revolutionary tools.  groundbreaking articles.  proven results.

Using Sabermetrics to Project 2018 NFL Team Wins

Earlier this week I looked at which teams got lucky in 2017 from a sabermetrics perspective. Using each team’s Pythagorean win percentage, we can use their 2018 schedules to give us an estimate for how they might perform.


Utilizing the Log5 formula in conjunction with each team’s Pythagorean win percentage from the prior year, we can estimate each team’s expected wins for the upcoming season. The Log5 formula formula states that if Team A has a Pythagorean win percentage of P­and Team B of P­B, then the probability of Team A winning (denoted W­A) is as follows:

A = (P­A – P­A * P­B) / [P­A+P­B – (2 * P­A * P­B)]

Note that these win percentages indicate how the team should play at a neutral site. Additionally, these projections are strictly a formulaic approach to calculating wins. The formula knows nothing regarding player development, personnel changes, coaching changes, or scheme changes that affect the true win percentage of a team.

However, given that there have to be 512 games every year, NFL wins are a net zero-sum game.1 Since we know that there have to be 256 wins every year, the Pythagorean wins reflect the true averages that we expect from each team in the long run. In other words, assuming we could simulate the offseason changes multiple times over, we should expect that the long-term winning percentages for each NFL team mirrors its Pythagorean expectation.

To find Pythagorean Win percentages from 2017, check out this link. Let’s dive into the 2018 projected team wins.



A few things stand out from the table.

  1. The Green Bay Packers, Indianapolis Colts, San Francisco 49ers, and Houston Texans could all ostensibly outperform their projected 2018 wins. In the five games where Aaron Rodgers was healthy the Packers went 4-1, and their average winning point margin was a robust nine points. Once Brett Hundley took over, the Packers suffered several blowout losses. Since Rodgers took over as the starter in 2008, the Packers have routinely won 10 or more games.2 Note that we have seen a similar situation play out in the past. Rodgers missed the bulk of the 2013 season with a collarbone injury. In 2014 the Packers went 12-4, before losing to the Seattle Seahawks in the NFC Conference Championship game. With Rodgers poised to return fully healthy for 2018, it wouldn’t surprise me if the Packers outperformed their paltry 5.9 win projection.
  2. As mentioned earlier, the Pythagorean formula does not take into account improvement in personnel. The 49ers best represent the limitations of the Pythagorean formula. With Brian Hoyer/C.J. Beathard under center, the 49ers lost 10 games. Although they lost five games by a field goal or less, they were also blown out by 20, 30, 23, 10, and 11 points in the other five. However, once Jimmy Garoppolo took over, the team ended the season on a five game win-streak. The team posted convincing wins against the Titans and Jaguars, both of whom were jockeying for playoff spots.3 If Garoppolo is the stud QB we saw over those five weeks, the 49ers should outperform their Pythagorean projection.
  3. Once Deshaun Watson is fully healthy, the Texans should be able to take advantage of the soft schedule. A fully healthy Watson should be able to support both DeAndre Hopkins and Will Fuller. It’s easy to envision Hopkins finishing as a top-3 WR, and Fuller could potentially crack the top-24 WRs in PPR scoring. Additionally, the favorable schedule should be good news for Lamar Miller’s fantasy stock. Assuming D’Onta Foreman is unable to start the season fully healthy, and the team doesn’t draft a top-ranked rookie RB, Miller should exceed expectations.
  4. The Pittsburgh Steelers project for approximately 2.6 fewer wins than last year. Interestingly, the Baltimore Ravens project for approximately 2 more wins than they did last season. The battle for the AFC North will be one to keep an eye on.
  5. Given that the Ravens face one of the softest schedules in 2018, Alex Collins is in a great spot. If the Ravens don’t spent a ton of draft capital on RB, Collins could end up as the Ravens’ workhorse. Collins earned a larger role in the receiving game over the last 9 games, and he’s poised to build upon his successful 2017 campaign.
  6. As previously mentioned, the Los Angeles Chargers are prime improvement candidates. Currently set to face one of the weakest 2018 schedules, the Chargers’ offense should be targeted in all fantasy formats. Expect Keenan Allen and Melvin Gordon to pick up where they left off last season. Barring a surprise Antonio Gates signing, 2018 is shaping up to be Hunter Henry’s breakout season. Lastly, Mike Williams and Tyrell Williams are two players you should consider stashing in deeper leagues.
  7. Not only are the Cardinals and Bills prime regression candidates, they get a bit of a double whammy. The teams are expected to face two of the toughest schedules next season. Given the QB issues both teams currently face, I expect that they both underperform their projected number of wins. Although David Johnson, Larry Fitzgerald, and LeSean McCoy are stud players at their respective positions, expectations should be tempered. I anticipate that all three players will see an outsize workload in 2018. If their respective offenses struggle, the lack of red-zone opportunities for Johnson, Fitzgerald, and McCoy will prevent them from reaching their ceilings.
  8. The fate of the Indianapolis Colts rests entirely on Andrew Luck’s injured shoulder. Currently Luck is not throwing at veteran minicamp, and has not thrown a pass since January 1, 2017. Luck’s absence torpedoes the fantasy value of T.Y. Hilton, Eric Ebron, and Jack Doyle. Even if Luck returns to action this season, it’s fair to ask if he will be the same player we remember.
  9. The Browns are expected to improve by the most wins in 2018. Not only did the Browns underachieve last season, they also underachieved in 2016. Given that Hue Jackson has underperformed two seasons in a row, it’s fair to wonder whether the Browns can win three or more games. Maybe Jackson should purify himself in the waters of Lake Erie?

What else strikes you as interesting from the Pythagorean win projections from 2018? Let me know in the comments below.

  1. Ties count as half a win for each team.  (back)
  2. 2008 was Rodgers’ first season as the starter and the Packers went 6-10.  (back)
  3. Their Week 17 31-13 win against the Rams came against the backup players.  (back)

recent and related...

in case you missed it...

Quick Hits – Dare Ogunbowale, Tampa Bay Buccaneers

Maybe you’re a longstanding Ronald Jones truther. Perhaps you love the fact that Peyton Barber had a top-10 opportunity share among all NFL RBs last season and delivered only 9.4 fantasy points per game. If you are, or even if you’re not1 then it’s hard to get really excited about the

Read More

Why Buy Tarik Cohen When This RB is Cheaper?

“You pay for this, but they give you that.” That classic Neil Young lyric came to mind as I was researching this year’s “Why Buy?” series. We’re often so confident of what we’re going to get when we spend that sixth-round pick on the next big thing, but in reality,

Read More

Sign-up today for our free Premium Email subscription!

© 2019 RotoViz. All rights Reserved.