New York RB vs Toronto

The model's predicted total of 10.91 corners clears the MLS baseline of 9.9, which means the structural signal here is pointing toward an elevated count. That gap is real. It's not, however, enough to lift this fixture out of medium-confidence territory, and the reasons for that restraint are embedded in both teams' recent form.

New York RB's last five games produced corner totals of 8, 11, 7, 11, and 5, averaging 9.7. That spread is wide. Toronto's equivalent figures are 8, 12, 12, 7, and 5, for a 9.9 average across those matches. Both teams carrying lows of 5 and highs of 11 or 12 in their recent sample give the model a dataset that's harder to compress into a stable projection, and that's the primary driver of the medium-confidence rating, not weakness in the underlying signal itself.

The 10.91 projection rests on genuine structural inputs, but a combined dataset this volatile means the model's read carries more noise than a high-confidence read would tolerate. Stake sizing must reflect that. Medium confidence carries its own defined protocol, and departing from it because the projected total looks compelling is how exposure accumulates without justification.

Spectral's medium-confidence picks in MLS have returned a 35.1% win rate from 77 settled picks, and that figure is the honest baseline for this tier in this league. Work from it.