Newcastle United vs West Ham United

The model's predicted total of 10.74 sits close to the Premier League baseline of 10.5. That margin is thin. A prediction this near baseline means the model is finding only modest signal in what either side brings to this fixture, and West Ham's recent data does little to sharpen the picture: their last five match totals of 10, 8, 10, 12, and 11 show a contained range and a season average of 11.2, neither of which pulls the model's output far from where it starts.

Newcastle's data carries more uncertainty. Variance is the headline. Their last five corner totals, 10, 12, 5, 14, and 6, span a range that makes any single-match estimate uncomfortable, and the season average of 11.7 absorbs that oscillation rather than explains it. When a home side's recent output moves that sharply between matches, the model has less stable material to work with. The medium-confidence rating captures this directly.

The predicted total of 10.74 is above baseline, but not by a margin that makes this a settled call. The model's output hasn't moved far enough from 10.5 to justify treating the rating as anything other than what it is: a medium-confidence read with real variance on the Newcastle side. Stake discipline applies. Treat the confidence tier as a ceiling on exposure, not a floor.

In the Premier League at medium confidence, Spectral has settled 101 picks at a 46.5% win rate.