How turnover differentials can predict the future
In Week 6 of the 2011 NFL season, Drew Brees and the New Orleans Saints traveled to Tampa Bay as 6-point favorites for a showdown with Josh Freeman and the Buccaneers. In an interesting turn of events, the Bucs defeated the Saints 26-20 in what would be Tampa’s last victory of 2011. Meanwhile, New Orleans would go on to finish the regular season with a 13-3 record and a league-best 12-4 mark against the spread.
So what happened to the Saints on October 16?
Despite finishing the season with a league-worst -16 turnover differential, the Buccaneers beat the Saints 26-20 in Week 6 last year.
You can point to several reasons for New Orleans’ somewhat surprising defeat, but what stands out the most from this game is the fact that Tampa Bay—who would go on to finish the year with the worst turnover differential in the NFL at -16—forced four New Orleans turnovers. Brees threw three interceptions, Pierre Thomas lost a fumble and the Buccaneers didn’t commit a single turnover the entire day.
And those who took the points with Tampa Bay looked like geniuses.
As all of you are already well aware, turnovers play an integral role in the successes and failures of NFL teams each and every week. And while it may seem as if these sudden changes of possession are just random, completely unpredictable occurrences, the truth of the matter is that there is some historical data we can use to get an idea of which teams could be headed for big things in the turnover department this season.
Justin Zovas wrote a great piece over at our friend Todd Fuhrman’s website ToddsTake.com a few weeks back and we’ve decided to expand on the research. But before we dive into which NFL teams could be worth an investment for the 2012 campaign, let’s first take a look at just how big an impact a significant turnover differential can have on a team’s season.
BEST & WORST TURNOVER DIFFERENTIALS BY SEASON (2002-2011)
Well how about that? The 11 teams that posted the best turnover differentials by season since 2002 combined to post a straight-up record of 133-43 (.755) and an ATS mark of 106-65-5 (.619). To put it another way, if you had laid a flat $100 on every game played by each of those teams at 11/10 odds, you would have turned a profit of $3,450.
As for the teams on the other end of the turnover spectrum, the 13 organizations that posted the worst turnover differentials by season since 2002 combined to post a straight-up record of 64-144 (.307) and an ATS mark of 66-132-10 (.333). Betting a flat $100 on each game played by all of those teams at 11/10 odds would have cost you the hefty sum of $7,920.
So how does any of this help us to determine what’s going to shake down in 2012? For starters, if you look at what happened the year after each of the aforementioned teams either led or finished last in the NFL in turnover differential, you’ll notice that a drastic change took place.
Excluding the 2011 49ers and Buccaneers, 21 of the 22 teams listed in the above chart experienced a turnover differential adjustment of +/- 11 or higher in the opposite direction the season after posting such extreme results.
Tony Sparano's Dolphins crashed back down to earth with a turnover differential of -8 in 2009 after posting a +17 in 2008.
For example, after the Miami Dolphins posted a turnover differential of +17 in 2008, the team came back in 2009 to record a turnover differential of -8 for a swing of 25 turnovers. Check out the chart below and you’ll notice the same thing took place for every team on the list. The only team that didn’t experience a swing of +/- 11 turnovers was the 2005 Indianapolis Colts, who went from +19 to +12.
In other words, teams that record extreme highs or lows in turnover differential one season will regress towards the mean the following season. In this case, the 22 aforementioned teams averaged a change in turnover differential of 20.31 the year after either leading or finishing last in the NFL in turnover differential.
Think about what we could accomplish by being able to predict when these significant turnover differential swings will take place?
Let’s take this study one step further and see what we can learn from a larger sample size. The following chart represents the turnover differentials recorded by each NFL team over the last ten years. Here’s what we’re looking for:
1. Teams that experience a change in turnover differential of +/- 20 or more from one year to the next.
2. Once we’ve identified those teams, our goal is to identify what happened in year 3.
NFL TURNOVER DIFFERENTIALS BY TEAM FROM 2002-2011
Here’s what the research tells us:
1. From 2002-2011, there were 34 instances in which an NFL team experienced an increase or decrease in turnover differential of 20 or more from one year to the next.
2. In 30 of those instances (88.2%), the teams went in the opposite direction in turnover differential in year 3. What’s even more shocking is that those 30 teams averaged a change in turnover differential of +/- 15.2 from year 2 to year 3. That’s an average of almost one more or fewer turnovers forced or committed per game.
3. The four teams that did not revert back in year three were the 2005 New York Giants, 2007 San Diego Chargers, 2010 Miami Dolphins and 2011 Detroit Lions.
