Win Shares Wednesday: Predicting Florida’s 2018 season by percentages

It’s a time-honored tradition here at Alligator Army.

Each year for the last several, we have come together and given our predictions on how Florida’s season will go based on win shares.

Loosely, this is the idea that we can assign Florida a percentage chance of winning each of the games on its schedule, then tally that up to see what approximately how many games we think the Gators will end up winning.

This isn’t a scientific process, of course, and it’s a lot more guesswork now than it might be in October or November, when we know more about what teams are. But it’s an exercise in measuring expectations and sharing our understanding, and I enjoy it each year.

So I’m going to try to keep publishing win shares predictions each week. And here are mine for the 2018 season with just three days to go before Florida’s season opener.

I think a lot of these are self-explanatory, and I’ll first note that I made my predictions on instinct and rudimentary understanding of what Florida and its foes will be, then used Bill Connelly’s preview — and specifically the schedule predictions based on S&P+ — to help augment those thoughts.

But doing this really helps cleave Florida’s season into what looks to me like a couple of distinct three-game stretches to start — first, a trio of three home games that Florida should be favored in, and then three tougher games that I think Florida would do well to go 1-2 in — and then two really difficult games at the end of less challenging months in October and November.

I don’t give Florida a 1.0 win share for anything, you may recall from previous years — thanks, Georgia Southern — and while I think I’ll probably bump up that Idaho number if the Vandals are as bad as they could be in their first year since dropping back to the Football Championship Subdivision, I also keep the numbers on late-season cupcakes artificially low as of September to account for the possibility of injuries.

But I’m also more bullish than S&P+ for six of Florida’s eight SEC tilts, with the exceptions being Tennessee (because it’s this team’s first road game, and because Feleipe Franks was especially ungood on the road last year) and Georgia (where I’m only slightly more bearish than the ratings system). I’ll be overstating Florida’s chances against Kentucky until the Wildcats actually break through for a win in my lifetime, am granting a bit of a bonus for Dan Mullen’s institutitonal knowledge in regards to Mississippi State, think LSU has many hallmarks of a program about to have a down year, and feel pretty good about Florida’s chances at Vandy (where it hasn’t lost in my lifetime!) and in home games in November after Mullen’s had a chance to iron things out with this team.

And, really, I’m not that much more bullish than S&P+ anywhere. The greatest divergence between its win probability and my win share for any given game is for the LSU game, for reasons stated above, but the second-greatest divergence is me granting Tennessee — S&P+ hated 2017 Tennessee almost as much as Tennessee fans did — a coin-flip chance against the Gators. The total difference between my 7.65 win shares and its 7.2 projected wins is really not a huge one.

But that also means that I’m thinking of 2018 as a typical first year for a new coach, rather than Mullen immediately jolting a sleeping giant into title contention. If Florida enters its game against Georgia 5-2, it will be basically where I thought it would be; if the Gators go 9-3 or better, it will be a very pleasant surprise.

That’s enough rambling from me for now. So I turn the discussion over to you: What do your win shares look like? What about mine look weird to you? Where do you think Florida’s season could turn? And what, based on this exercise, are your expectations for this team?