Started From The Bottom

In a sample space as minuscule as sixteen NFL games, there are bound to be statistical outliers--teams that were unusually lucky or unlucky by certain measures over the course of the season. It is the unavoidable curse of football. No matter how good the players are or how well they are coached, the team is at the mercy of the bouncing will of an oblong sphere. 

These teams often find that this remarkable luck, good or bad, is often unsustainable due to the statistical tendency of regression to the mean. Over a large enough sample, statistics subject to randomness (e.g. fumble recovery, record in close games, etc.) approach an even 50% rate. In this article, we will take a look at teams that significantly under-performed in these areas of randomness, likely to see their luck improve. For some of them, normalized performance in these areas might mean the difference between a winning or losing record, or even making or missing the playoffs.

Record in Games Decided by 7 Points or Fewer

Teams' records in one-score games gravitate towards .500, so any squad that loses an inordinate number of them might hope to get some of those wins back the following season. Perhaps outside of having an excellent time-wasting run game, luck and intangibles like "grittiness" are the clearest deciding factors in one-possession games. Below are the NFL teams who had a winning percentage below .400 in such games:

TeamRecord (W/L)Win Pct
New Orleans Saints(3-5).375
Jacksonville Jaguars(1-2).333
Baltimore Ravens(2-4).333
Houston Texans(2-4).333
Atlanta Falcons(2-4).333
New York Jets(3-6).333
Oakland Raiders(2-5).286
Tennessee Titans(1-4).200
Tampa Bay Buccaneers(1-8).111
New York Giants(0-3).000
Chicago Bears(0-4).000

Most of these teams could have really used the one or two wins that would have brought them to .500 in one-score games. Had the Texans closed out at 3-3 instead of 2-4, they would have finished 10-6, a wild-card-worthy record. The same goes for the Saints: if they had pulled out the victory in just one of those five losses, they would have finished 8-8 on the season, good enough for the division crown in the putrid 2014 NFC South. 

Speaking of the NFC South--my goodness, Bucs. Of Tampa Bay's 16 games, nine came down to a single possession. This reveals two things about the Buccaneers: (1) their fans are probably still in pain from having their hearts broken so many times, and (2) their league-worst 2-14 record is probably far from representative of their true talent level. Regardless of how well QB Jameis Winston plays in his rookie year, Tampa is a big candidate to improve, simply because they will not fare as poorly in close games next year.

While the Buccaneers are the most dramatic case, if you see your favorite team in that table, you might have an extra couple of wins coming your way in 2015.

Fumble Recovery Percentage

Even the most agile, observant player cannot jump on a loose ball if it lands in an opponent's lap. Fumble recovery almost entirely comes down to luck, which creates an interesting dynamic given how impactful a turnover can be during a game. Again, since the chances of recovering a fumble should hover around 50%, the teams grabbing loose balls at a significantly lower rate could see fewer turnovers, more takeaways, and maybe more wins.

TeamRecovery Pct
Cleveland Browns 44.74
Chicago Bears44.19
Oakland Raiders42.55
Pittsburgh Steelers42.50
New Orleans Saints37.84
New York Giants36.96
San Francisco 49ers32.43

Unlike with one-score games, which might lend themselves somewhat to ability (just ask Andrew Luck and his career 18-4 regular season record in those games), players have virtually no control over whether they are the ones at the bottom of the pile. Therefore, there is good news around the corner for teams in this table, especially the bottom three.

Strength of Schedule

It is not quite accurate to call a difficult slate of games "unlucky" in the same way as losing a fumble is unlucky, but analyzing strength of schedule in 2014 (and projected SOS for 2015) gives us another metric to predict improvement. Below is the more difficult half of schedules; each of these teams faced opponents with a collective winning percentage above .500.

TeamOpp Win Pct
Tennessee Titans.506
San Diego Chargers.512
Kansas City Chiefs.512
Miami Dolphins.512
New York Giants.512
Jacksonville Jaguars.514
New England Patriots.514
Buffalo Bills.516
Denver Broncos.521
Arizona Cardinals.523
Seattle Seahawks.525
San Francisco 49ers.527
Chicago Bears.529
St. Louis Rams.531
New York Jets.543
Oakland Raiders.570

Obviously, this is not quite enough information. These teams had tough roads, but not all will get a reprieve in 2015. All the teams from the NFC West appear on this list by virtue of playing in a tough division, which does not change, and they play the NFC North and AFC North next season, each with at least two teams that will give them some trouble. Eight of these sixteen teams will see a significantly easier projected schedule (sorted from smallest to largest difference in SOS):

Team2014 Opp Win Pct2015 Opp Win Pct
Miami Dolphins.512.492
Oakland Raiders.570.545
Buffalo Bills.516.486
New York Giants.512.478
New England Patriots.514.477
Jacksonville Jaguars.514.463
New York Jets.543.488
Tennessee Titans.506.435

Each of these differences is significant but will clearly help some teams more than others. The Raiders, for example, still have an opposing winning percentage over .500, going from the hardest schedule in the league to the 7th-hardest -- not exactly cause for celebration. The Patriots, meanwhile, face an even easier schedule one year after going 12-4 and winning the Super Bowl. Square up AFC East!

Tennessee once again find themselves on the right side of these stats; they did not have an exceedingly difficult schedule in 2014 but now get the 7th-easiest projected slate this season by virtue of playing the Jaguars twice, the Browns, the AFC East, and the NFC South. You cou ld not do much better custom designing a schedule for a rookie QB.

Estimated and Pythagorean Wins

The last big predictive stat we will use here tries to more accurately understand a team's talent and how many wins they would merit under normal circumstances. If either of these measures indicates a win total higher than a team's actual wins, then we know that squad to be better than their record would suggest. Pythagorean wins is a relatively simple calculation, judging an appropriate number of wins based on point differential. Estimated wins (from "emphasizes consistency as well as DVOA in the most important specific situations: red zone defense, first quarter offense, and performance in the second half when the score is close. It then projects a number of wins adjusted to a league-average schedule and a league-average rate of recovering fumbles." I limited the teams in this table to those with six wins or fewer, or those with the most room for improvement--that is the cutoff where you really begin to see some disparity between real wins and estimated/Pythagorean wins.

TeamActual WinsEst. WinsPyth. Wins
Atlanta Falcons67.27.1
New York Giants67.07.5
St. Louis Rams
Chicago Bears56.44.9
New York Jets45.94.8
Washington Redskins44.44.5
Jacksonville Jaguars33.33.6
Oakland Raiders34.83.1
Tampa Bay Buccaneers24.14.4
Tennessee Titans24.03.3

Notable disparities between estimated and Pythagorean wins--the case with the Jets and the Bears--make prediction difficult. Both agree that Gang Green was better than 4-12; while they had the score differential of a 4.8-win team, estimated wins puts them closer to six wins. The Bears, meanwhile, did slightly better than their Pythagorean expectation.

On the more straightforward side, 2015 might hold some good tidings in the form of regression to the mean for teams like the Falcons, Giants, and Titans, each of whom under-performed their win total by at least a win by both stats--the Bucs underachieved by at least two wins in each. 

If at any point in this article you saw your favorite team, congratulations! However, it seems some teams were particularly unlucky, and in several areas, they are. In no particular order, the preseason picks for Most Improved: Atlanta Falcons, New York Giants, New York Jets, Tampa Bay Buccaneers, and Tennessee Titans.

Wish them and all other teams luck in the 2015 season; everyone needs some of it.

Upcoming Games

Aug 5th, 8:00 PM




Aug 12th, 7:30 PM


New England


Aug 12th, 7:30 PM