Last Week: 4-4-1
Season: 25-15-2
Mortal Lock: 3-1
What happened last week?
In a word, coaching.
Hugh Freeze never adjusted to Syracuse’s loaded front. BYU, Michigan decided to play Dave Rader: Up by three scores at the half, both teams crawled into a shell rather than working on areas that still need improvement. What happened then? Michigan ran 17 plays the entire second half; 13 were dives right into the LOS. Even Rutgers can figure that out. BYU cleared the bench and promptly decided to give up 20 unanswered points to one of the most anemic offenses in the country. At home. Paul Chryst looked at ND’s secondary, the unqualified best part of the Irish, and decided he would gameplan to throw the ball 3 times more than run it — against a team that has allowed 4.7 YPC on the ground — and do with a QB who is only hitting 56% of his passes. In Chicago.
And so on down the line. From 11:00 a.m. to 11 p.m., it was a clinic in professional malpractice.
Such is life. An algorithm can explain 92% of the variance as to why teams win a game and do so by X points, and it’s fairly effective at predicting which games will (Mine do so in about 4 out of 5 cases, and at 90%+ confidence). But when you get a perfect storm of those 8% all falling on the same day? Plus bad coaching? There’s not much you can do about it except pucker up, take the L and regroup. That’s why you play for ROI over a season; week-to-week agonizing gets you nowhere, except making desperation picks. You’re already up almost 2:1 on the season. Don’t panic.
And, well, if you bet on this first set of games, you are panicking:
Not with a 10’ Pole and Stolen Money
Pitt -3.5 at Georgia Tech — Want the definition of a #GoACC game? Here it is. One team that wins with defense, forcing a lot of TOs, and praying they don’t have to swap scores...but that is also so flaky they can lose to a MAC team at home one week and beat UNC soundly the next? Or do you want the team with a fantastic explosive offense that hasn’t played a G5 defense with a pulse yet...and that also gives up a ton of explosive plays? The Panthers secondary can make even Tennessee look good. That’s all I’m saying. Calculated final: Pitt 24 GT 20. And nope, nope, nope.
You can also throw Tennessee +3.5 at Mizzou in here as well. These two teams are fighting with Vandy for worst defense in the conference. The Tigers’ defensive front is a particular concern; it is the worst in the SEC. I expect if Doughboy has any brains he’ll just line up and run right at and over a bad MU front. But is he that smart? And can Tennessee’s defense actually get a stop either? It will probably be a fun game, but it won’t be a good one.
Alabama -14.5 vs. Ole Miss — When this game briefly got to 16, I was going to mortgage the baby on the Rebels to cover. As a fan, it’s the one that has scared the hell out of me since last October. In a cold, calculating analytical sense, Alabama really is the better team — even if it doesn’t always appear to be.
The Rebels limit explosive plays and gain plenty themselves. But they have been susceptible to teams lining up and imposing their will with possession football; and, absent their giant explosive runs over 20+ yards, Ole Miss is averaging a meager 2.9 YPC outside of the homerun balls: so, it’s feast or famine in Kiffinland — and that’s against a very bad early schedule for the Rebs. Alabama, absent its explosive plays, is still nearing 5 YPC. It’s not sexy, but it just might be effective.
In toto The Numbers say Alabama by 13-14. I have ran five simulations, using a variety of calculations, and only one gets you over 14.5. The Rebels cover the other 4, but they don’t win a single one — and they all go below the ridiculous o/u of 80.
I don’t want to bet either against my team or bet on the Tide to suddenly put it all together. 37-30 is the closest one I’ve gotten, 47-30 is the largest Alabama MOV. And something in the neighborhood of 41-28 is composite number I get, with a junk score for the Rebels in a hard fought loss. And analytically that sounds about right. But don’t put a dime on this one either way. For all the talk about how this is the one that Lane Kiffin most wants to win (and it is), it’s also the one that gave Nick Saban nightmares for the last 11 months. And I’m not betting against him to have been working on this game intermittently all month.
Alabama 37 - Ole Miss 30, junk time score for the Rebels to make the scoreboard pretty.
There are about half a dozen more to steer clear of over here, as well as why !
