The Model Speaks: 2021 NFL Week Three Betting Analysis (Part 2 of 2)

THE MODEL SPEAKS 2021 NFL Week Three Projections and Analysis- Part 2 of 2

In the first part of this week’s projections and analysis, I recognized that there will be many teaser bets this weekend. Additionally, I cautioned about the math necessary to make a teaser bet logically. In part two, I will give the model’s projections on week two and point out some potential teaser plays.  

I also want to share some interesting statistics I learned this week. If you look at the previous 3000 NFL dogs receiving 7 points or more on the spread, you will not be surprised to learn that approximately 50% of those teams covered. What will likely surprise you is that of the 50% that covered, two-thirds of those teams won the game outright. 

Given that the average line on a seven-point dog is +275, had you bet all 3,000 games on the dog moneyline, you would have returned a profit of over 12%. 

I am not advocating that you blindly bet all 7 point dogs on the moneyline. I am trying to reinforce, as I have in several articles, is that we tend to exaggerate the perceived differences between teams in our minds. 

Indianapolis Colts v Tennessee Titans (-5, 48)

I hate this game. The model has a terrific track record when it believes a dog is going to win the game. Here, it believes the Colts are going to beat the Titans. It also believes this outcome, only if Wentz Plays. If you have read my articles, you know I am not exceptionally high on the AFC South or Wentz. Now my model is suggesting I bet both and bet Wentz as a dog. 

My trepidation is eased to a degree by the fact that only 18% of the spread tickets are on the Colts, but that represents 47% of the handle. I have grabbed a little taste at +5.5. I will wait for news on Wentz later in the week. 

The quarterback situation is worth monitoring. Wentz has two sprained ankles, so it may be Eason and or Hundley. The model removes just over two points from Indianapolis if Wentz is out. However, the line will move with the announcement. 

Wait for the last minute and take the Colts +5.5 with Wentz, or +6 or better if he is out.

Arizona Cardinals v Jacksonville Jaguars (-7, 52)

Arizona is fun to watch. If you are a fantasy player, you likely value Murray more than perhaps those of us that are not. Either way, he has been fantastic to start the season. The Cardinals were lucky to beat the Vikings last week, and the hype is baked into the spread. 

If the model had to take a side on this game, it would take the Jags. Mercifully, the model can pass, and it is passing on this game. 

If you are considering taking this game, I would think about teasing the Cardinals down and pairing it with another team that you can tease across seven and three. There are plenty this week. 

Trevor Lawrence throwing no picks at +176 has too much value for me to pass on personally. This is not a model play but a personal play.  

The model is not taking a position in this game. 

Los Angeles Chargers v Kansas City Chiefs (-6.5, 55)

The model grabbed the Chargers at plus seven but is not interested in the lessor number. Kansas City has been placing close games for over a year now. The fact that they keep getting large numbers is a testament that people more vividly remember extremes. 

For example, many people remember being stuck in line at the grocery store by someone moving glacially slow. Perhaps they are paying with pennies or writing a check. We remember these events and then tend to normalize them. You will often hear people say they always choose the slowest line at the store. That is not mathematically likely. 

Rather what happens is that we remember those times we do get the slowest line because we are generally in a hurry and wanting to do or be somewhere else. We do not recall the countless times we sped right through the line. 

That is our collective association with Kansas City, a dominant high scoring offense. They may be, but they play close games, and this year, like last week, they will lose a few more. 

The model has this side at +7. It does not suggest it at +6.5. Kansas City could be a decent teaser option. 

Atlanta Falcons v New York Giants (-3, 47)

The model is looking incredibly foolish for being high on the Atlanta preseason. It is always important to reevaluate your initial positions in light of new data. Too often, we cling to our priors. I tend to admit the error and move forward, looking for value. 

Having said this, the model is off this game entirely, but I am highly interested in it. We will learn a great deal about both teams. 

Arthur Smith is finding his footing as head coach. The first week he looked out of his depth. Last week he returned to some of the trademark things he does well, especially with the tight end. Pitts had a great game last week.  I expect him to continue to do that in this game. Unfortunately, the lines are expecting this too, and Pitts player props do not offer much value. 

The model has no position in this game. 

Cincinnati Bengals v Pittsburgh Steelers (-3, 44.5)

The model has no side on this game, but it would lean heavily to the under. Neither team is going to want to put the ball in the hands of their quarterbacks. This is a game that should move exceptionally fast and stay under even this low total.  

Pittsburgh drops this game to Cincinnati, I believe we will finally start hearing the discussion that should have already started – Ben held on too long. 

The model is not taking a position in this game, but it would lean to the under. 

Miami Dolphins v Las Vegas Raiders (-4, 45.5)

I have written about this game in another piece for BLS. I have pasted the relevant portion below.  

The line movement has been unusual. Driving the move was the Dolphins’ poor performance, the Raiders upset in Pittsburgh, Tua’s injury, and speculation around Carr’s injury. Carr is a go this weekend.  

The Dolphins received a ton of the early action in this game, especially from smart money. Those bettors moved in at +5.5. They pressed the line down across 4 to -3.5. Then they bought the Raiders back at -3.5. They are sitting with a middle of 4.  

This line is too mature for the model to suggest action this week.  If anything, it would consider a position on the over.

New York Jets v Denver Broncos (-10.5, 41)

Denver is going to continue to play well and be efficient with Bridgewater. I strongly advised an over bet on Denver’s win total before the season, expecting them to emerge 3-0.  Wilson has struggled to stay upright and as a quarterback through his first two games. Both of those trends should continue in the altitude in Denver and against this defense. Fangio will give him fits. 

The model is not taking a position in this game. 

Seattle Seahawks v Minnesota Vikings (1.5, 55.5)

The model does not want anything to do with this game.  

From the Seattle perspective, the model is concerned that Carroll will do what he has always done after surrendering a fourth-quarter lead – take on a significant role in the offense. That role will undoubtedly involve taking the ball out of Wilson’s hand and running the ball at an irrationally high rate. Historically after such losses, Seattle tends to go run, run, pass, punt. 

Wilson is the best quarterback in this game, and the ball should be in his hands. Minnesota is vulnerable in the secondary, as we have seen all season. 

The model believes the Vikings are more similar to the team that played last week against Arizona than the team that played Cincinnati. Cousins is boring but significantly underrated. The model rates him as the league’s ninth-best quarterback in the league through two games.  He is not elite, but he is firmly in the second-tier quarterbacks. 

Zimmer historically gets the most out of his defensive talent and will have his unit prepared. I hate to call a game in week three a must-win, especially for a team in the NFC North, but the Vikings will go for the jugular early against Seattle. Minnesota is one of the few teams in the league that has a decided home-field advantage. 

Teasing Minnesota up and across seven is another possibility for teaser bettors. 

The model has no position in this game.

+ posts