Anyone who follows football closely has a take on which league is “best” or “most competitive.” That is the wrong question. The right question is which league is hardest to predict. The answer, after running the same model across all five for the 2024-25 season, surprised us.
It is not the one most people would name: the Premier League, La Liga, Serie A, Bundesliga, and Ligue 1 each have a different answer.
What we measured
For each of the five leagues, we tracked four numbers we could verify cleanly from the published 2024-25 season totals.
Home win rate. The share of matches the home side won. Higher means more predictable, more home-dominant.
Draw rate. The share of matches ending level. Higher means more outcomes where the model is least confident.
Away win rate. The share of matches the away side won. Higher means a flatter, harder-to-predict league.
Goals per match. Higher means matches are more open and the scoreline underneath a probability call is wider.
The data
| League | Home wins | Draws | Away wins | Goals/match |
|---|---|---|---|---|
| Ligue 1 | 47% | 20% | 33% | 2.98 |
| La Liga | 44% | 26% | 30% | 2.62 |
| Premier League | 41% | 24% | 35% | 2.93 |
| Serie A | 40% | 28% | 32% | 2.56 |
| Bundesliga | 39% | 25% | 36% | 3.14 |
Numbers from each league’s 2024-25 published season totals.
What this says
Ligue 1 is the most home-dominant league. Across 306 matches, 143 ended in home wins, 62 in draws, 101 in away wins. The home side wins nearly half of all matches. Combined with the lowest draw rate of the five leagues, Ligue 1 matches follow the most predictable script. PSG’s dominance at the top of the table is a feature, not a bug. The whole league leans home.
La Liga is second-most home-dominant. The home side wins 44% of matches, with 26% draws. The away side wins only 30%. Real Madrid and Barcelona have been dominant for a decade, but the home-leaning structure of the league is older than either of them.
The Premier League is the most balanced of the five. The home win rate of 41% is the third-lowest. The away win rate of 35% is the highest. The draw rate of 24% is the second-lowest. Put together, the Premier League is the flattest distribution. That is the structural reason anyone can beat anyone on a Saturday.
Serie A has the highest draw rate of the five leagues at 28%. The Italian game is the most tactical. The 1-0 and 0-0 outcomes pile up. The model has to assign more weight to the draw than for any other league.
The Bundesliga has the lowest home win rate (39%) but the highest goals per match (3.14). German football is structurally the most volatile. The home advantage is the weakest of the five leagues, but the matches themselves are the most chaotic. A 65% home-win forecast in the Bundesliga lands in a much wider range of scorelines than the same forecast in Ligue 1.
The surprise
We expected the Premier League to come out hardest to call. On the away-win metric, it does. But on the metric that matters most for model calibration, the draw rate, Serie A is the harder league. Twenty-eight percent of Serie A matches end level, against twenty-four in the Premier League. A Serie A forecast has to price in a draw more often than a Premier League one.
That sounds small. Over a season it is large. A model that prices Serie A draws at Premier League rates is wrong on roughly 14 more matches per season than it should be.
What the draw rate really means
A 28% draw rate is not a 28% chance of any one match being a draw. It is the season aggregate. But it is the right starting point for a calibration. If the model says 30% on a given Serie A match, the benchmark it should be checking against is the league’s 28%, not the Premier League’s 24%.
For a reader following predictions across leagues, this matters. A 30% draw call in Serie A is a baseline forecast. A 30% draw call in Ligue 1 is overconfident. Same percentage, different league, different meaning.
What makes the Bundesliga feel most chaotic
The Bundesliga has the highest goals per match at 3.14. That is not just slightly higher than the rest. It is meaningfully higher than La Liga (2.62) and Serie A (2.56). The Bundesliga is the only league above 3.0.
Some of that is structural. The bottom sides are aggressive. The top sides press high. The transitions are vertical. Matches open up fast and stay open. The Bundesliga produces 60% of matches with over 2.5 goals, the highest of the five.
For the model, that means a 70% home-win forecast in the Bundesliga is genuinely 70%. But the scoreline underneath that 70% is much wider than the same forecast in Serie A. The Bundesliga’s 1.50 away goals per match is the highest of any league, and the 1.64 home goals is the highest. Goals are flowing in both directions.
For a reader, the Bundesliga is the league where the action is most consistently end-to-end. The model is sharp on outcomes. The scoreline is harder to call.
What makes Ligue 1 feel most predictable
PSG’s structural dominance means most Ligue 1 matches follow a script. The top of the table is settled early. The middle is compressed but predictable. The bottom mostly loses to the top.
The 47% home win rate is the headline. Combined with the 20% draw rate, Ligue 1 has the fewest “open” outcomes of any of the five leagues. The model handles Ligue 1 well because the underlying dynamics are simpler. There are fewer upset paths.
That is not a criticism of Ligue 1. It is a feature of the league. For a reader, the predictions are sharper because the league itself is more structured.
What makes the Premier League feel hardest to call
It is not actually the home win rate alone. It is the combination of three things.
The away win rate of 35% is the highest of the five leagues. English sides win on the road more often than any other league. That is unusual. It breaks the standard model of home advantage.
The fixture pile-up. English sides play more competitions than their continental counterparts. FA Cup weekends, EFL Cup midweeks, and rescheduled Premier League matches all stack up. Squad rotation is higher. A Premier League side playing three matches in seven days is harder to forecast than a La Liga side playing one.
The middle of the table is genuinely deep. Eight or nine sides are usually separated by goal difference alone, not points. Any of them can beat any of them on form. The model has to treat the middle of the Premier League as a single, dense cluster, and that cluster produces the upset rate.
The press. Premier League sides are pressed higher up the pitch and pressed more intensely than in any other league. Matches that look settled on paper get broken down by second-half pressing runs the model cannot fully capture in pre-match data.
What this means for following predictions across leagues
If you follow Goalsforge across all five leagues, expect to see the same probability band produce different reliability across competitions. A 60-70% home-win bucket in Ligue 1 will settle over a narrower range than the same bucket in Serie A. A 30% draw call in Serie A is a baseline. A 30% draw call in Ligue 1 is overconfident.
That is not the model being wrong. It is the model accurately capturing that Ligue 1 is structurally home-dominant and Serie A is structurally draw-prone.
What makes each league worth following
The Premier League’s middle of the table is the most entertaining eight or nine sides in European football. Any weekend can produce a result nobody saw coming.
The Bundesliga produces the most goals. The action is end-to-end and the scorelines move fast.
Serie A is the tactical puzzle. The model has to be sharp on the marginal call, because the outcomes are close. Twenty-eight percent of matches are draws, the highest of the five.
La Liga rewards form reading. Real Madrid and Barcelona have been dominant, but the league below them is competitive and predictable in a structural way.
Ligue 1 is the cleanest read. PSG wins. The model handles the rest with confidence.
A single-league prediction model would never surface these differences. A five-league model has to handle all of them.