Oh what a difference a week makes. After last week, things were looking grim. The performance just wasn’t what I or anyone else was hoping for. Then, last weekend came, and suddenly everything went just great. Every optimistic prediction seemed to come true, and now things are looking up. The pendulum has been swinging a bit back and forth so far this year, and here’s to hoping that the current upswing continues,
I am, of course, referring to the general accuracy of my weekly betting advice so far this year. What did you think I was talking about?
Oh, right, MSU looked pretty good last week too. And, how about those Wisconsin Badgers? It seems like I read a prediction somewhere how they were going to win by a little over 3 touchdowns last week, despite a spread of only 3-4 points. Who made that crazy prediction...? Oh, that’s right, I did. Yeah, that was totally me. I remember now.
This week, the Indiana Hoosiers are coming to East Lansing for homecoming and that means that the Old Brass Spittoon is once again up for grabs. The Vegas lines opened at -15.5 for MSU based on my source, but it seems to have settled in at -14 almost immediately. That corresponds to a 84% chance that MSU escapes unscathed, with their favorite spittin’ bucket still in the trophy case. My algorithm is slightly more optimistic, but only by a point or two (-16.7). The FPI came to almost exactly the same conclusion at -15.2.
Both data sources suggest that MSU will barely cover, so I will go with that, but again, the signal is no where strong enough to issue a recommendation. As for the over/under, my data suggests it should be 46 or 47, while Vegas has it around 44 or 45. I would not get within spitting distance of that one. Overall, the math suggests that MSU will win something like 31-14. That sure sounds like 31-7 with a late garbage time TD for the Hoosiers. I would certainly take that.
In order to put this data in perspective, here is my standard weekly preview graph for Week 5, based on my algorithm's projections relative to the opening spread:
For reference, here are ESPN’s FPI picks for the week
If you think that there are a lot of games crammed down into the lower left corner of both plot, you would be correct. A full 49% of the games this week (25/51) have a spread that opened at 7 points or lower. This has all the hallmarks of an exciting weekend. Also note how similar the FPI predictions are to the Vegas line at this point in the season.
As for bets that might be more appealing than betting on MSU (which is always a bad idea), here are my recommended bets for Week 5
Overall, my spreadsheet likes 6 teams to cover this week, while the FPI adds two additional bets. As for a parlay, the strongest signals are coming from the Ohio State and Wisconsin games this weekend. All-in-all, that seems like a good bet based on what we have seen so far this year, although Ohio State at Nebraska makes me pause every so slightly. This was a game in the preseason where I predicted an upset loss, and odd things can go down in Lincoln sometimes (see MSU, c. 2015).
As for upsets, my picks for this week are shown below:
In total, the two systems like 7 upsets total, and 3 of those picks are shared by both systems. My math suggests an additional 3 upsets, while the FPI like 1 more. In general, the upset picks this week are almost all from the Group of Five, with the exception of my pick of Georgia Tech over Temple. But, the mere fact the Temple is favored in the first place is perhaps the bigger story there.
I am also getting to the point of the season where I almost have enough data to completely remove the preseason rankings from my calculations. If I run these numbers too soon in the season, I can get some very odd predictions. That said, they sometimes wind up being prophetic. When I remove the preseason rankings this week, I get a couple of interesting upset picks that we may want to keep an eye on, including:
- Maryland over Penn State
- Purdue over Minnesota
- Oregon State over Stanford
- Washington State over Utah
- Kentucky over South Carolina
I should not be shocked to see 2-3 of those upsets hit as well. Finally, we weekly simulation suggests that we will see 12.8 ± 2.9 upsets in Week 5. Once again, that is the new record high number of upsets predicted in a single week so far this year. It looks like it could be a fun weekend, once again.
That is all for now. Check in after the games to see once again how bad (or good?) my advice was.