Bournemouth vs Manchester City

The cautious read starts with the projection: 9.32 total corners, below the Premier League baseline of 10.5. No clear pattern at the Vitality. Bournemouth's last five corner totals, 13, 8, 13, 14, 7, don't point anywhere definitive, and the season average of 10.0 per match smooths over a range that runs from 7 to 14.

City's numbers are cleaner. Their last five totals, 13, 14, 14, 16, 16, trend higher and the season average of 10.1 sits broadly in line with the league baseline. City contribute corner volume consistently. The model runs those inputs against Bournemouth's context and still lands at 9.32, which is where to look for the drag in the projection. Whether Bournemouth's lower-corner outputs in that last-five stretch reflect something structural in how they set up at the Vitality, or whether the variation is noise without a signal, the data doesn't resolve cleanly. That specific uncertainty is what pushes the grade to medium rather than high. The model's raw projection output grades low; the pick grades medium. Both grades are doing separate analytical jobs.

Medium confidence means medium exposure. It's not a watered-down high-confidence call; the grade reflects genuine uncertainty in the data, and position sizing should account for that. Stake to the tier.

In the Premier League, Spectral's medium-confidence corners picks have settled 134 times at a 45.5% win rate.