Tuesday, September 4, 2012

Detroit Lions 2012: Is Regression Coming?

The 2011 Detroit Lions season was something special. The Lions made the playoffs for the first time 1999, and only four years removed from that year. Matthew Stafford and Calvin Johnson broke all sorts of franchise records. The Lions won for the first time in Minnesota since 1997. They were featured on a Monday Night Football game after a ten year absence and made their debut on NBC's Sunday Night Football.

And the games. Man the games. Comebacks, blowouts, upsets, comebacks, more comebacks. And the plays. Man the plays. Bombs, pick-sixes, game-winning drives, blocked field goals, Calvin Johnson, Calvin Johnson and more Calvin Johnson. When creating the top 50 plays of the 2011 season, the trouble was not finding 50 moments, but rather cutting down to 50.

But now its 2012, and though the 2011 season will never be forgotten, we find ourselves in much different times now. The postseason is not the golden, shiny reward at the end of the journey, but the expectation. The Lions aren't the surprise contenders, but a decent team that is expected to play well on a weekly basis or they are a failure.

However, the common motif surrounding 2012 previews is that Lions were a good story last year and great for the city, but, ultimately, they had their time to shine and what lies ahead is the dire reality of mediocrity similar to the Lions of the 90s. The Lions don't rank in the top ten of most preseason power rankings, and are as low as 18 in some. Perhaps the most damning piece on the fate of the 2012 Lions is a study done by National Football Post, which found an interesting, but disturbing trend in the NFL.
Since 2002, there’s an 89.6% chance that a team who bounces back from a losing season to post ten or more wins the following fall is headed for a step in the wrong direction come year three.
The sample size of 29 teams is obviously a bit small, but the statistics are strikingly strong. Even more alarming is the 26 teams that "regressed" in the study lost an average of nearly four more games the following season. However, this study included teams that won more than ten games; teams that had more room to fall than the Lions.

I wanted to narrow this study a bit, so that it was more applicable to the Lions' situation. I looked at the past ten years and found every 10-6 team. Of the 33 teams that went 10-6, nine of them (27.3%) improved the next year, 22 regressed (66.7%), and two went 10-6 again the following season. On average, these 33 teams won 7.9 games the following season.

But this doesn't really capture the essence of the Lions. The Lions are a team that has been improving for the last few years, and these 10-6 teams in the study come from all varieties. So from those 33 ten-win teams, I picked out the teams that had done worse in the previous season and improved to 10 wins. I'll call these teams "improving teams."

Over the past ten year, there have been 24 improving 10-6 teams. Those improving teams continued to improve after their 10-6 season only 20.1% of the time (five of 24), averaging a total of just 7.4 wins the following season.

The numbers aren't quite as pessimistic as the ones in the linked study, but they don't exactly give Lions fans a reason to get excited. Still, it's clear that it is possible for teams to continue success. So instead of accepting our doom, let's examine the what factors contribute to sustained success versus ones that cause teams to fall back to irrelevancy.

The 2010 Tampa Bay Buccaneers is one of the most striking examples (and the most recent) of a team that saw extreme improvement, only to see it all dissipate the following season. The Bucs jumped from 3-13 all the way to 10-6 in 2010, just missing out on the playoffs. With a young team and an emerging quarterback, the Bucs were expected to again compete for the NFC South. But in reality, the team struggled drastically and finished a meager 4-12.

At the opposite end of the spectrum, the 2008 Minnesota Vikings jumped from 8-8 to 10-6, but didn't stop there. After adding Bret Favre, the Vikings finished the 2009 season with a record of 12-4 and made it to the conference championship (only to lose in an awesomely hilarious fashion).

On the surface, it appears the Lions most resemble the 2010 Bucs team: a young team looking for their players to improve the next season, rather than the Vikings, who seemingly improved by adding a much-needed veteran quarterback. However, looking at four key aspects of these teams, it becomes clear what makes an improving team better and what causes a team to stunt their growth.

Roster Changes
Stability is key to success in the NFL. A stable roster develops thorough knowledge of the playbook, creates chemistry between players, and makes a coach's job easier. After the Bucs' 10-6 season, Tampa Bay returned only 13 of 22 starters. [For clarification purposes, a "starter" is defined as the player who started the most games at that position throughout the season. So if a projected starter was injured and did not play the majority of games at his position that year, he is not considered a "starter" in this study] The Vikings, on the other hand, returned 16 of 22 starters.

The Lions are projected to start 21 of 22 starters from last year, with the only exception being cornerback Aaron Berry (although, it is very possible Mikel Leshoure replacing Kevin Smith will be another). If these 21 players can stay healthy for the majority of the season, the Lions' stability will only help maintain their success.

Roster Age
Age is a fickle thing in the NFL. Teams strive to keep their youth, but gather a team full of inexperienced youngsters, and you are doomed to inconsistency and mistakes. The 2010 Bucs starters were considerably young, averaging a birthday of 6/11/1984, or around 26 years old. In 2011, their average age was only two months older.

