BGA Weekly: How Unlucky Can You Get?

Over the course of the offseason, I’ll be writing a weekly analytical piece, looking at statistical data from the 2011 season and revisiting some of my analysis from 2010 to see if any trends identified are still holding, or whether I can identify any new ones. As always, I would appreciate any suggestions for future BGA pieces.

How do you begin to write an analytical article about luck? If there’s one thing you’d think was impossible to represent statistically, luck would be it. (Unless, exceptionally, you’re talking about Andrew Luck). While there are some sabermetric stats in baseball that attempt to account for or even quantify luck, football doesn’t lend itself to a pure numerical analysis in quite the same way that the purity of baseball allows.

There are lots of misconceptions about luck, anyway. If you were flipping a coin ten times, every possible sequence has an equal chance of happening, whether or not it formed a uniform pattern or the results did not represent anywhere near the expected 50:50 outcome. The same could perhaps apply if there was something related to the NFL that had a 50:50 chance of happening and was unaffected by external factors.

Is there anything significant that is as random as a coin flip though? Probably not. Therefore, my approach in this article will be to consider some areas where the Jets could be considered to have been unlucky last year, figure out what factors affected the results and try to predict whether the Jets will have better “luck” in 2012. More after the jump…

Injuries

Injuries happen. That’s a fact of life in the NFL. However, last season, the Jets offensive line – which had been blessed with comparative good health over the previous few years – seemed to suffer some significant losses that had a major impact on their season. Let’s just focus on the offensive line, then.

Just how “unlucky” were the Jets in terms of injuries on the line? Nick Mangold was the only starter to miss significant time and he only ultimately missed two-and-a-half games. That’s fewer games than Damien Woody missed in 2010. Having said that, they also lost a key reserve in Robert Turner (for the whole season) and Mangold’s injury had longer-lasting issues in terms of his effectiveness over the rest of the year. That’s where it’s difficult to just evaluate how much of an impact injuries have had, because while your starter might not miss time, they might still be slowed by an injury, as Matt Slauson was towards the end of the year and Brandon Moore was at the start. Again, that is the case every year – Slauson had a bad knee and Moore needed hip surgery during the 2010 season.

Football Outsiders had been pointing out for some time how fortunate the Jets had been to not see any starters on their offensive line miss any significant time from the beginning of the 2008 season to late in the 2010 season and were convinced that this would “regress to the mean”. However, that doesn’t take into account whether the players on the line are capable of playing while hurt (or even too stubborn to take themselves out of the lineup when they should). Maybe a team’s medical staff has influence here too – and the attitudes of different teams’ staffs could vary wildly across the league.

Maybe the Jets deliberately targeted linemen who were durable. Mangold and Ferguson didn’t miss any starts in college and Brandon Moore hasn’t missed a start since 2004, for example. That could mean that you’d expect the Jets to get fewer injuries on their line than the average team. Sometimes injuries are virtually inevitable – would anyone be backing a team that decided to build around Chad Pennington, Darren McFadden and John Abraham?- but the nature of injuries is that you never know when they will strike.

As Mangold’s ankle sprain showed, a freak injury can happen to any player at any time. You could say the Jets were particularly unlucky that the injuries they did suffer on the line were both by players that play the same position early in the year at a time when the backup up at that position was not ready, just as they were due to head off on their toughest three game stretch of the season. In addition, neither of their backup guards were ready to play at that time, so moving Matt Slauson to center wasn’t an option either. However “unlucky” they were, I don’t think you can use this to forecast what will happen going forward.

They’ll get some injuries on the line this year, no doubt. In some cases, the linemen will probably play through the pain as they’ve done in the past and the cohesion of the line will probably benefit from that, although they might be shortening their careers to some extent. I’m sure the Jets have a backup plan for each position (and contingency plans in place for multiple injuries). They probably learnt a lesson or two from last year, as well.

Penalties

Penalties are a good example of something where almost no luck is involved in any individual call, but you can still feel slighted if all the calls seem to go against you. However, some players are more physical than others, some star players get away with calls on certain occasions and some officiating crews appear to be either biased or incompetent, if not both. If you end up on the wrong end of a combination of those factors, you might find penalties costing you the game on any given Sunday.

There were plenty of instances last season where I felt the Jets were on the wrong end of a bad call (or calls), but this is always an area where it’s difficult to separate personal biases from objectivity. I’m sure fans of just about every team felt the same way at some point last year (and every year).

For their own part, the Jets have constantly been taking dramatic measures to try and remove any chance element from the equation. They scout officiating tendencies and instruct their practice referees to call the practice in that way. Since 2001, they’ve been the least penalized team in the NFL off the back of this.

