and most tests below use stated approximations; the small-discordance McNemar p-value is exact conditional. Every result is for transparent reporting, not a substitute for a pre-declared protocol.

Single rate
Wilson score interval at 95%; one-sample normal score test against the chosen reference
Independent difference
Newcombe-Wilson score interval for A minus B; pooled two-sample z statistic under equal null rates
Paired difference
Normal interval from per-match correctness differences only when at least 25 pairs are discordant; otherwise the interval is withheld and the p-value uses the exact conditional McNemar test
Required sample
Two-sided normal planning approximation at the selected alpha, power and meaningful effect
Minimum detectable effect
Smallest effect whose approximate required sample is no larger than the current sample
Paired Brier plan
Paired-mean normal approximation using the supplied SD of per-match A-minus-B Brier differences

Paired probability quality

Brier sample size needs the variation of paired differences

Evaluate both models on each same match, calculate one Brier score for each, subtract B from A and estimate the standard deviation of those differences from a separate pilot or earlier time window. Use the same summed or class-averaged convention in the pilot and the planned study.

Marginal Brier means or separate standard deviations are not enough because they omit the within-match covariance. A pilot SD chosen after inspecting the final evaluation can make the sample plan look more precise than it was. The calculator also enforces the bounded-difference feasibility limit SD ≤ √(M² − δ²), where M is the convention's largest absolute per-match difference and δ is the planned mean-difference magnitude. The inverse bound, |δ| ≤ √(M² − SD²), also prevents the current-sample MDE from being reported when no bounded per-match difference distribution could attain it at the supplied SD.

Limits before claims

Approximate power is not a performance guarantee

Required sample size is a design aid under stated assumptions. It does not guarantee significance, future accuracy or betting value in the realized sample.

  • Small discordant counts need an exact paired test.The normal McNemar approximation can be inaccurate when few matches differ.
  • Match observations may be dependent.Repeated teams, leagues, seasons and time periods can create clustering that these formulas do not model.
  • Tuning and selection consume evidence.Choosing the model, threshold, baseline or metric on the same matches invalidates a clean confirmatory interpretation.
  • Multiple comparisons inflate false positives.Testing many leagues, horizons or model versions needs a declared multiplicity policy.
  • Power assumptions can drift.Reference rates, paired discordance and Brier-difference variance can change after deployment.
  • Publication timing still matters.A statistically strong result is not a forecast record unless its probabilities were publicly fixed before kick-off.

Primary references

Intervals, paired tests and proper scoring rules

This is an editorial technical note, not externally peer reviewed. The cited papers define the statistical building blocks; they do not certify any entered evidence or model claim.

  1. Wilson (1927), score interval for one proportion.Original paper and DOI
  2. Newcombe (1998), interval for the difference between independent proportions.Original paper and DOI
  3. McNemar (1947), paired binary response comparison.Original paper and DOI
  4. Brier (1950), verification of probability forecasts.Original paper and DOI
  5. Gneiting & Raftery (2007), proper scoring rules.Journal paper and DOI
  6. Cohen (1992), statistical power planning.Method paper and DOI

Cite this page as: Football Proof AI (2026), “Football Prediction Sample Size & Significance Calculator,” version football-sample-size/1.0.0. Access the canonical page at https://footballproofai.com/tools/football-prediction-sample-size-calculator.