Juventus vs Fiorentina

The model outputs 9.31 total corners for this fixture, placing the prediction below Serie A's baseline of 10.2. That divergence from the league average isn't noise; it reflects what the recent match data is actually showing. Variance is the problem. Juventus's last five corner totals run 7, 10, 13, 15, and 8, a range so wide that anchoring on any single figure misrepresents the distribution, and it's the primary reason the confidence tier sits at medium rather than high.

Fiorentina's profile is tighter. Their last five totals run 8, 9, 7, 10, 10, producing a match average of 8.6 that reflects genuine consistency rather than a smoothed outlier. Juventus average 9.8 over the same window, but the 15-corner game is doing considerable work in that number. Strip that match, and the underlying Juventus tendency looks materially closer to what Fiorentina have been producing on the road. Full distributions feed the model without outlier suppression, which is correct methodology, but that means the output carries wider uncertainty bands than a 9.31 headline suggests.

Medium confidence means something specific here. Not a data failure. The distributional spread, primarily driven by Juventus's variance, caps the confidence tier and demands proportionate stake sizing. A high confidence tier signal sits at 56.1% win rate across 98 settled bets; this is a different proposition from that.

Spectral's medium confidence record in Serie A: 49 settled, 44.9% win rate.