Everybody likes to think they can predict the future. The Mayans thought we’d all be dead by now, Napoleon figured his army could weather the Russian winter just fine, and Microsoft’s CEO once said “There’s no chance that the iPhone is going to get any significant market share.”
When people try forecast the future with little reasonable basis for making such predictions, they’re usually wrong (With the notable exception of our own Freddie Maggard prophesying Charles Walker’s punt return last Saturday). ESPN has tried to do just that with their “matchup indicator” tool, and what they’ve come up with (its legitimacy notwithstanding) is textbook Debbie Downer.
If you add up all of the matchup percentages that ESPN predicts, the Cats are likely to win just 2.55 more games.
Obviously, you can’t win a fraction of a game, so that means that the worldwide leader in sports most likely pegs us to finish at 4-8, with a strong possibility of 3-9–the thought of which makes me slightly nauseous.
Hopping onto the fad of Nate Silver‘s fivethirtyeight.com, a statistically driven site that just so happened to be bought by ESPN, the regular site has tried to combine a number of criteria to effectively use it’s FPI (Football Power Index), based on Expected points added to predict the outcome of a game. In their own words:
“Expected points added per game takes into account yards, turnovers, red zone efficiency and more to determine how many points each unit is contributing to its scoring margin. For example, if a team wins by an average of 10 points per game, maybe plus-7 of that is offense, plus-4 is defense and minus-1 is special teams. Because expected points added is built on play-by-play data, it’s fair to say that FPI looks at every play of every game in the season.”
That’s all well and good, but some of their matchup predictions seem wildly off base and, maybe this is just the UK homer in me, anti-Kentucky. Mrs. Tyler’s earlier post didn’t try to get all fancy and dig into stats, but for my money it’s a better, if slightly homer, predictor. ESPN’s numbers are crap, and here’s why:
This isn’t as bad as the next ones, but since it’s at Commonwealth, UK should be around 50%. South Carolina escaped by the skin of their teeth against Vandy and ECU, and they’ve yet to score more than 20 points against anyone. Though their defense is strong they’ll likely need to exceed that season-scoring high to beat the Cats–not unlikely considering our defensive woes, but far from guaranteed under true freshman QB Brandon McIlwain. Kentucky at -2 is a better guess
Vandy was a media darling heading into the season. If “saying all the right things” won football games, they’d be a CFB Playoff contender, but instead they’re reeling from a sloppy-yet-close loss to South Carolina and an embarrassing 38-7 rout by Georgia Tech. Granted, UK lost by more than that to Florida, but Georgia Tech isn’t half as good as the Gators. Also, the fact that the contest is in Commonwealth should push the odds in Kentucky’s favor.
To me, this is the worst offense. I get it, Missouri almost beat Georgia, but are we even sure Georgia is that great? They almost got beat by an FCS team the week before Missouri almost beat them, so I don’t think it’s fair to immediately catapault the Tigers far ahead of the Cats. Yes, Missouri still has a strong defense and they’re hosting, but with a new coach and a lackluster rushing attack (3.6 yards per carry) giving them an 83% chance to win is kind of ludicrous.
Expected points, Ekshmected Shmoints.
Follow me on twitter @LilbighornKSR