by youngterrier on Wed Dec 06, 2017 8:13 pm
The reason why they have Wofford that much of an underdog is because Sagarin and Massey have us at about that much. The problem is a case study in why the computers aren't the best at this level. They don't account for many variables, some very hard to measure.
For example, the average MVFC team runs for more yards per game and stops the run better than the average Socon team. Why? Because we have 4 teams in the Socon that run it better than the bottom 5, but not enough to compensate for the bad running teams. Meanwhile, everyone got to defend those running teams so everyone's run defense is kind of inflated. If you were to just look at the stats without context to how each team plays philosophically (not something computer algorithms are adept at as they just put in variables from each game), you would assume 1) MVFC teams are better at running the ball than Socon teams, on average and 2) MVFC teams are better at stopping it. Looking at the data, that's a valid assumption, but if you actually know how these teams work, that's another issue all together.
So, I won't say that the stats are useless, but the algorithms don't translate between conferences, especially with the Socon, because the Socon is probably the one conference that both 1) has good teams and 2) manages to have a diversity of offensive styles. If you look at the MVFC, Big Sky and CAA, they are pretty monolithic (with some exceptions), but in the Socon you have 3 option teams (each with different and statistically verifiable strengths and weaknesses), a fast pace team (Western), an air raid team (Samford), a pro set team (Mercer), a bad team (VMI), a start up (ETSU), and a team that never found its ground and was wholly inconsistent this year (Chatt). The computers can't compensate for it in the same way.
Look at our numbers for run defense, for instance. We haven't allowed an FCS team to break 100 yards on us since the Citadel did it. We haven't allowed a player to rush for 100 yards on us since Youngstown. We've played 5 good rush teams in the top 25 run offense--Furman 2x, Western, Citadel, Gardner Webb and South Carolina and held all but South Carolina below their average running the ball (in some cases well below). But looking at our overall run defensive stats, you'd think we were only slightly above average. But the reality is, when it comes to run defense strength of schedule, we had one of the hardest in the country. Computers don't care about that.
That doesn't necessarily translate to us playing well or winning, but you get the point. Don't trust computers. Anyone who says that they know with certainty that one team is going to piledrive the other is sipping the kool-aid. The safest bet is that this is going to be close unless one team loses the turnover battle by 2+
Study hard, Work Hard, Party Hard, Go Terriers!