Pisa operate as a fragile attacking unit, generating just 1.15 xG per match whilst conceding 1.49—a profile that leaves little margin for error. A four-match losing streak compounds the vulnerability, though the underlying shot data suggests variance rather than structural collapse. With no fixture scheduled in the immediate window, focus remains on stabilising both ends before the next assignment. Bawler's model has identified Pisa banker opportunities at a 75% strike rate, indicating reliable edge detection when the conditions align.
How to read this: the green bar shows the average goals Pisa were expected to score per match (their xG output). The red bar shows what opponents are expected to score against them. The diamond on each bar marks the Italy: Serie A average — if the bar extends past the diamond, Pisa are above average there.
How to read this: the solid line is the goals Pisa actually scored each match. The dashed line is the goals the model expected them to score (xG). When the solid line is above the dashed, they overperformed — they finished better than the chances they created suggested. When it's below, they underperformed. Persistent underperformance often regresses; a one-off gap usually doesn't.
How to read this: each tile is one settled match, most recent first. Green = win, amber = draw, red = loss. Numbers show the actual scoreline from Pisa's perspective. Tap a tile to see Bawler's full prediction for that match.
When Bawler issues a Banker pick on a Pisa fixture, the model lands 4 out of 5 (80%). This is well above the cross-league baseline of ~65%. Every pick is logged before kickoff and settled publicly.
How to read this: each row groups settled Banker picks Bawler issued on Pisa fixtures by market type, so you can see where the model has the strongest read on this team. Higher hit rate = more reliable category for Pisa matches.