The 2009 Vikings, on the other hand, were around 29 years old on average, and were exactly four days older in 2010. So where do the Lions land? Last year, the Lions average birthday was 4/30/1984 (feel young yet?), or a bit older than 27. This year, with nearly the exact same lineup, they are unsurprisingly almost an exact year older. So the Lions are getting older, gaining experience and hoping to rid themselves of the inconsistencies and mistakes that plagued last year. Of course, it's worth noting that some of the Lions' best players are still young and therefore still prone to mistakes.

Quality of Wins (and losses)
In the NFL, the bottom line is always wins vs. losses. But rarely does that tell the entire story about a team. Did they blow out every team they beat and just barely lose against good teams? Was every game a nail-biter? Were there any warning signs that this team wasn't as good as their record implied? It is reasonable to conclude that if a team struggled mightily to get to ten wins, they may not be able to sustain their success next year.

With the Tampa Bay Bucs, you see exactly that. Though they won by an average of 9.3 points per victory, they only outscored their opponent by 23 total points at the end of the season. Also, the Bucs had a couple of weeks where there were clear warning signs. They were blown out by two different teams, losing both games by 25 points each. And five of their wins were by three points or less. Small margins of victory plus large margins of defeat (11.7 on average) equals danger.

The Vikings were in almost every game they played in 2008. Their biggest loss was by 13, and every other defeat was by seven or less. At the end of the season they had outscored their opponent by exactly double of what the Bucs did (46 for the math-impaired). Still, six of their wins were by seven points or less.

The Lions, as we all know, had some very close calls as well. They had five wins by seven points or less, but had the highest margin of victory of the three teams (Lions - 15.4, Bucs - 9.3, Vikings 8.7). They also outscored their opponent by 87, more than the Bucs and Vikings combined. However the Lions, too, had some warning signs. They lost by 24 to the Bears, 14 to the Saints and 12 to the Packers. They also fell behind big in games quite frequently and though they came back in most of those, that sort of thing is not sustainable for multiple seasons.

Overall, it's pretty clear the Lions were above the lower-level teams in the NFL, but their play against above average and elite teams was inconsistent at best. They competed with the NFL's best, but failed to record a win against a playoff team outside of the 8-8 Broncos.

Turnover Margin
Sometimes an oddity in record can be explained by plain luck. Turnover margin is often key to victories, but countless studies have proved its randomness. It is possible then, that a 10-6 record may be the result of mostly good fortune and that a regression is more likely if the team experienced a high turnover margin during their ten win season.

For the 2008 Vikings, this was not the case. They actually lost the turnover margin for the season, averaging 0.4 more giveaways than takeaways per game. They won games without the benefit of winning the turnover battle, a great sign for future success.

Looking at the 2010 Bucs, it's not too surprising to see that they ranked sixth in turnover margin, forcing 0.6 more takeaways than giveaways. Lo and behold, in 2011, when they ranked dead last in turnover margin (-16 for the season), they regressed considerably.

The 2011 Lions ranked fourth in turnover margin, also forcing 0.6 more turnovers than takeaways per game. This doesn't mean that the Lions are doomed next year. In fact, assuming that the Lions will lose the turnover battle in 2012 is the exact wrong way to interpret this data. That is akin to believing that a coin is more likely to land heads after landing tails five times in a row (the odds are still 50/50). The Lions are just as likely to win the turnover battle as they are to lose it in 2012. What this data does tell us is that the Lions were aided significantly by the luck of turnovers last year, and if they don't receive the same fortunes in 2012, they may not see the same success.

While history is stacked against the Lions, the 2011 team had many positive traits that may contribute to the teams' continued progression. Their roster is nearly the same as it was last year, giving the team stability and chemistry. Their players are also a good mix of young and old; players in their physical prime and veterans with valuable experience. Last year, they dominated a few teams and were in nearly every game they played.

But there are some warning signs as well. They Lions were clearly aided by turnovers last season, and without that benefit, they may not have reached the postseason. Also, they failed to beat a very good team last year, and were clearly over-matched against Green Bay and New Orleans.

In the end, I don't expect a huge regression. The 2010 Bucs season seemed completely random. Their jump from 3-13 to 10-6 was a wild jump in the standings largely aided by turnovers and winning games by the skin of their teeth. The Lions, however, have slowly and steadily been improving their team. The past three years have been an improvement from the season before, and because it has been a gradual improvement it is likely more sustainable than the wild seven-game improvement that the Bucs experienced. If the Lions are to regress this year, it will be no worse than an 8-8 season.

Special thanks to http://www.jt-sw.com/football/pro/ and http://www.teamrankings.com for their extensive historical data.


