2017 WEEK 15 PICKS

I continue my slow and steady descent into mediocrity: I went 8-8 for the second week in a row. Somewhat symbolic of this “balanced approach” to picking games, is the fact that while I was a victim of a bad beat with the CLE/GNB game – the Hundley-to-Adams 25-yard touchdown in OT blew my Browns to cover pick – I got lucky when Bills RB LeSean McCoy ran in for a score in OT against the Colts, giving me the Bills cover. That being said, there’s some good news: I hit two of my three Super Hot Sizzling picks (ARI +2.5 against TEN, JAX -2.5 against SEA).

Where I’m at:

All Picks: 111-97, 53.4%
ESPN Leaderboard Rank: 8507, 95%
Super Hot Sizzling Picks: 21-18, 53.8%
Bankroll: +$43 (betting $11 on every game)

Feels good to be spending countless hours on a spreadsheet with 55 columns and have $43 to show for it. Awesome.

WEEK 15 PICKS

DayDateAwayHomeMy
Spread
ESPN PP
Spread
My
ATS
Pick
Pred
Score
Thu12/14/17DENIND-1+2.5IND16-17
Sat12/16/17CHIDET-4-5.5CHI19-23
Sat12/16/17SDGKAN+3+1.5SDG23-20
Sun12/17/17MIABUF-1-2.5MIA19-20
Sun12/17/17CINMIN-12-10.5MIN15-27
Sun12/17/17BALCLE+7+7.5CLE24-17
Sun12/17/17GNBCAR-2+0.5CAR26-28
Sun12/17/17NYJNOR-16-15.5NOR16-32
Sun12/17/17PHINYG+12+7.5PHI26-14
Sun12/17/17ARIWAS-7-4.5WAS18-25
Sun12/17/17HOUJAX-9-11.5HOU17-26
Sun12/17/17LARSEA+3-2.5LAR26-23
Sun12/17/17TENSFO-1-1.5TEN20-21
Sun12/17/17NWEPIT+2+2.5PIT26-24
Sun12/17/17DALOAK+5+2.5DAL26-21
Mon12/18
/17
ATLTAM+5+6.5TAM26-21

Not much to say except that The Machine is liking the underdogs this week. Can’t believe I’m taking Cleveland; not feeling great about laying all those points with New Orleans.

QUICK AND DIRTY WAY TO ADJUST YOUR PREDICTION SPREAD FOR CHANGES AT THE QB POSITION

I’m going to lay out a possible solution to a problem that I’m assuming a lot of novice handicappers (like myself) have been dealing with this season: changes at the quarterback position. Off the top of my head, we’ve had the following changes at QB:

Tyrod Taylor/Nathan Peterman/back to Taylor
Aaron Rodgers/Brett Hundley/back to Rodgers
Sam Bradford/Case Keenum
Deshaun Watson/Tom Savage/T.J. Yates
Eli Manning/Geno Smith/back to Manning
Carson Wentz/Nick Foles
C.J. Beathard/Jimmy Garoppolo

Losing any player at any position can affect the expected performance of a team, but it’s safe to say that a QB change carries the most significant impact. From a handicapping perspective, the problem isn’t knowing which QB is better; for example, we all know Rodgers is better than Hundley. The question is by how much? Let’s look at the Packers/Panthers game. The last time I checked, the Panthers are favored by 3 points over the Packers (the ESPN spread is locked at Panthers +0.5). Now, my model, The Machine, has Panthers winning the game by 9 points, and the reason for that difference is obvious: The Machine, using past results, doesn’t know that Rodgers is back in. So, how much do I need to adjust this spread to reflect this QB change? What’s the difference, in points, between Rodgers and Hundley? Or Wentz and Foles? Or Savage and Yates for that matter? Here’s one way to estimate this:

The key is to use a QB’s ANY/A as a scoring model. If we can convert that stat to points per game, we’ll have at least a rough idea of the difference between the two QB’s. This isn’t my idea at all, it comes from this post on the website Code and Football, and in any event, the idea of connecting yards to points has been around for a while (Expected Points, etc.). To convert ANY/A to points per game, all we need is two pieces of information:

1. The conversion factor between points and yards. We’ll use 0.075 points per yard, since this is what we get when we reduce the current ANY/A formula to a scoring model (again, see the Code and Football post).

