THE MODEL SPEAKS: 2021 NFL Week Four Projections and Analysis, PART TWO

116
THE MODEL SPEAKS- 2021 NFL Week Four Projections and Analysis PART TWO

Readers will notice that the model is very selective with its plays. It takes, on average, four positions a week. In an attempt to offer more value to the reader, especially in the games where the model is not taking a position, I have included player props in which the model sees value. They are sides that are worth your consideration. If a prop is listed at the bottom of the analysis and highlighted, I am advocating you tail the side. Only those sides will be tallied in the model’s results. It is necessary to point this out as transparency is critical.  

New York Giants v New Orleans Saints (-7.5, 42)

This is the season where you back the Saints in odd-numbered weeks and fade them on even weeks. 

This week Judge was asked why he does not use analytics more in his game plan. He responded, “if Excel won games, Bill Gates would be killing it.” This is a fascinating response. First, I have no problem believing that Bill Gates would run the Giants far better than Gettleman. Judge may have just given the Mara and Tisch families an idea.  

Second, you only hear this nonsense from losing coaches. I have a follow-up question for Judge, “You do not use analytics, Coach Judge, and you are not killing it? You are quite good at losing. Your win percentage is just .316. Why do we only hear this nonsense from losing coaches that do not use analytics? Coincidence?” I have never heard McVey say, “all these damn analytics … they do not help me win at all! 

Joe Judge wins 31.8% of his games as a head coach. Here is a list of things you are better at than Joe Judge is at winning football games:

  1. Picking the correct door on Let’s Make A Deal
  2. Picking the correct “third” on a roulette table
  3. Picking the winning side ATS in the NFL

Just think about that for a minute. You have a better chance of winning a -110 bet on any NFL game than Joe Judge has in winning any NFL game he coaches. 

If I were Joe Judge, I would think that I would be open to anything at all that might make me better at winning football games. Not Joe Judge. He is content to be the latest in a long line of the catastrophically horrendous Belichick coaching tree. Hey coach, your ignorance is showing! The Giants brain trust, Gettleman, Judge, and the Clapper stretches the definition of the word brain. 

The Saints are okay.  They are not the team that whipped Green Bay in week one. Neither are they the team that the Panthers laid out in week two. They are exceptionally well-coached. They have an excellent defense. They play a ton of man defense and stop the run. On offense, it is apparent that Payton does not fully trust Winston, at least not yet. They are going to run the ball, utilize the short passing game and try to win ugly.   

As much as I would love to bet against them, there are just too many points here. The smart play is either staying away or holding your nose and taking the Giants +7.5. 

The only player prop with value is Winston Over 199 yards, -115

The model is not taking a position in this game. 


Kansas City Chiefs v Philadelphia Eagles (7, 54.5)

After seeing what Dallas did to Philadelphia on Sunday, it is hard not to think that Philly will get boat raced in this game. The model has been highly suspect of Hurts. His numbers looked great against Atlanta. He only completed eleven passes against an injured San Francisco defense. He imploded against a Dallas defense that has its own question marks. The Eagles quickly became everybody’s darling. 

Besides, it is just and proper that Philadelphia lose. They claim to be a great sports town, but I have a hard time taking that seriously. Philly is a town that put a statue of a fake boxer on its streets (Rocky) before it even erected one to its own, raised if not homegrown, actual Heavyweight Champion of the World, Smokin’ Joe Frazier. Unforgivable, yet it perfectly encapsulates everything I know about Philadelphia. It is the curse that nobody talks about. 

Everybody knows by now that the Chiefs have not been able to cover spreads dating back through last season. Everybody knows that defenses are putting a roof on the backend of their defenses with exotic combinations of cover two. They are forcing the Chiefs to play small ball, 

Still, if I were going to take a side, it would be the Chiefs in an absolute thrashing.

One aspect of the cover two defense that the Eagles employ is that it leaves the intermediate middle of the field vulnerable. This is where Kelce lives. Kelce has the most targets of any player against cover two in the last two years.  

The model is taking Kelce Over 7.5 Receptions +105


Houston Texans v Buffalo Bills (-16, 47)

I am not going to lay 16, even though the model sees a 24 point victory. The problem, of course, is that the teams let off the gas in these situations. There is also no way on earth I am taking the points with Culley’s Texans.

If you want to bet on this game, I suggest working in the derivative markets. Look at first-half team totals and sides. You can fully trust that the Bills will have Allen spinning the ball often on early downs in the first half.

I would also look to player props for action. I have listed the ones with value below for your consideration:

Davis Mills throws no interceptions: +171

Davis Mills throws no touchdowns: +180

Josh Allen throws Over 2.5 Touchdowns: +120

Mark Ingram Rushing/Receiving yards Over Over 37.5 -116

I would love to find one on Mitch Trubisky to complete a pass. He may play the entire fourth quarter. 