  1. "...12 to the Vikings..."
    Didn't think they lost to the Vikes.
    Good article overall; a thoughtful investigation into the numbers. As a scientist, I would love to see some prob + stats applied to see the confidence levels of these trends, but that doesn't take too much away from the article.

    1. Gah, should read Packers.

      And back in the day, I could've brought you some R-squared and significance levels, but its been a bit too long since my last statistics course. I've always meant to keep up with all that math stuff, but it's hard removed from college.

  2. Good analysis. Though I think there's 2 metrics in determining whether a team is prime for some regression - even if it is mild. Turnovers (which you rightfully mentioned above) and QBs. Elite or borderline elite QBs are pretty much immune from wild fluctuations. Most of the teams on the list NFP provides have had mediocre at best QBs. The exceptions being the '07 Eagles (McNabb) '06 Giants (Eli), '05 &'09 Falcons (Vick/Ryan), '05 Chargers (Brees) '11 Bears (Cutler). Though you could argue that none of those QBs were elite at the time. All had mitigating circumstances go against them.

    The '07 Eagles didn't run as well. Plus McNabb was ailing...the '06 squad was the fluke Jeff Garcia run after McNabb got hurt.

    The '06 Giants was Super Bowl hangover. And, as you mentioned above, certainly direct example of turnovers being a huge part of the equation Eli truly is maddening to figure out with his QB ratings in the mid 70s.

    The '05 Falcons had Vick get hurt which accounted for the dropoff and then in '09 Ryan in his 2nd year regressed a bit. Couple that with a huge drop in running production, that explains the mild regression.

    The '05 Chargers were actually a good team an quite befuddling. A good D, Brees having a great season and LT at the top of his game. BUT Turnovers and a slow start and poor finish hurt this team the most and sent Brees packing to NO.

    Speaking of NO, we all know what the Saints with Drew Brees did the season after Katrina ('06). It was a magical year. But that was a case of getting too good too soon. The defensive side of the ball was terrible.

    The '11 Bears had everything to do with injuries and little to do with an actual regression.

    The 3 that 'bucked the trend' as NFP put it 2 of those QBs (Rodgers, Roethlisberger) were elite. The Steelers buoyed by elite rushing offenses and defenses. The '05/'06 Bears were at their defensive apex.

    So that brings us to the '12 Lions. Will they take a step back? The biggest component is Matthew Stafford. If he regresses, the team will. Given the clear improvement Stafford made throughout last season it seems unlikely. Only a major injury can really derail an elite QB.

    As far as turnovers, it's entirely possible the Lions won't do as well. Especially when you consider the Lions recovered 13 of 17 forced fumbles. That's a high %. That being said, the Lions have built their entire Defense around getting after the quarterback. The 17 FF were the lowest in the Schwartz era and INT totals have increased every year as well. So while unpredictable, the Lions are seemingly doing something right.

    PS: Didn't mean for this to go on so long, but this is what happens when you're unemployed.

    1. Love it! Thanks for commenting. Really interesting find with the high fumble recovery percentage, didn't catch that.

      I think a lot of people predicting regression for the Lions are also predicting regression for Stafford. He caught many people's attention for his 5K yards, but I know some people over at Football Outsiders think he's being a bit overrated for being a good fantasy QB. Their argument being that for how many passes Stafford threw, he should've put up that many numbers. (just look at the back-and-forth about Stafford in this article: http://www.footballoutsiders.com/scramble/2012/scramble-ball-afcnfc-north-overunder )

  3. There are soooo many things wrong with that Football Outsiders article. I genuinely enjoy the myth of Calvin Johnson being the only reason Stafford is good. Without question he helps greatly. The same way Rice helped Montana/Young, Irvin helped Aikman, etc. Stafford passed for 3357 yards to guys not named Megatron. That would've been good for 16th in the league or approximately 250 yards behind Joe Flacco. Remember - there were a lot of people (not just Cris Carter) that were doubting Calvin was a Top 5 WR going into last season. Let's ask the classic AC/DC question: Who made who? Considering Calvin had averaged 1050yds/season the previous 2 years, you might have to say Stafford helped him quite a bit.

    I even agree with the argument that Stafford's numbers are inflated because he passed so much. As long as you're willing to do the same for Brees and Brady. But here's the thing: Average QBs don't pass for 35+ TDs. Yes, there have been a couple of flukes (Beurlein, Culpepper) but 41 TDs and a 6.2% TD Pass rate (4th in league) is impressive. Also, these stats could have been better if not for the Lions being 9th in dropped passes.

    And anyone, ANYONE who continues to make the claim that Stafford isn't on his way to being an elite QB just hasn't watched the Lions play a lot. Some of the throws he makes are downright jaw dropping. About the only throw he still has issue with is the short pass to the left flat/sideline. Anything else he's on. And only going to get better.

    I actually hope Stafford never passes the 5000yd mark again mostly because that means the offense is more balanced and the defense is much improved.