1581 plays 1544; the model splits it 52.0% / 24.4% / 23.6%. The split is one apiece: Aston Villa leads the attacking rate on 1.09, Hull City the defensive on 0.80.
Taken to a single scoreline the model lands on 1-1. Relative to the 1500 seed that every rating begins at: +81 and +44. 37 points of rating separation. It ended 0-0. Between them, the home side holds the higher rating. Our pre-match read, logged before the game: 54.2% / 23.9% / 21.9%. That one did not come off. The three results cover 28.4 points. Attack/defence pairs: 0.96/0.80 and 1.09/1.16. The model gives 75.6% to someone winning. The leading outcome clears the second by 27.6 points. Ratings update at full time; these are the post-match figures. Expected goals come out at 1.66 to 1.18. 52.0% goes to Hull City, the model's lean. 4 scorelines clear 8% individually. Hull City's expected share on the Elo curve comes to 64.3%. No scoreline stands out; 1-1 only leads on 11.5%. Sample sizes behind the two numbers are 54 and 44 matches.
Computed by elo-poisson-v1 from the whole corpus rather than republished. Analytical only.