Risers and Fallers for 2016 NFL Season (02/18/16)

By John Ewing Director of Research and Analytics @johnewing
Three stats that can (help) predict the 2016 NFL season.

The Super Bowl just ended and the new league year doesn't even start until March, but it is never too early to look ahead to the 2016 season. The NFL Draft and free agency will help determine how teams play in the upcoming season (and we will use updated rosters, depth charts and schedules to get our final answers), yet here are three stats that can be useful in predicting which teams will improve and decline when football kicks off in the fall.

Pythagorean Wins

In the NFL, with only 16 games on the schedule, winning seems like everything. While an excellent record across a 16 game regular season can get a team in the playoffs, it can also be completely meaningless when it comes to predicting how a team will perform the next season. A better indicator of future success is point differential.

Football's Pythagorean Theorem is a formula that converts the points scored and allowed by a team into an expected winning percentage (with historically strong accuracy). For example, in 2014, Detroit went 11-5 to earn a Wild Card berth while scoring 321 points and allowing 282 points. Based on the Pythagorean formula, the Lions should have won 9.2 games. The boys from Motown outperformed their true level of talent by 1.8 games in 2014. This difference in expectations hinted at regression for the Lions in 2015.

A year later, Detroit went 7-9 and finished third in the NFC North. There were eight teams in 2014 that exceeded their Pythagorean expectations by a game or more. Those clubs combined records were 84-43-1 in the 2014 campaign. In 2015 they combined to 67-61. Six teams underperformed Pythagorean expectations by at least a win in 2014. Five of the six teams improved and by an average of two wins over their previous record in 2014.

What teams can we expect to improve and decline in 2016 based on Pythagorean Wins?

Likely Improve Pythagorean Differential Likely Decline Pythagorean Differential
San Diego -1.9 Minnesota +1.2
Seattle -1.8 San Francisco +1.2
Tennessee -1.8 Indianapolis +2.0
NY Giants -1.5 Denver +2.3
Dallas -1.2 Carolina +2.6

Point differential suggested that San Diego would decline in 2015 but falling to 4-12 was surprising. The nightmare season saw the Chargers go winless in the division. By the second month of the season, San Diego's starters, including three offensive linemen, had missed more than 20 games. Philip Rivers was able to overcome the uneven play up front, but the running game never got going. The numbers indicate that the Bolts should improve in 2016.

Another candidate for improvement is Seattle. This might seem unfair, as the Seahawks have had more than their fair share of success the last few seasons. Still their 10-win campaign in 2015 underperformed expectations by nearly two full wins. This is one of the reasons why the Seahawks were installed as the early favorites to win Super Bowl LI.

Which teams should we expect to decline? When you look at these five franchises, one could make a case for each to continue exceeding expectations. The Panthers have the reigning MVP, Denver will return many pieces of an all-time defense, Indianapolis will have a healthy Andrew Luck, Chip Kelly could reinvigorate Colin Kaepernick's career and Teddy Bridgewater and a young Vikings defense could take another step forward. Those are compelling arguments, but history tells us it is exceedingly difficult to outperform point differential year in and year out.

Record in Close Games

It is almost impossible to consistently win games decided by one touchdown or less. Teams that have a good or bad record in close games one season usually don't perform to the same level the following year.

In 2014, the Packers and Bengals combined to go a perfect 8-0-1 in one score games. In 2015, Green Bay and Cincinnati combined to go 6-6 in such games. Thirteen teams had winning records in games decided by seven points or fewer in 2014. In all, those teams combined to go 57-19-2, a 75 percent win rate in 2014. In 2015, those same teams went 59-45 in tight games, closer to a 57 percent success ratio.

Teams that perform badly in one score games are just as likely to regress to the mean (improve). The five teams (Jets, Raiders, Titans, Bears and Giants) with the worst records in close games in 2014 combined to go 6-23 (20.7%). In 2015, their win percentage nearly doubled (19-30, 39%) in close games.

What teams can we expect to improve and decline based on record in close games?

Likely Improve Record in Close Games Likely Decline Record in Close Games
Cleveland 1-5 Carolina 6-1
Dallas 2-6 Arizona 4-1
Tennessee 2-6 Denver 9-3
NY Giants 3-8 San Francisco 4-2
San Diego 3-8 Minnesota 4-2

Cleveland ended the year as the worst team in our Power Rankings, so seeing the Browns on this list isn't much of a surprise. The Browns were a dysfunctional squad with Johnny Manziel, Josh McCown and Austin Davis at quarterback. Cleveland has the second pick in the draft; maybe it will find a franchise quarterback that can help the team win the close ones.

Dallas, San Diego, Tennessee and the NY Giants are all teams that should improve based on point differential; they also make the cut because of lousy records in close games.

The same teams (Carolina, Denver, Minnesota and San Francisco) that exceeded expectations based on Pythagorean wins also had success in one score games. This continues the narrative that these teams are candidates for regression in 2016.

Turnover Differential

Like records in one score games, a team's turnover differential is rarely consistent from year to year. This is especially true of the teams on the extreme ends.

Here were the teams with the largest turnover margin in 2014.

Improve Turnover Differential Decline Turnover Differential
Oakland -15 Green Bay +14
New Orleans -13 New England +12
Washington -12 Houston +12
NY Jets -11 Seattle +9
Tennessee -10 Arizona +8

As we would expect, the majority of these teams regressed towards the mean. Four of the five worst teams actually had positive turnover differentials in 2015. The Packers, Patriots and Texans were at the top of the leaderboard with a combined +38 in the turnover department in 2014. They were not as successful in 2015 as their combined positive differential was cut by more than half.

What teams can we expect to improve and decline based on turnover differential?

Likely Improve Turnover Differential Likely Decline Turnover Differential
Dallas -22 Carolina +20
Baltimore -14 Kansas City +14
Tennessee -14 Cincinnati +11
Jacksonville -10 Arizona +9
Cleveland -9 New England +7

Dallas, Tennessee and Cleveland, all teams that have already popped at least once before in this article, likely won't be as unlucky with the bounce of the ball in 2016. Carolina was the best in the NFL at forcing interceptions and recovering fumbles, but it's doubtful the Panthers will be quite as good as they were a year ago.

These stats won't predict the future, but they do give us an idea of what to expect from teams going forward. Tennessee and Dallas are two clubs that will likely improve in 2016 after underperforming point differential, having bad records in close games and poor luck with turnovers.

Carolina was a surprise team of 2015. The Panthers rode excellent play from Cam Newton and an opportunistic defense to the Super Bowl. Just about everything went right for Carolina, and it has to for any team to reach the big game. However, these indicators of future success suggest that the Panthers will regress.