Media Days in general bother me. Not only do you get the regular dose of coach-speak and cliches from a press conference, you get to hear it over and over and over again one after the other. In order to get any sort of valuable answer, you have to get the subject in question in an isolated setting where they feel more comfortable. (That’s why whenever a big shot like Gary Parrish or Jeff Goodman or Andy Katz come to press conferences after games, they rarely ask questions, but rather wait for the coach to finish and then walk out with them and interview them on their own.)\
But even I can’t resist the temptation of a preseason poll. It gets fanbases all riled about how their team is getting slept on, signifies live basketball is tantalizingly close and is a general guide post as to how this particular season might play out. I really like getting this sense of hierarchy and only wish that the individual polls were made public, instead of just the combined one.
Marquette is predicted to finish second behind Villanova for this upcoming season, the best finish in this poll since 2014. A pretty fair projection and one that seems to align with the national consensus. But one thing rarely gets analyzed in the discussion. How accurate are the predictions anyway?
Here are the results since the reformation:
And here’s a visual representation of how far off the preseason predictions were to the actual results:
Without doing any fancy statistical work, here are some quick reactions.
Coaches actually do a pretty good job overall. On average, any individual team’s projected finish will be off by only 1.74 spots any given season. When the difference between 3rd place and 7th place was 1 win last year, this lack of variance makes the projections look even better. That’s not to say you have to take these results as gospel, only 16 of the 50 positions have been hit on the head, but in general, they are close enough to the mark to be considered accurate.
On average, Marquette has finished 2.4 spots away from the preseason projections, tied for 3rd worst in the conference. It took some time, but it seems like there was a market correction on Marquette’s program as it finished below expectations for the first 3 seasons after the reformation. However, it has met or exceeded expectation the past 2 years, which would help explain why so many coaches are bullish on Wojo’s squad this season.
Georgetown has finished below expectation 4 of the 5 years, and doing so by at least 5 spots for 3 of those 4 years. Guess the name on the front was carrying a lot of weight. Last season, the first under Patrick Ewing, was the first time the Hoyas beat the projected spot, finishing 8th instead of 9th.
DePaul has been the team projected with the most accuracy. Seeing as they have been predicted to finish 10th 4 out of 5 years, that’s quite a bummer for them.
Villanova has been the team projected with the 2nd most accuracy. Seeing as they were projected to finish 1st 4 out of 5 years, that’s quite an accomplishment for them.
Xavier has probably the strangest profile. The have beaten expectations, where the sum of their projected finishes is dead on with their actual finishes (4th for both). However, they have been at least 2 spots above or below the predicted ranking each season. The coaches just can’t seem to get a grasp on how good the musketeers are.
St. John’s has never finished above their preseason projection, just put that in the back of your pocket before putting any money down on them. (Though I’m very bullish on them myself this season.)
On the other hand, Butler has most consistently beaten their projected rank, and by the most spots as well. Sleep on them at your own peril.
At the end of the day, the actual performance on the court will have the final say. If Marquette finishes 1st or 8th, it will have 0 to do with what the other coaches thought of them. But until we get that sweet taste back in our systems, acknowledging that the preseason poll is a good bellwether of what’s to come will have to do.
Paint Touches 
Leave a Reply