3 NFL Win Totals You Should Bet Based on Pre-Draft Season Simulations
With the lull in sports going on right now, it's easy to get a little stir crazy. Of course, while I can't bring back the NBA or NHL or MLB, or PGA Tour, I'm doing what I can to get my sports fix.
Lately, that included building out season simulations for the 2020 NFL season because -- why not? FanDuel Sportsbook has win totals for all 32 teams and plenty of other NFL Draft and season props.
In order to see where the value stands out, I wanted to take a good hard look at the upcoming schedule and early offseason changes to get out ahead.
Here is the bare minimum of what you should probably know about the process as well as the results of the best values on the market.
A big piece to projecting an upcoming season is using data we already have but also not accounting too heavily for the prior season of results. Things like win totals and Super Bowl betting odds are tied more strongly to last season's results than the upcoming season. On that note, we can't just rely on last year's winning percentages to project the following season.
As for more specific betting trend information, the overs for teams projected with high win totals (like 11.5 for the Kansas City Chiefs) hit at a bit of a lower rate than teams with win totals around 6.5 (like the Detroit Lions). There are also reasons to be bullish on teams projected to improve.
So, what does all this really mean? It means that I'm not accounting at all for a team's 2019 win-loss record but instead am leveraging numberFire's Net Expected Points (NEP) data to account for team performance from 2019. The more advanced data is less tied to results and more tied to expectation. After all, numberFire's data has been pretty reliable in predicting next year's Super Bowl winners.
Using historical regression and some manual updates to offensive and defensive predictions for teams with major overhauls -- such as Tom Brady's joining the Tampa Bay Buccaneers and the New England Patriots' uncertain quarterback situation -- I have simulated out the 2020 season schedule 10,000 times.
These teams were the most and least likely to hit the over or under in that span.
Over/Unders to Target
Dallas Cowboys (Over 9.5 Wins)
Last year, the Dallas Cowboys ranked sixth in point differential, outscoring opponents by 113 points, or 7.1 points per game. They finished just 8-8 despite a Pythagorean win mark of 10.7. Via our nERD metric, the Cowboys were 7.57 points better than the average NFL team on a neutral field. That ranked them fifth-best in the league.
So, again, while we can't just look to last year and project it forward, we're looking at the underlying data here, and it checks out.
Due mostly to really strong data, Dallas' average win total in my projections is 10.7, and they hit the over 74.5% of the time. Because they finished second in the NFC East, they benefit from a softer schedule than they probably deserve (more on that when we talk about the Philadelphia Eagles).
They did lose stud cornerback Byron Jones to Miami and edge rusher Robert Quinn to Chicago but shored up Dak Prescott and Amari Cooper, a dominant duo in terms of efficiency. With a promising schedule and two poor teams in their division, Dallas is an early favorite of my simulations to notch double-digit wins.
Philadelphia Eagles (Under 9.5)
The Philadelphia Eagles have had one double-digit win season over the past five years, and while that doesn't really mean anything, it's always interesting to see how upcoming perceptions pan out compared to recent history.
In 2019, the Eagles "won" the NFC East with a 9-7 record. This is despite ranking 16th in numberFire's year-end power rankings with a nERD rating of -0.27, which indicates expected point differential against an average opponent on a neutral field. So, yeah, they were slightly below average in efficiency but happened to win nine games. Their Pythagorean win total was 8.8 games. I don't want to make it sound like they over performed significantly, but their division crown has earned them a tough schedule in 2020, insofar as we can project which teams will remain competitive this season.
Anyway, because they won the division last year, they have a tougher draw than a roughly .500 team typically gets, and when we compare them to the 8-8 Dallas Cowboys, there's a pretty clear discrepancy in schedule. Philadelphia hosts the New Orleans Saints, who project to be one of the NFL's toughest outs, and face the Green Bay Packers on the road. They're also at the San Francisco 49ers and Pittsburgh Steelers.
By contrast, the Cowboys don't have the Saints or Packers but instead the Los Angeles Rams and Minnesota Vikings. Even with four nearly free wins over Washington and the New York Giants, the Eagles hit the under roughly 70% of the time in my simulated seasons. We shouldn't chase the record last year and instead expect negative regression for Philadelphia. Even a nine-win season won't get them the over.
Jacksonville Jaguars (Over 5.0 Wins)
The Jacksonville Jaguars have a low win total of just 5.0, but they averaged 6.7 wins in my 10,000 simulations and hit the over on 5.0 wins in 75.9% of the outcomes. Why? Well, it's not because they're a particularly great pre-draft team. They rank only 27th in expected point differential, which translates to an expected win percentage that puts them on pace for 6.1 wins. That's pretty rough, but even that is also above the 5.0 wins needed for the over.
Jacksonville also has just the 22nd-toughest schedule based on my 2020 team estimations when I factor in expected steps back from the Tennessee Titans, who lost top-tier tackle Jack Conklin and should face regression from Ryan Tannehill's breakout season, and the Houston Texans, who got rid of DeAndre Hopkins.
The main caveat here is that Gardner Minshew wasn't good in 2019, as he ranked 29th in opponent-adjusted efficiency based on numberFire's metrics. However, Jacksonville has two first-round picks and a lot of draft value and is once again in possibly the NFL's weakest division from top to bottom. Another six-win season will get us the over, and the simulations think that happens often enough to back the rebuilding Jaguars.