Vikings 16, Colts 10: The Field Had the Winner, Not the Margin
Drafted by a language model from MissedTakes' graded data, then reviewed and approved by Ben Garves before publishing. How our posts are made
The Minnesota Vikings beat the Indianapolis Colts 16-10 in Week 7, and we graded 19 calls that this game settled. Minnesota was the overwhelming pre-kickoff consensus, and the consensus was right about who would win. It was far less sure about by how much: a 16-10 final means Minnesota won by 6, and the six pundits who put a score on it had a median predicted margin of 10.5.
How many pundits picked the Vikings?
Seventeen of the 18 pundits with a graded voice on this game picked Minnesota outright, and one took Indianapolis. Bill Huber summed up the consensus simply: "The Vikings are 5-0 and look primed to be 6-0 with a date on Sunday against the lowly Colts." The lone dissenter was Chris Broussard, who argued the desperate Colts would rise to the occasion: "Indianapolis' season is on the line... Minnesota, this is your big trap game." The Colts lost by 6, and Broussard's pick graded to a flat miss, zero accuracy and a skill score of -167, the worst mark on the board for this game.
Because Minnesota was also the game's naive favorite, the 17 winner-only picks all graded as expected (100 accuracy) but added no skill above simply taking the home team. The skill separation on this game came entirely from the six pundits who also named a score.
Which score call came closest?
John Breech and Pete Prisco tied for the sharpest score call, both missing the final margin by 3 points and both posting a skill score of +13.06, the best marks of anyone who predicted a score for this game. Breech wrote: "Vikings 19-16 over Colts." Prisco's pick: "Pick: Vikings 21, Colts 18." Both undershot Minnesota's actual margin of 6, but came closer than anyone else on the slate.
Jake Arthur landed next, predicting "23 to 13" for a margin of 10 (missed by 4, skill +9.08), followed by Richard Sherman at "Minnesota 21-10" for a margin of 11 (missed by 5, skill +5.11). Both beat the game's baseline projection, which barely favored either side on points.
| Pundit | Score call | Predicted margin | Missed by | Skill |
|---|---|---|---|---|
| John Breech | Vikings 19-16 | 3 | 3 | +13.06 |
| Pete Prisco | Vikings 21-18 | 3 | 3 | +13.06 |
| Jake Arthur | Vikings 23-13 | 10 | 4 | +9.08 |
| Richard Sherman | Vikings 21-10 | 11 | 5 | +5.11 |
| Chris Simms | Vikings 24-10 | 14 | 8 | -6.82 |
| Tyler Sullivan | Vikings 33-19 | 14 | 8 | -6.82 |
Who missed, and by how much?
Chris Simms and Tyler Sullivan both expected a rout that didn't happen. Simms said flatly, "I got the Vikings 24 to 10," and Sullivan projected "Vikings 33, Colts 19." Both predicted a margin of 14, eight points wider than the Colts actually lost by, and both graded with a negative skill score (-6.82), worse than simply assuming the game would play out close to form. Combined with Broussard's outright miss on the winner, three of the eight pundits who went beyond a bare winner pick finished this game with a negative skill grade, even though the Vikings won exactly as expected.
Did this fit these pundits' track records?
Broussard's whiff extends a rough season of game-outcome calls rather than breaking from one: he's down to 54.55% accuracy and a -26.23 average skill score across 28 graded game-outcome picks this year, well below the field. This one fits that line, not against it.
Breech and Prisco, by contrast, both beat their own norms. Prisco's score-prediction average sits at 56.6% accuracy and a modest +5.76 skill score across 132 graded calls; this call's 92.86 accuracy and +13.06 skill topped both. Breech's typical score-call outing runs 57.27% accuracy and +6.31 skill over 126 graded calls, also short of what he delivered here. For two pundits whose overall numbers are unremarkable, naming the closest number on a game most people got directionally right is a clean outlier in the good direction.
What does this result touch beyond this one game?
Pundits' game-outcome picks on Minnesota have graded at 90.7% accuracy with a +46.02 average skill score across 43 graded calls this season, a run this result extends rather than complicates. Indianapolis is the opposite case: game-outcome picks involving the Colts have graded at just 31.8% accuracy this season, reflecting a team pundits have mostly stopped believing in and have mostly been right not to. Preseason win-total calls on the Colts clustered in the 7-10 to 9-8 range; a loss like this one makes the lower end of that range look more plausible with the season roughly a third gone. For the full season arc on both clubs, see the Vikings and Colts team pages, and see how this week's sharpest and weakest callers stack up on the overall leaderboard.