Now we get to the good stuff…
1. Of the 30 teams that reverted back in turnover differential in year 3, 14 of those teams trended upwards (more favorable turnover differential) while 16 teams trended downwards (less favorable turnover differential).
2. The 14 teams that trended upwards posted a straight-up record of 84-140 (.375) and an ATS record of 85-133-6 (.390) in year 2.
3. After bouncing back in the turnover differential department in year 3, those same 14 teams posted a straight-up record of 126-98 (.563) and an ATS record of 120-100-4 (.550). That’s good for an average of 3.0 more straight-up wins and 2.5 more ATS wins in year 3 than in year 2.
4. The 16 teams that trended downwards posted a straight-up record of 170-86 (.664) and an ATS record of 156-91-9 (.632) in year 2.
5. After regressing in the turnover differential department in year 3, those same 16 teams posted a straight-up record of 122-134 (.477) and an ATS record of 108-145-3 (.427). That’s good for an average of 3.0 less straight-up and ATS wins in year 3 than in year 2.
For a complete breakdown of the straight-up and ATS swings that took place for each of the 30 aforementioned teams from year 2 to year 3, check out the chart at the end of this story.
SO HOW DOES THIS INFORMATION HELP US?
Ok, so we’ve established that teams that record a significantly high t urnover differential will also usually experience a high amount of straight-up and ATS wins, while teams that record a significantly low turnover differential will usually post a low amount straight-up and ATS wins.
In addition, we’ve also learned that in 88.2% of the 34 instances since 2002 in which a team experienced an increase or decrease in turnover differential of 20 or more from one year to the next, those teams swung in the opposite direction at an average of +/- 15.2 in the turnover differential department in year 3.
Take a look at the chart above and you’ll notice five teams (in red) that experienced a swing in turnover differential of 20 or more from 2010-2011. In addition, four of those teams finished the 2011 campaign at one of the two extreme ends of the turnover differential spectrum.
2010: TO Diff of +17, 12-4 SU, 10-6 ATS
2011: TO Diff of -13, 12-4 SU, 7-9 ATS
No team in the NFL experienced a bigger shift in turnover differential last season than the Pittsburgh Steelers, who went -30 from 2010 to 2011. Mike Tomlin’s crew posted the same amount of SU victories in both ’10 and ’11, but the Steelers covered the number in three fewer games last season. If these guys could win 12 games with a -13 turnover differential, imagine what happens in 2012 if they find themselves on the plus side.
San Francisco 49ers
2010: TO Diff of -1, 6-10 SU, 7-9 ATS
2011: TO Diff of +28, 13-3 SU, 11-4-1 ATS
History tells us Alex Smith and the Niners won't be able to recreate last season's success in the turnover department.
Take note that this is now the second time the 49ers have appeared in one of our articles discussing teams headed for a step in the wrong direction in 2012. The hiring of head coach Jim Harbaugh produced fantastic results last season, but history shows us it will be damn near impossible for the Niners to maintain in the turnover differential department in 2012. Exercise caution when looking to make any futures bets on this team.
Tampa Bay Buccaneers
2010: TO Diff of +9, 10-6 SU, 9-5-2 ATS
2011: TO Diff of -16, 4-12 SU, 4-12 ATS
A league-worst -16 turnover differential led to a miserable 4-12 season that resulted in a complete overhaul of the coaching staff. In a division in which the Buccaneers are staring up at the Panthers, Saints and Falcons, there aren’t many people out there predicting big things for Tampa Bay in 2012. That means this team is likely to be undervalued, which would almost certainly result in more covers against the number.
2010: TO Diff of +9, 10-6 SU, 8-8 ATS
2011: TO Diff of -14, 8-8 SU, 8-8 ATS
Credit a late season run of four consecutive wins and covers for salvaging what was a disaster of a campaign for the Eagles. With an improved defense and all the key offensive skill players happy and under contract, things are looking up for the Birds in 2012. Of course, Michael Vick’s ability to stay healthy will likely be the difference between a postseason run and a coaching change.
2010: TO Diff of -15, 8-8 SU, 9-7 ATS
2011: TO Diff of +5, 5-11 SU, 7-8-1 ATS
Despite producing a much more favorable turnover differential in 2011, the Jaguars took a step back in both the SU and ATS departments last season. Much of that can be attributed to a change at the quarterback position in which inexperienced rookie Blaine Gabbert was given the keys to a car that didn’t work all that well in the first place. Also take note that the Jaguars are the only team on this list that wasn’t at one of the extreme ends of the turnover differential spectrum last season, meaning it wouldn’t be surprising to see these guys hover around the mean for another year.
Hit me up on Twitter: @JoeFortenbaugh