Ridiculously Large Spreads I’m Taking Nonetheless:
UNC -20 vs Duke — I know, I know. We don’t trust Mack Brown here at GAM, and for a damned good reason. But we do trust Mack Brown when it comes to lighting up bad teams. The man lives for running it up on the Directional Texases of the world to distract from the inexplicably flat performances in meaningful games. And, pals, do the numbers love this one.
UNC should have a field day throwing the ball. They’ll give up a few freebies on defense, sure. But that Duke secondary and offensive line are catastrophic sieves. This is another of those “please ignore my bad loss to Georgia Tech” beatings that UNC has come to expect under Brown, even if it’s a rivalry. The data don’t suggest anything resembling a close game. This is the closest score I’ve gotten: 45-20 Carolina.
New Mexico State +27 1/2 at San Jose State — A buddy of mine in Las Cruces at the NMSU athletic department told me that the Aggies are secretly dreading this trip to Silicon Valley. They fear that this is going to be the Spartan’s “unf*** themselves game” at the expense of New Mexico State (that’s a great phrase, BTW).
The algorithm suggests they have a lot to fear, as well. In a game of matchups, it’s a bad one. NMSU gives up a ton of pressures (which SJSU thrives upon), and is very susceptible to explosive passing plays (again, which SJSU thrives upon). Indeed, San Jose State has to rely upon it, because these two teams have dreadful rushing attacks. The difference is, SJSU has a quality passing game.
The Sparty front, the Sparty secondary, Sparty’s better passing game, and licking their wounds after a tough opening month all finally pay off with an “unf***themselves game”. SJSU 44-13.
Coastal -34 vs. ULM — Year One Terry Bowden is good for two things. The improbable upset, because guys are playing harder but they’re not his guys yet. And then getting murdered the rest of the schedule.
Tick off box one: Last week the sluggish Troy offense again couldn’t get off the ground, but this time, their defense abandoned them en route to a S/U loss.
This week, tick off Box Two: This week no such heroics shall be forthcoming from the Warhawks. Nope, our spread-covering BFF Jamie Chadwell will line up and run for about 400 yards and force several TOs. The numbers say: 51-10 Coastal, and given his record of going full Tom Osborne on crappy teams (and covering at home), I’m riding it. This game could be out of hand by halftime.
Games The Pewter Likes:
USC -7 at Colorado. The most mercurial team in CFB, the USC Trojans. If you can punch this team in the mouth, you can beat them. The problem is, the Buffs can’t; it is is the worst offense in the Pac 12 by a country mile. It’s not even close. But, hey, at least they also struggle with explosiveness — creating and stopping it.
The road trip to Boulder is tough to adjust to, and coupled with a general softness that permeates Troy, will be worth some points. Enough? Nope. Not nearly enough. And the secondary has nowhere near the size and speed to keep up if and when USC goes to the air. I hate betting on USC; I really do. But even with their road demerits, and accounting for the flakiness of the Trojans, they’ll win what should be an ugly game. Nothing comes easy for this Trojan team: USC 24 Colorado 14 (Nota Benne: I added almost a full score to CU’s performance and docked USC a full TD. The composite score was actually USC 29 CU 9).
Houston +5 at Tulsa—Few teams have had as rough a start as the Golden Hurricane: From FCS Semifinalists to the Horseshoe to Stillwater to the facing the Nation’s No. 1 offense, there’s not been a real breather. To call a conference game a breather is a stretch, but this is the first team Tulsa has faced that’s not been...weird. No weird scheme, no weird talent gap, no weird opponent motivation. It will show.
Tulsa is the better team here, despite the record. Houston is not a good team. It’s a 1-3 team wearing the skinsuit of a 3-1 team. Their wins are illusory, to put it mildly. And against the first real team they faced, they lost by three scores...at home. The Cougs make too many mistakes; Tune (QB) is especially one of the sloppiest guys you’ll see this year. And that’s the best way to describe UH: undisciplined, sloppy, and just a team.
The Numbers say Tulsa forces 3 TOs (+2 TOM), gifting them several short fields, and Golden Hurricanes win 31-23. I’ll go with it.
