The table says Cagliari; the last result says caution
Saturday's Serie A fixture at the Unipol Domus matters because it pits a table position against a warning sign. Cagliari are seventh with 13 points from seven games; Bologna are 15th with five. Both have scored six goals. The gap is built on Cagliari's ability to turn narrow matches into points, not on a high-scoring attack. That is the storyline, and it is also why Nutmegly's AI read deserves scrutiny rather than a simple nod.
The model leans home: 77.9% Cagliari win, 17.0% draw, 5.2% Bologna win. That is a strong lean, not a guarantee. The prediction was generated on 24 September, nearly a month before kickoff, and the data contains no confirmed lineups, no injury reports, no odds and no external forecast. Those gaps do not prove anything about availability; they simply mean the estimate is uncalibrated by market or team news.
What the stored form actually shows
Cagliari's five most recent stored results are a 0-4 loss at Monza, a 1-1 home draw with Juventus, a 1-0 win at Udinese, a 2-1 win at Atalanta and a 1-0 home win over Lecce. The sequence is newest first, so the Monza defeat is the latest evidence, not an old blemish. Before it, Cagliari had taken seven points from three matches. The current table adds context: at home they have one win, one draw and one defeat, with two goals for and two against in three games; away they have three wins, no draws and one defeat, with four goals for and five against in four.
Bologna's five most recent stored results are a 1-3 home loss to Inter, a 2-1 away win at Lecce, a 1-1 home draw with Torino, a 0-1 away loss at Napoli and a 2-2 home draw with Sassuolo. Their away split is one win, no draws and two defeats, with two goals for and three against in three. The Lecce win is the counterweight to the table: Bologna have already taken a road result in this sample. Their broader problem is at the other end: 10 goals conceded in seven league games, compared with Cagliari's seven.
The model's case and its soft edges
The case for the model is straightforward. Cagliari have 13 points to Bologna's five, a better goal difference at minus one against minus four, and a home defense that has conceded two in three. Bologna have lost four of seven and conceded 10. In a low-scoring match, the team with the better points return and the tighter defensive record is a reasonable favorite. The model's over-2.5 probability of 42.2% also fits the profiles: both sides have scored six in seven, and Cagliari's home games have produced four total goals in three.
But the soft edge is Cagliari's attacking volume. Six goals in seven is not the profile of a 77.9% favorite, even against a 15th-placed side. The model's confidence is built on points and table position, while the most recent result is a four-goal defeat. That does not make the prediction wrong; it makes it an estimate that needs the first goal and the team sheet to behave. The supplied counterpoint is explicit: lineups are not fully settled, and late selection or shape changes can alter the result surface.
Bologna's strongest supplied counterevidence is the 2-1 win at Lecce. It shows an away performance with two goals, which is exactly what Cagliari's home record has struggled to produce consistently: two goals in three home games. If Bologna score first, the match becomes a test of whether Cagliari can generate more than their home average. If Cagliari score first, Bologna's away record of two goals in three games becomes the relevant pressure point.
The watchpoint: first goal, not forecast
The watchpoint is not another probability recap. It is the first 20-25 minutes and the first goal. A Cagliari opener would put the emphasis on Bologna's away attack, which has produced two goals in three away games. A Bologna opener would put the emphasis on Cagliari's home attack, which has produced two goals in three home games. Those are conditional scenarios, not tactical predictions. The data does not tell us how either team will set up, only what the scorelines and table have produced.
One more qualification: the stored results may span seasons, and the recent-form sample is not a complete season record. The table is the current-season anchor: seven games played for both, 13 points for Cagliari, five for Bologna. The model's read is a pre-match estimate, generated before the latest round of results. It should be treated as a lean, not a verdict. The most interesting question is whether Cagliari's table edge is real enough to survive the kind of result that just happened to them.