Wigan Athletic vs Reading

The model's predicted total of 10.66 lands just below the League One baseline of 10.7, and the gap between the two is small enough to be functionally negligible. Close. That proximity is the primary reason the confidence rating sits at medium: when the prediction and the baseline converge like this, the model is working in a narrow corridor where limited separation exists between signal and noise.

Wigan's recent corner match totals, 9, 11, 10, 7, produce a home average of 9.2, running below the league norm. Reading's recent totals of 7, 12, 10, 14 yield an away average of 10.8, pulling in the opposite direction and partially offsetting Wigan's subdued output. The two averages don't resolve cleanly. Reading's range of 7 to 14 is wide within the same sample, and when a team's per-match corner count swings that dramatically, pinning down the combined total with confidence becomes harder.

The home average of 9.2 is the softest data point in this build, not anchored by any consistent pattern and ranging from a low of 7 to a high of 11. Dispersion matters here. That kind of variance doesn't eliminate the read, but it complicates it, and the medium rating directly reflects that.

Proportional exposure is appropriate. Medium confidence carries its own variance profile and isn't a diluted version of a high confidence tier output. The data supports a measured position here.

Across 101 settled medium-confidence picks in League One, the model's win rate is 48.5%.