Derby County vs Stoke City

The gap is narrow. The model produces a predicted total of 10.37 corners against a Championship baseline of 10.2, and when output and baseline sit this close, the model is not identifying a clean setup. That's not a model failure; it's an accurate read of a fixture where the inputs are pulling in competing directions, and medium confidence is the appropriate categorisation. The model is tracking near-baseline territory, not a clear departure from it.

Derby's home corner totals across their last five Championship matches read 9, 14, 14, 11, and 14, averaging 12.4, which sits well clear of the 10.2 league baseline. Stoke complicates matters. Their last five away totals run 14, 13, 12, 8, and 6, and it's that volatility, not just the headline away average of 10.6, that compresses the model's confidence. Derby's elevated home corner average would push the predicted total higher in isolation, but Stoke's away pattern resists a clean directional read, and the output at 10.37 reflects that tension directly.

Medium confidence warrants measured positioning. The model output doesn't carry the separation from baseline that characterises a high-confidence tier read, and the Stoke away variance means the uncertainty is structural, not incidental. This isn't a fixture to treat as settled. Approach the tier proportionately.

Spectral's medium-confidence record in the Championship: 140 settled picks, 47.1% win rate.