The model is not taking a position in this game. 


Arizona Cardinals v Los Angeles Rams (-4.5, 54.5)

This game will tell us a lot about the Cardinals. This is where they get their “bona fides.” The model thinks Arizona is suspect:

Still, the model sees them hanging in this game. It is the odd result where the most common outcome, according to the model, is that Arizona covers, the second most likely result is that Arizona loses by double digits. 

Smart money came in heavy on Arizona. They are looking for a Rams’ letdown after last week. Or simply a small regression. 

Like the Buffalo game, this is a game where I am primarily interested in the derivative markets.  

I would like to buy the total down to 50.5 and bet the over. The model has this game at 56.12% to go over. By buying down to 50.5, I pick up 51, 52, 53, and 54. That equates to 9.32% of value. So I can take my win percentage to 65.44% (56.12 + 9.32). The best alternate line I have seen is 50.5, -190. That is a slightly worse bet than 54.5 -110. I will need to find -170, and this is unlikely. 

The model is taking Stafford Over 2.5 TD Passes +145.  


Seattle Seahawks v San Francisco 49ers (-3, 52)

I am not interested in this game. I will say three is probably the correct number. There will likely be some highly improbable play that determines the winner. I do not like to bet on highly unlikely plays determining the outcome. 

I am half-joking, but I might be interested in a position of three being the exact final margin of victory. I would need +550. 

I am content to watch this game. 

The model is not taking a position in this game. 


Baltimore Ravens v Denver Broncos (-1, 45)

The model is high on Denver from a total win perspective, it has a ton wrapped up in the over. However, it sees Denver as a bit of a paper tiger. Three impressive wins by the numbers, but they came against the Jets, Giants, and Jaguars. 

The model is not as high on Baltimore as most. It picked them to finish second in the AFC North. Baltimore’s schedule starting week 10 is brutal. They have to have these early wins. They squeaked two out against the Chiefs and the Lions. 

The model has the game in line with the market, so it is not taking a side. Moreover, according to the model, nothing in the derivative market nor the player prop market is worth mentioning. 

The model is not going to take a position in this game. 


Pittsburgh Steelers v Green Bay Packers (-6.5, 45.5)

If you have read anything I have written, you know how little the model thinks of Pittsburgh. Now Big Ben takes his Hall of Fame EPA per pass play of -.06 to Green Bay. Yes, Ben has a negative EPA on pass plays. He also has a 1 to 2 ratio of elite throws to interception-worthy throws. Rodgers had one horrific game but has been lights out since. 

This projects to be a low-scoring game. Green Bay does not have the offensive line to hold up against the Steelers’ defensive front long enough to take a ton of deep shots. I expect them to continue with their RPO offense and sustain drives early. Pittsburgh does not have the skill, especially at quarterback, to take advantage of the suspect Packers’ defense. 

The spread is right in line with model projections, but it does see some value on the under. There is also value on the following props:

Rodgers Yes to throw an interception, +205

Harris Over 14 Carries, -114

The model is not taking a position in this game. 

Tampa Bay Buccaneers v New England Patriots (5.5, 48.5)

This game is going to be hyped beyond belief. Brady returns to Gillette. Brady v Belichick. The bad news for the Pats is that the Buccaneers are far superior, and Brady. 

This game will share several similarities with New England’s game last week against the Saints. I predicted last week that Mac Jones would throw his first pick against the Saints, and he did. He could throw another in this game. Both the Saints and Tampa Bay are structured to exploit Jones’ inexperience as both teams are tough to run the ball against. Jones will have to do more with his arm again this week.

It is not as tough of an assignment for Jones because Tampa Bay’s secondary is neither healthy nor to the level of the Saints. If he can hang in the pocket, he will make more throws than he did against New Orleans. 

Everybody is on the Patriots. The model agrees but is begging off this game. 

The model is not taking a position in this game.


Las Vegas Raiders v Los Angeles Chargers (-3.5, 52.5)

The Raiders have become a popular play with smart money. The reasons are relatively straightforward. The Chiefs are at home, off an impressive win against the Chiefs, the Raiders struggled with the Dolphins, and Justin Herbert is the NFL’s latest Kid Gorgeous. 

While the above is all correct. I even wrote that if you gave me a team, put all the players in a common draft, and gave me the first pick, I would take Herbert over Mahomes. That is primarily due to the contract, but it is also a reflection of Herbert’s talent.   

Still, Carr has had the better season thus far. If you haven’t noticed, Carr is playing out of his mind. The advanced metrics are undeniable. Moreover, while Carr is limited in the number of weapons at his disposal, his connection with Waller, and the devastation it wreaks on defenses, has proven to be a repeatable advantage. 

It is too many points to give away in a divisional game. The smart money is correct. The side in this game is the Raiders. 

The model is on the Raiders +3.5

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