A one-point gap with conflicting splits
Lecce and Genoa arrive at Stadio Via del Mare for Round 10 separated by one point and one place: Lecce are 14th with 9 points from 9 games, Genoa 13th with 10. That makes this a small but real separation fixture in the early Serie A table. The headline, though, is not the table. It is the clash between Nutmegly's AI read, which leans away, and Genoa's away record, which is 0 wins, 0 draws and 3 defeats with 1 goal scored and 5 conceded.
The model gives Genoa a 39.3% chance, the draw 31.4% and Lecce 29.3%. That is a lean, not a majority: the home and draw outcomes together account for 60.7%. The AI's away preference is understandable on form, but the away sample is the strongest counterargument. Genoa's three away games have produced one goal; Lecce's four home games have produced one win, four goals and ten concessions. Both splits are bad in different ways, which is why this match is more interesting than a simple 13th-versus-14th line suggests.
What the recent results actually show
Lecce's five most recent listed results are a 1-5 defeat at Frosinone, a 0-2 home loss to Juventus, a 3-0 win at Udinese, a 1-2 home loss to Bologna and a 0-3 defeat at AC Milan. That is one win in five, with 5 goals scored and 12 conceded. The stored eight-match sample puts Lecce at 0.75 points per game. At home in the league they are 1-0-3 with 4 goals for and 10 against. The data does not tell us why the home defense has leaked; it only establishes that it has.
Genoa's five most recent listed results are a 2-0 home win over Juventus, a 1-0 home win over Venezia, a 0-2 away loss at AS Roma, a 2-1 home win over Fiorentina and a 1-2 away loss at Parma. Three wins in five, 6 goals scored and 5 conceded. The stored eight-match sample gives Genoa 1.25 points per game, and the stored five-match Serie A sample gives them 1.80 points per game against Lecce's 0.60. That is the model's best evidence. The caveat is location: three of those five listed results were at home, and both away games in that run were defeats.
The AI read and the away-record problem
Nutmegly's AI read is away, with Genoa at 39.3%, the draw at 31.4% and Lecce at 29.3%. The model also puts over 2.5 goals at 34.1%, so its goal environment is low-scoring rather than open. The away lean is built on efficiency signals: Genoa's recent points rate, their same-competition points rate, and a rest signal that the data itself treats cautiously. The table signal, by contrast, points to a draw because the teams are one point apart. That mix explains why the away probability is a plurality rather than a majority.
The strongest counterevidence is Genoa's away column: 0 wins, 0 draws, 3 defeats, 1 goal for and 5 against. A team that has not won away is being asked to justify a 39.3% away probability. Lecce's home column is not much better — 1 win, 0 draws, 3 defeats, 4 goals for and 10 against — but it does include a home win, and the table gap is only one point. An external forecast also points in a different direction, which the model's own counterpoints flag as a sample-scope issue. None of that invalidates the lean; it does mean the away probability should be read as an estimate, not a guarantee.
Team news adds another layer of uncertainty. Seven potential absentees are flagged across the fixture, and no starting lineups are confirmed. The data does not identify the players, their roles or the likely replacements, so any claim about how the absences will shape the match would be invented. What can be said is that the model's own counterpoints treat the injury count and the unconfirmed lineups as reasons for wider single-match variance. Missing lineup data does not mean both squads are fully available; it means the picture is incomplete.
The watchpoint: Genoa's away goal column
The match turns on which bad split bends. If Genoa's away scoring stays at one goal in three, the away lean has no obvious route to three points. If Lecce's home concession rate — 10 in four — dominates, Genoa's 39.3% looks reasonable. The model's low over-2.5 probability of 34.1% suggests it expects a tighter game, which could bring the 31.4% draw into focus. These are conditional scenarios, not tactical predictions; the data does not supply lineups, roles or in-match plans.
Watch Genoa's away goal column and Lecce's home concession rate. Genoa have one away goal in three; Lecce have conceded 10 in four at home. If Genoa score first, the model's lean has a path, though their away sample offers little evidence of how they would manage a lead. If Lecce score first, Genoa's away record offers little evidence of a comeback. The first goal is not a prediction; it is the clearest single event that would tell us which of these conflicting samples is driving the match.