2. The expected number of drop backs by the QB. This can vary depending on team styles and in-game conditions; to keep things simple, I’m going to just use the league average of 36.6.

Let’s start with Rodgers and Hundley.

If we take Hundley’s current ANY/A of 4.28, multiply that by 0.075, we get 0.32. Multiply that by 36.6 (the average number of drop backs per game), and we get 11.8 points. We’re going to say Hundley is “worth” about 11.8 points per game.

For Rodgers, we’ve only got a few games to look at this year, so, we’ll just use his career ANY/A of 6.89. Multiply 6.89 by 0.075 and then by 36.6, and you get 18.9 points. So, per this quick and dirty method, we can say that Rodgers is about 7 points “better” than Hundley. I probably need to add about 7 points to my Green Bay/Carolina spread. That puts it at Carolina -2, which is close to the Vegas spread.

Let’s do Carson Wentz and Nick Foles. For Wentz, we’ll use just this year’s ANY/A; after all, he’s playing a lot better than he did last year, and there’s no indication (in my mind anyway) that his performance would take a nose dive against the Giants. So, at 7.43 ANY/A, Wentz is worth 20.4 points per game. Now, getting Foles’ points per game is going to be tricky because he had that ridiculous 2013 season where he had an ANY/A of 9.18. That season seems like an anomaly, but let’s leave it in and see what we get. His career ANY/A is 6.32, meaning he’s worth 17.3 points per game. Is Wentz really only 3 points better than Foles? Maybe. I’m thinking since we really don’t know what we’re going to get (I don’t anyway), let’s just say he’s an average QB. This year the average QB has an ANY/A of 6.0, which translates to 16.5 points per game. Using that number, we could say that Wentz is about 4 points better than Foles, which feels right.

Finally, we might as well look at Tom Savage and T.J. Yates. For simplicity’s sake, let’s just use their career numbers:

Savage: 4.63 ANY/A, 12.7 points per game
Yates: 5.01 ANY/A, 13.8 points per game

This says that Yates would be about a 1-point improvement over Savage. In my mind, not enough to even bother with, so no need to adjust my spread.

As noted, this isn’t meant to be definitive, but it does at least give you some way to estimate this. And the cool thing is, you can adjust it any way you want – if you think Eagles OC Frank Reich isn’t going to let Foles throw it that much, you can adjust the drop backs. If we say Foles is going to throw it 30 times, then maybe the difference between Wentz and Foles is closer to 7 points, etc. And finally, you’ve probably noticed that I’ve left out rushing performance, that can also be figured in somehow in the future as well.

Alright, I’m set: I’m adjusting the Packers spread by 7 and the Eagles spread by 4.

SUPER HOT SIZZLING PICKS OF THE WEEK

Denver at Indianapolis, Broncos by 2.5
My ATS pick: Colts

Yep, I lost this one. Both of these teams stink, and The Machine figured the home team would take it. It actually was looking pretty good after Seimian threw that ugly pick and McManus missed a field goal, but then Brock caught fire and that was that. My (incorrect) prediction: Colts 17, Broncos 16.

Philadelphia at New York Giants, Eagles by 7.5
My ATS pick: Eagles

If you read the above discussion, you know that I’m docking the Eagles 4 points for Wentz being out, but even after doing that, I’ve still got the Eagles crushing the Giants by 12 points. This seems like too many points, but I’m thinking the Eagles defense can make this happen: Eagles 26, Giants 14.

Los Angeles Rams at Seattle, Seahawks by 2.5
My ATS pick: Rams

Can’t believe it, but the Rams are my top pick of the week. All year, I’ve been a non-believer, fading them every chance I get, and now I’m all in…meaning, I’m probably too late and there’s no more value on them! In any event, the Rams have been great on the road all year (+8 point differential) and Seattle is banged up. I don’t think the Rams will be intimidated at all playing in Seattle; they take one more step towards winning the division by winning by a field goal: Rams 26, Seahawks 23.

Have a great week all!