The Knish’s Model NFL Week 6: Projections, Picks, and Analysis, Part 1

104
The Knish's Model NFL Week 6- Projections, Picks, and Analysis

Every time I write out the number “six,” (In the title above, for example) I flashback to Clark W. Griswold in European Vacation stating that he and his family are looking for “sex.” That and him driving around in circles, “look kids, Big Ben, Parliament.”  That was apropos of nothing. 

Actually maybe not. In that movie, Clark took his family to a foreign country and acted with reckless abandonment. Perhaps it was a time for caution. I have a great deal of caution about this week for the following reasons:

  • It is the first bye week so our slate is reduced
  • We had two top 15 quarterbacks injured last week (Wilson and Jones)
  • We had a coach resign in disgrace
  • We have important starters (maybe) coming off of injury (CMC and Tua)
  • We have uncommon marquee matchups (Browns/Cardinals; Chargers/Ravens)
  • I think a number of medium favorites are vulnerable. 

There is one other reason, the lines are simply tighter this week. Notice we had only one teaser option, I was able to grab it Sunday night, Minnesota +7.5 and Cardinals +8.5. That play is no longer available.    

My advice is to exercise caution this week and block out the narrative as much as possible. Not having a side in a game beats the hell out of losing a side on a game you should have just stayed away from. 

Every week I do a video with Razor. I have gone 12-2-1 on that show. The picks in the video are generally not included in the write ups, so be sure to watch that video on Instagram.

Tampa Bay Buccaneers v Philadelphia Eagles (-6.5, 52.5)

Buccaneers 32.07 – Eagles 22.13

The model’s power ranking algorithm (RES) makes the line Buccaneers -9.38 on a neutral field. So the line at seven makes inherent sense. The simulations have the Buccaneers winning 9.94 points, and going over by a hair. 

If Leftwich stays married to running the ball on early downs in the first half, look for this game to stay close and finish at or around the number. If so, the under is a live play. If Tampa opens up their offense on early downs in the first half, it could get ugly quickly. In which case, look for Hurts to respond by throwing the ball 45+ times. I lean to the Buccaneers. I am taking Brady over 2.5 touchdowns, at +100.

The model is taking Tom Brady over 2.5 touchdowns +100. 


Miami Dolphins v Jacksonville Jaguars (3.5, 45)

Jacksonville 20.06 – Dolphins 19.42

Oh boy! The simulations say the Jags pull the upset and get their first win. The model’s power rankings (RES) make the Dolphins a 3.7 point favorite on a neutral field. So the line is probably thin at 3.5. The simulations have the Jaguars pulling the upset by .64 points and the moneyline is touching +170.  

The model knows Tua will be back, his brief performance to date ranks him as the league’s 34th best quarterback. The model will put one-half unit on the moneyline at +170.

The model is taking the Jags +3.5


Minnesota Vikings v Carolina Panthers (-1, 47)

Vikings 15.20 – Panthers 20.61

The model’s raw expected spread has the -.5 on a neutral field. The simulations have the Panthers -4.41.

I am not going to touch this game for two reasons, but oddly both of the reasons should make me more inclined to bet on this game. First, CMC could be back for the Panthers. That is a massive infusion of talent for a team that desperately needs it. Second, the Viking offense is explosive, but absurdly, tragically married to inefficient play-calling. Everything about this game screams Panthers. 

The model’s priors (preseason predictions) had the Vikings as the better team. I still believe they are the better team. Zimmer can concoct a scheme that makes Darnold look like NY Jets Darnold, and if they uncuff their offense, Minnesota could rout the Panthers. I am not going to put my money on the worse team with the worse quarterback. I will just sit this out. 

The model is not taking a position in this game. 


Los Angeles Chargers v Baltimore Ravens (-3.5, 50.5)

LA Chargers 29.39 – Ravens 25.00

The model’s (RES) makes the Ravens -2.29 on a neutral field. The current spread is -3, or -3.5. The simulations have the Chargers winning by 4.39 points. 

Staley is coaching a semi-masterpiece with the Chargers. His fourth-down decision-making is elite. His early-down play-calling is too restrictive on Herbert. Yes, this offense can be much better. The Ravens will be able to run, but not as well as the Browns did last week. Do not look for Lamar to recreate his MNF magic through the air. He did that against a bad and injured Indy secondary. 

The model is taking the Chargers +3.5 (Still okay at +3) and the moneyline at +130 or better)


Kansas City Chiefs v Washington (6, 54)

Chiefs 31.69 – Washington 23.88

The model’s raw expected spread (RES) makes the Chiefs -8.89 on a neutral field. The model’s simulations have them winning by 7.81 points. 

It is difficult to lay points with this Chiefs team from a mental perspective given their performance, but this isn’t supposed to be easy. In games like this, I return to my checklist: The Chiefs have the better coach, better quarterback, and better-skill-position players. The RES has value on the line, as do the simulations. This may be the final “buy-low” option for the Chiefs all season. 

The model is taking the Chiefs -6


Dallas Cowboys v New England Patriots (3.5, 51.5)

Dallas 25.81 – New England 21.76

The model’s raw expected spread makes this game Dallas -3.09 on a neutral field. That makes the line a bit heavy-handed for the popular Cowboys. The model makes this game Dallas -4.05.

This game opened at a total of 48.5. It was immediately bet up over 51 to 51.5. The number 51 is a key number in totals. Games end on 51, 3.77% of the time. That represents significant value. Additionally, this number was bet up because of Dallas’ explosive offense. The Patriots are going to drop seven or eight each play and make Dak go down the field methodically. Moreover, this is a Dallas team that wants to, on a biological level, run the ball anyway. I think this speeds up the game and keeps the total under 51.5. 

Do not bet the under at a number worse than 51.5. 

The model is taking UNDER 51.5 

About the author:

+ posts

I write about data and sports. I created my first model in 1997 using nothing more than Excel. Currently, I have data-driven models for the NFL, NBA, and World Cup Soccer.

Mathematics is the music of reason.
— James Joseph Sylvester, English mathematician

Leave a Reply