The Bundesliga returns tonight, and the model that projected the Premier League, La Liga and Serie A tables has now been run over Germany: 20,000 simulated seasons, all 306 fixtures played out in every one of them. The champion it picks will surprise nobody. The margin might.
How the Supercomputer thinks
Each club's attack and defence come from last season's Bundesliga table — goals scored and conceded per game — pulled 30% back toward the league average. One 34-game season is a real sample but not a precise one, and ratings left unregressed overfit the champion and the bottom club alike.
Every fixture is then played as two independent Poisson draws, home advantage applied at ×1.15 to the home side and ×0.9 to the away side. Twenty thousand times. What comes out is not a prediction of one season; it is the shape of twenty thousand of them.
Bayern do not have a title race
Bayern München win the league in 93.2% of simulated seasons. Every other club in Germany, added together, shares the remaining 6.8%. Dortmund are the best of them and they take the title in 4.6% — roughly one season in twenty.
The projected gap is 16 points across a 34-game season. That is not a close race described politely; it is a one-horse race the model cannot talk itself out of.
The real contest is fourth
Behind the top two, four clubs are separated by 3 projected points: VfB Stuttgart (57), Bayer Leverkusen (57), RB Leipzig (56) and 1899 Hoffenheim (54). Their Champions League odds run 33% to 54% — close enough that the order between them is noise, and honest enough to say so.
What the model does not know
FC Schalke 04, SC Paderborn 07 and SV Elversberg come up with no Bundesliga record at all, so each inherits the same prior: the average of last season's bottom three. That is why all three land on 35 projected points with almost identical risk.
Read that as an admission, not an analysis. The model is not saying these three clubs are equally good. It is saying it has nothing to separate them, and inventing a difference it cannot support would be worse than showing none. FC Schalke 04 in particular arrive with a history that no promoted-side prior can capture.
The one at real risk who is not promoted
Werder Bremen finish 16th or lower in 37.3% of seasons on 37 projected points — the highest danger reading of any club that was in this league last year. Hamburger SV (19.2%) and 1. FC Köln (16.7%) sit in the next tier of exposure.
One point of care with those numbers: in Germany the drop is not two clubs but two and a half. 17th and 18th go down automatically; 16th plays a two-legged play-off against the third-placed side in the 2. Bundesliga. Every “in danger” figure on this page counts 16th place and below, so it is the risk of finishing in trouble — not the risk of being relegated.
The projected table
| # | Club | Proj. pts | Last season | Move | Key number |
|---|---|---|---|---|---|
| 1 |  Bayern München93.2% title | 80 | 89 | — | 93.2% title |
| 2 |  Borussia Dortmund4.6% title | 64 | 73 | — | 4.6% title |
| 3 |  VfB Stuttgart0.8% title | 57 | 62 | ▲ 1 | 0.8% title |
| 4 |  Bayer Leverkusen0.7% title | 57 | 59 | ▲ 2 | 0.7% title |
| 5 |  RB Leipzig0.4% title | 56 | 65 | ▼ 2 | 0.4% title |
| 6 |  1899 Hoffenheim0.2% title | 54 | 61 | ▼ 1 | 0.2% title |
| 7 |  Eintracht Frankfurt5.9% in danger | 46 | 44 | ▲ 1 | 5.9% in danger |
| 8 |  SC Freiburg8.0% in danger | 45 | 47 | ▼ 1 | 8.0% in danger |
| 9 |  FSV Mainz 0511.5% in danger | 43 | 40 | ▲ 1 | 11.5% in danger |
| 10 |  Borussia Mönchengladbach13.9% in danger | 42 | 38 | ▲ 2 | 13.9% in danger |
| 11 |  1. FC Köln16.7% in danger | 42 | 32 | ▲ 3 | 16.7% in danger |
| 12 |  Union Berlin18.6% in danger | 41 | 39 | ▼ 1 | 18.6% in danger |
| 13 |  Hamburger SV19.2% in danger | 41 | 38 | — | 19.2% in danger |
| 14 |  FC Augsburg20.7% in danger | 41 | 43 | ▼ 5 | 20.7% in danger |
| 15 |  Werder Bremen37.3% in danger | 37 | 32 | — | 37.3% in danger |
| 16 |  SC Paderborn 0748.8% in danger | 35 | — | new | 48.8% in danger |
| 17 |  SV Elversberg49.2% in danger | 35 | — | new | 49.2% in danger |
| 18 |  FC Schalke 0448.8% in danger | 35 | — | new | 48.8% in danger |
Points are on a 34-game season across 18 clubs. They are not comparable with our Premier League, La Liga or Serie A projections, which run to 38 games across 20.
Squad churn
1,040 completed moves in and out across the eighteen squads this window, with 1899 Hoffenheim the busiest at 92. We report what moved rather than pricing it: we have no market-value source, and a points-impact column built on guesses would look like a measurement while being nothing of the kind.
What we would actually back
The title is not a market worth entering at 93.2%. The fourth Champions League place is, because four clubs sit inside 3 points and the model declines to rank them confidently. And the bottom is a genuine three-way question in which the model openly holds no view on the promoted trio.
Every call we make on this league goes on the record, win or lose, the same day it settles.