Ideas worth keeping.
Explore how a probability-backed analysis becomes a useful piece of football intelligence, from the first signal to the final audit.
Most late goals do not change the outcome
Across the Big Five in 2024-25, only 19% of goals from the 80th minute onwards were winners. The 44% lead-extenders are the largest bucket.
Set-piece goals across the Big Five, not just the PL
Set-piece goals are roughly 20% of all goals in every Big Five league, 2019-20 to 2024-25. The PL's 29.3% 2025-26 spike is the only published deviation.
Draw share is splitting apart across the Big Five. Six seasons show where.
Draw share across the Big Five from 2019-20 to 2024-25 is not shrinking on average. It is splitting. Serie A and Bundesliga rose; La Liga and Ligue 1 fell.
Away sides keep more clean sheets than home sides across the Big Five
Across the Big Five from 2019-20 to 2024-25, clean sheets favour the home side almost everywhere. Here is the league-by-league split.
From the spot: Bundesliga 2024-25 hit a 69.9% conversion rate
Across the Big Five from 2019-20 to 2024-25, penalties convert at 79.5%. Bundesliga 2024-25's 69.9% is the worst single-season figure on record.
Half-time is not a halving line: 55.1% of Big Five goals come after the break
Across 30,120 goals from 2019-20 to 2024-25, 55.1% come after half-time. Here is how the second-half share moves league by league and season by season.
A 1-0 half-time lead is won 69% of the time
Across the Big Five from 2019-20 to 2024-25, a 1-0 half-time lead converts about 69% of the time. Here is the conversion curve, by league and by scoreline.
Set-piece goals now drive Premier League football
Premier League set-piece goals made up 29.3% of all PL goals in 2025-26, the third-highest share in competition history. Here is the long-run baseline.
The red card is not a death sentence
Across the Big Five from 2019-20 to 2024-25, the team reduced to 10 men still wins roughly 28% of matches. Here is what changes league by league.
Against the Form Guide: How AI Forecasts Challenge Traditional Football Wisdom
Six traditional football beliefs a calibrated analysis quietly down-weights, and why the pre-match form guide still treats them as fixed.
Decoding audited status: how a fixture becomes a row in the record
Every Goalsforge fixture passes through four states: scheduled, published, in-play, audited. Here is what each state means, and what audited actually covers.
Fixture Spotlight: Key Matches to Watch in the Upcoming European Cups
Six categories of fixtures across the Champions League, Europa League, and Conference League that test the model hardest this autumn.
Home-field advantage is splitting apart. Six European seasons tell us how.
Home accuracy in Europe's top five collapsed in 2020-21, recovered to a 2022-23 peak, then diverged in 2024-25. Here is the trend.
The new-manager bounce is mostly noise
Across 147 mid-season manager changes in Europe's top five leagues, the average bounce is +0.12 points per game, only +0.07 above teams that did not change.
What 'expected goals' actually means, and what it doesn't
xG is the most cited and most misunderstood football statistic. What it is, what it is not, and how Goalsforge uses it as one input among many.
How promoted teams actually fare across the Big Five
Six seasons of data across the Big Five on promoted clubs: survival rates, top-half finishes, and the structural reasons behind both.
The Premier League is a graveyard for promoted teams
Six promoted clubs in a row went straight back down in 2023-24 and 2024-25. Here is the Premier League survival record across six seasons.
When promoted teams finish in the top half
Across the last six seasons, eight promoted clubs finished 10th or higher in their first year back. Heidenheim is the only one to crack the top eight.
The Bundesliga title race is over by April
Ten seasons of Bundesliga data show the title is usually decided in April. Here is why the pattern is structural, not coincidental.
What makes the three European club competitions different
How promotion and relegation between the Champions League, Europa League, and Conference League changes how an analysis model handles each one.
Cup football is a different game
Why the FA Cup, Copa del Rey, DFB-Pokal, and Coupe de France break every analysis model trained on league football, and how to handle them.
Which European league is hardest to call
Premier League, La Liga, Serie A, Bundesliga, Ligue 1: a 2024-25 comparison of home wins, draws, away wins, and goals to rank who is hardest to call.
How to read a probability-backed football analysis
A practical guide to looking past a single result and reading what a probability-backed analysis is really saying about the match.
Why coverage should be published before kickoff
Timing is part of the evidence. A read is more useful when its inputs are fixed before kickoff, so the result can be checked against the record.
What a useful football model can, and cannot, tell you
A useful football model should make uncertainty easier to see, not hide it. Here is what a model can, and cannot, tell you about a fixture.