NFL Week 8 Projections & Best Bets
This is part of a recurring series on VSiN.com in which VSiN host Jonathan Von Tobel tracks the progress of a model he created for the NFL season
Top NFL Resources:
- NFL Betting Hub
- NFL 2024 Week-By-Week Schedule
- Parlay Calculator
- NFL Betting Splits
- NFL Betting Odds
Week 7 Best Bets: 2-3 | Overall Best Bets: 20-15
Week 7 Overall: 8-5 | Overall Record: 62-42-2Â
Week 7 Recap
It was a different week for Jon Von Model in Week 7. The biggest edges went 2-3 for a second consecutive week, but the overall edges bounced back with an 8-5 record. For the season the model continues to operate at a decent pace. Overall edges are hitting at a 59.6% clip and the SuperContest entry is 20-15 on the season. However, the pace has obviously started to slow down.
Since the 11-4 start to the contest entry the model is 9-11 on best bets. This is about the time where I believe we need to take a hard look at what the model has done to this point and figure out our next course of action.
Week 8 Adjustments
This is a massive week for Jon Von Model. After consulting with Mitch Moss I have decided to make the transition to the 2023 version of the model.
Up to this point of the season the model has been using some form of data from the 2022 season. It has led to some success, but the farther away we get from last season the less accurate the best bets have become. It is time to give the updated version of JVM a test drive, and this week will be that test.
In the upcoming projections you’re going to see some pretty interesting numbers. What you will see is projections based purely off of 2023 statistics. These will have much more variance from what the projections have been since the start of the season. My hope is that by jumping over to the 2023 version of the model we will be ahead of some edges and be able to kick-start what has been a middling effort the last few weeks.
Week 8 Projections
Take it all in! This is what projecting the upcoming week of NFL action looks like based on just 2023 statistics via JVM.
Let’s start with the big picture before we get into some of the individual projections. Overall, the average edge this week is about 5.4 points which is pretty high. If you remove the biggest edge – San Francisco – then you get an average edge of about 4.2 points. That is still somewhat high, but it is part of the risk we’re taking by jumping over to the 2023 model this week.
The hope is that as the results come in the average edge will tighten up, and it has over the course of the season. Last week the overall average edge from the 2023 version of the model was about 6.6 points, so we’re seeing the model tighten up every week.
The elephant in the room is obviously the biggest edge, which is that of the 49ers who are projected to win by 25.85 points according to JVM. I’ll say immediately that Mitch and I will not be using that as a play this week.
There is clearly something off about how JVM is reading both San Francisco and Cincinnati. The Bengals have been below average in a lot of the offensive metrics I use for the model, so it is not a surprise that they are rated poorly. Having said that, this is extreme. When these two teams don’t play each other next week we will get a better sense of whether the issue is with the Bengals or the 49ers, but I believe it to be the former.
As far as the quarterback difference between Brock Purdy and Sam Darnold: There is none. Purdy has not played particularly well this season by the metrics that I use to measure quarterback play. He is the 20th-ranked passer via PFF passing grade (67.3) and his turnover worthy play rate (4.3%) ranks 32nd among qualified quarterbacks. Darnold’s career averages are nearly identical to Purdy’s current numbers, thus there is no difference between the two.
It is also obvious how much the 2023 version of Jon Von Model likes the best teams in the 2023 season. Kansas City, Miami and Baltimore are all projected to win by double-digits and all have at least 6.83 points of value according to model. All three are going to be part of the contest card this week. Some would obviously point to laying all of those points with those teams as an issue, but it should be pointed out that the market has obviously been low on these teams as they are all 5-2 ATS heading into Week 8.
So, with that in mind our five biggest edges of the week – excluding what we believe to be an extreme in San Francisco – looks like this:
Chiefs (+9.48)
Dolphins (+8.44)
Ravens (+6.83)
Texans (+6.28)
Titans (+6.14)
We are going to have to really evaluate the final selection on the contest card this week. Personally, I do not think we should use the Titans as a true edge either. Quarterbacks are factored into the metrics for JVM, but there are many team-based statistics as well. A vast majority of the profile JVM is reading is a team with Ryan Tannehill at quarterback. Will Levis – the reported starting quarterback – is a complete unknown. We do not want to go to battle with an incomplete picture.
So, if we do not use Tennessee we get to Cleveland. The Browns have their own quarterback issues, but each quarterback the Browns have used are very similar to one another statistically. Couple that with a dominant defense and an opposing quarterback in Geno Smith that has not graded well and it makes more sense to roll with Cleveland.
Regardless of what we decide for the fifth entry, this week will be a massive one for this model journey. We either now have a simple model that just needs updated statistics and results each week, or I go back to the drawing board to figure out what to do next.
I can’t wait to see the results!
