Inter
Roma
Napoli
Serie A kicks off on 23 August, and Italian football arrives at it in a strange mood. In March the national team lost a World Cup playoff in Bosnia on penalties after a 1-1 draw — a third consecutive World Cup missed, the first former champion ever to sit out three in a row. Gennaro Gattuso left the Italy job three days later. Gabriele Gravina resigned in April, and Giovanni Malagò was elected FIGC president on 22 June with 68.58% of the vote. The league that opens under that cloud has responded the only way Serie A knows how: by firing, swapping and reshuffling almost every bench in sight.
Massimiliano Allegri was dismissed by Milan the day after they fumbled fourth place on the final day, then resurfaced at Napoli on a deal to 2029 — Aurelio De Laurentiis tweeted "Benvenuto Max!" — replacing Antonio Conte, who left by mutual consent a year before his contract expired. Ruben Amorim, sacked by Manchester United in January after a 14-month, 63-game spell, took the Milan job on 16 June. Maurizio Sarri went to Atalanta, Gattuso to Lazio, Domenico Tedesco to Bologna, and Fiorentina, Sassuolo, Torino and Monza all changed hands too. Only a few benches — Chivu's Inter, Gasperini's Roma, Fàbregas's Como, Spalletti's Juventus — carry over. Into that churn we ran 5,000 season simulations. The full Serie A forecast is live; here's the argument.
The title: Inter, and the question is only how much
Inter won the 2025-26 scudetto — their 21st — with 87 points, sealing it on 3 May with a 2-0 home win over Parma and three matches to spare. Lautaro Martínez took the Capocannoniere with 17 goals in 30 appearances, his second, with Marcus Thuram adding 13 more. The gap to second-place Napoli was 11 points. The model responds accordingly: 69.1% title probability, 96.6% top four, 80 expected points, 1888.6 Elo — nearly a hundred points of rating clear of anyone else.
Here's where we'll take a position against our own number: 69.1% describes the team that won 87 points, not the one that starts on 23 August. Denzel Dumfries is gone to Real Madrid, who triggered a €20m release clause on 5 July — a clause with a quirky escalator that rose every five days toward €30m, which Madrid simply beat to the punch. Sommer, Acerbi, de Vrij and Darmian all left as free agents when their contracts expired on 30 June. Josep Martínez is promoted to No.1 with Ivan Provedel signed for €3m as experienced cover — and Provedel is coming off shoulder surgery that may keep him out until August. Chelsea even hijacked Inter's move for Serie A's Defender of the Season. That's a spine being stripped, and Inter's Elo has already shed 39 points over six months. We'd argue the true title number sits closer to 60 than 70. The counter-argument, and the reason the model isn't panicking: nobody behind them has done anything to earn the difference.
The race that matters: four Champions League tickets, six claimants
This is the tightest market on the board: Roma 62.4%, Napoli 52.6%, Juventus 45.8%, Atalanta 40.7%, Como 38.5%, Milan 32.1% for top four. Thirty points separate first claim from sixth. Someone with a genuine case misses out.
Roma at 62.4% is the model's quiet conviction pick. Gasperini finished third with 73 points in his debut season, Devid Malen scored 14 goals, and year two of a Gasperini project has historically been where the pressing patterns compound. Napoli at 52.6% is the number we'd push back on hardest. Yes, their Elo fell 42 points as they finished 11 behind Inter — but they were defending champions, they won the Supercoppa in December (2-0 over Bologna, a David Neres double), and they've replaced Conte with Allegri, the most reliable floor-raiser in Italian football. Elo prices the late-season fade; it cannot price a coaching upgrade that happened in June. We'd take Napoli above Roma on football grounds even though our own simulation disagrees.
Juventus at 45.8% is a referendum on continuity as strategy. Spalletti finished sixth with 69 points and missed the Champions League entirely — and Juventus responded by extending him to 2028, reportedly worth upward of €6m a season, in April. Kenan Yildiz is staying: Arsenal's June enquiry was firmly rejected, the valuation reported around €100m, and Fabrizio Romano called the exit door shut. But Dusan Vlahović walked out as a free agent after talks collapsed over a reported €8m-a-season demand, and as of mid-July he's still unsigned, with Barcelona reportedly his preference amid Bayern, Chelsea and Newcastle links. A team that couldn't score enough to make the top four just lost its only recognised No.9 for nothing.
Atalanta at 40.7% is the number we trust least. They sacked Palladino on 9 June after roughly six months despite a seventh-place finish, and Sarri arrives on a three-year deal off a mutual-consent Lazio exit that followed clashes with Lotito. That's not continuity — it's a full stylistic rebuild dressed as a coaching hire, and the Elo (down 52 in six months) is closer to the truth than the simulation's residual respect.
And then Como. Fourth place, 71 points, one ahead of Milan, secured with a final-day 4-1 win over Cremonese while Milan lost 2-1 at home to Cagliari. A first-ever Champions League qualification, in only their second season back in Serie A after a 21-year exile. Fàbregas himself cited "15 players who are all under 23"; Tasos Douvikas scored 14 goals. Como are the only riser on our movers board, +36 Elo against a sea of red. The 38.5% is not a sentimentality discount — it's a Champions League-workload tax on the league's youngest squad. It's also the number we'd most enjoy being wrong about.
Milan: the biggest faller, the biggest variance
Milan's -67 Elo is the largest six-month decline in the league, and it's earned: fifth place, no Champions League, a manager sacked. But the number is entirely backward-looking. Amorim arrives on three years to 2029 with an option for 2030, and his first signing smashed the club record — Gonçalo Ramos for €74m plus add-ons, potentially €80m+, obliterating the €49.5m paid for Rafael Leão in 2019. Leão himself is the unresolved question: recent reporting says Amorim has sanctioned a departure, Milan want at least €60m, Tottenham have talked to his agent, and the player will decide after Portugal's World Cup. A Milan with Ramos, a settled system and no midweek Champions League football could blow through 32.1%. A Milan where the Amorim-at-United version shows up could undershoot it. That 32.1% is the single number on this page most likely to look silly by May — in either direction. Run your own version in the what-if lab.
The bottom: Monza drew the short straw
Monza came back up the hard way — a 2-0 home playoff win over Catanzaro, a 2-0 defeat in the return, promotion on superior regular-season finish at 2-2 aggregate — an immediate bounce-back from relegation, now handed to Ivan Jurić. The model is brutal: 62.0% relegation probability, 1544 Elo, the weakest team in the division. Behind them, Frosinone 47.1%, Lecce 46.8%, Cagliari 28.6%. Last season's floor was 18 points (Pisa); Jurić's job is to make Monza's fight look nothing like that.
What we'll be graded on
Four falsifiable calls: Inter retain the title (69.1%, though we think the honest number is lower); Roma make the top four ahead of Napoli (the one we'd overrule ourselves on); Atalanta miss the Champions League despite the 40.7%; Monza go straight back down. Every one gets scored in May, next to everything else on our track record. The model is a statistical engine, not an oracle — it can't see Allegri's whiteboard or Amorim's redemption arc. That's exactly why we write down where we disagree with it before a ball is kicked.