Michigan State -10 1/2 vs. Western Kentucky — This is one of those games that’s very simple to diagnose: middle of the pack G5 team visits more talented, more explosive P5 team that loves to force turnovers. Nothing Western Kentucky does well matches up well with what MSU does. The pewters love Sparty, and with such a forgiving spread, so do I. WKU has been good at late garbage time scores (IU, Army), so there’s that to consider. But the algorithm presumes 60 full minutes, and that still doesn’t get a WKU underdog cover. MSU is the call, 38-20
Texas -5 at TCU -- Want to see a very bad defense? I give you Texas. Want to see a very bad run defense? I give you TCU. The difference is that Texas is giving up a lot of junk yards and scores in the middle of blowouts. TCU is getting rang up like a Walmart Black Friday special by teams like Cal and SMU. This may be the worst Horned Frogs front of the Patterson era.
Texas just also so happens to have discovered that its strength lies on the ground: it has gone over 300+ yards each of the last two weeks. That’s a bad time for the Frogs.
If Texas shows up and continues to control the ground, then this is a W. Don’t screw this up, Sark. Texas 41 TCU 30
Boise State -6 1/2 vs. Nevada — Sometimes the better team is just better. And the Wolfpack’s vaunted pass rush has been a no-show against FBS teams, and its offense has not even approached 400 yards against those lower-tier Power 5 clubs. BSU is very much on par with a team like K State, and that game was instructive for this one.
Nevada is a one-trick pony, and if Carson Strong’s arm doesn’t have it that day, then neither does Nevada. UN can also be controlled on the ground, where Boise figures to do its worst damage. This game won’t be a one-sided affair, just a W. And the numbers say, coupled with a formidable home field advantage to-boot, that the Broncos pick up a quality win in a lower-scoring game than you’d expect for the MWC: Boise State 27 - Nevada 17
A few others that the Algorithm likes:
Kent State - 16 1/2 vs. Bowling Green
Memphis -11 at Temple
Charlotte +11 at Illinois
UTSA -20 1/2 vs UNLV
Ohio -9 1/2 at Akron
Mortal Lock — with a caveat:
There is an anomaly this week. There’s absolutely no game that shines by the numbers that suggests it should be the one guaranteed Mortal Lock. If I had to choose some wildly and randomly, I’d probably say that Wisconsin/UM go under; that Arkansas State lights up Georgia Southern; that A&M/MSU go under; that SMU rings up USF, and that WVU beats Texas Tech.
But in terms of analysis, all the games I selected this week all fell in the same QA zone — not a single game this week went above the 90% confidence interval. And there are usually 2-3 that do, then I throw those out as the Mortal Lock. The ones I listed above come close, at the 85% mark, but never exceed 90%, and thus cannot qualfy for Mortal Lock. That holds true for the pick below: The CI was only 87%; short of Mortal Lock territory, but based on my eyeballs and consistency, the Demon Deacons have been the most complete team in the ACC so far this season. So, we’ll go with Wake.
Wake Forest -6.5 vs. Luhvl -- If the Cardinals have shown Vegas something this year, I’ve not seen it. Margins on UL’s games have been grossly lower than expected (looking at you -9.5 Ole Miss). Casuals must be betting the Cards heavy, because there’s never movement on the lines, either. This is another such game. The Demon Deacons are just better in every practical category, and particularly defensively and with ball security; two of UL’s bugaboos.
Louisville is improving, so this won’t be a beating, but data suggests 34-23 Wake, and that sounds about right. Even more do I like Wake’s perfect 3-0 ATS vs. FBS teams.
I like this game, and especially at home.
—
Sorry this was so late. There were a lot of games that required going through by hand this week: 44, in fact. Usually there are about 28-30 that I run detailed analysis on based on preliminary data. The extra 50% workload mattered a lot.
If you want more of these, I usually pick another dozen or so and another Mortal Lock (when available) over here, at (Almost) Giving Away Money
Poll
Does Alabama cover -14.5 at home?
This poll is closed
-
45%
Nope, but gets a win and that’s all the matters
-
51%
Yep. The Crimson Tide are motivated and tired of hearing "dynasty’s dead!" takes.
-
2%
Not only do they not cover, they lose at home, as Lane Kiffin becomes the first Saban assistant to hang the L on him.