Last year was a down year, though, despite a penalty-free opener against the Cowboys. Hopefully some of the penalties on offense, especially pre-snap penalties, be eliminated just due to the relative simplicity of the offensive scheme. On defense, they already made improvements over the second half of last year, with Antonio Cromartie in particular (seven penalties, all over the first eight games) managing to make strides.

Fumble Recoveries

Fumble recoveries are particularly interesting because they are, according to Football Outsiders, entirely based on luck:

There is no correlation whatsoever between the percentage of fumbles recovered by a team in one year and the percentage they recover in the next year. The odds of recovery are based solely on the type of play involved, not the teams or any of their players.

So, a team can have excellent “fumble luck” one year and bad “fumble luck” the following year (or vice versa) but it’s not thought to have anything to do with how you play the game. You might think that defenses that swarm to the ball would have more chance of recovering a fumble, but there’s no evidence to support this. (I do wonder if an offensive lineman who finishes his blocks and plays to the whistle is less likely to recover a fumble than someone who gets beaten cleanly and is able to get an unobstructed view of the ball flying loose though!)

The Jets had seen a lot of balls bounce their way in recent years. Some attributed this to luck, whereas the more optimistic might have argued that the players were playing hard for their head coach, so they’d be more apt to hustle after a loose ball than someone less motivated. Football Outsiders would insist that it was just luck. That being the case, the luck went against them in 2011 (although FO would probably attribute this to “regression to the mean” as well).

Their 16 lost fumbles would have topped the league if not for the Denver Broncos who ran the ball a higher percentage of the time than anyone else. (And let’s not overlook the fact that one of Denver’s major culprits is now a Jet). More jarring was the fact that Mark Sanchez lost eight fumbles after having only lost one in 2010.

You might assume that was due to the fact that Sanchez fumbled the ball a lot more in 2011, but this isn’t the case. In fact, how often Sanchez has fumbled the ball has been an area where he’s shown rare consistency. 10 times in 2009, nine times in 2010 and 10 again in 2011. Looking closer at the splits shows that he was anything but consistent though, as fumbling is an area where he’s had lapses and then focused on protecting the ball at various times. Last year, he only fumbled once between week 4 and week 14 and, in 2009, he fumbled just twice after week 8 (including three postseason games).

Going back to that original quote by FO, they say that the odds of recovery are based “on the type of play involved” and here we can see some interesting differences. Of his 29 fumbles, 12 have been on running plays. There’s a clear progression here, too. Two in 2009, four in 2010 and six in 2011. That also means that he’s fumbled less each year when being sacked (eight in 2009, five in 2010, four in 2011). Also remarkable: He lost all six of his running play fumbles in 2011 and NONE of his six in 2009 and 2010.

One word of caution: When FO analyse fumble luck, they exclude fumbled snaps from the data. Official NFL numbers do not, so these are included in the running plays. There were three of these in 2011 and I’m sure there were a few in those 2009 and 2010 numbers. Usually, I’d imagine the chance of recovering a fumbled snap is pretty high, but Sanchez still went 0-for-3 in 2011.

I can see what FO means by the type of play affecting whether or not a fumble will be recovered. If a QB drops back to throw and then tries to scramble and loses the ball, all his linemen could be behind him and all his eligible receivers could be downfield. In any case, it seems like there will be more defensive than offensive players on hand to corral the loose ball.

It’s also interesting to note that on the four plays where Sanchez fumbled after having been sacked, two of these saw D’Brickashaw Ferguson get beaten for the sack that caused the fumble. Of the others, one came as he was hit by an unblocked Ed Reed on the blindside and one (as well as two of the running plays that saw him fumble) came when he scrambled and was tackled out of the pocket. So, as surprising as it may be, Wayne Hunter was not directly responsible for any of Sanchez’s fumbles.

So, the Jets recovered a lower percentage of Sanchez’s fumbles than you’d expect, but it’s not hard to see why. It shouldn’t enable us to predict future performance any better than when the Jets were among the league leaders a year or two ago. If they recover half of their fumbles next year, which is statistically probable, then that would be an immediate improvement.

Field Goals

Here are the statistics for field goal kickers against the Jets last year:

29 attempts, 28 made

That’s remarkable…especially when the only miss was a 56-yarder by Sebastien Janikowski. Let’s dig deeper into the numbers:

82.9% of field goals were made last year in the NFL. So, had the kickers performed to that standard, only 24 would have been made. Could those 12 points have helped a team that lost one game by four and one by two points? Maybe. It’s not like they were all chip shots either. Kickers were 7-for-7 from 40-49 yards and 6-for-7 from over 50 yards. That’s 85.7% from over 50 yards when the league average from that distance was 64.5%. Kickers were 22 for 22 from inside 50 yards (with the average around the league rising to 86%).

But was it unlucky?

Clearly when a kicker attempts a field goal, there is more involved than luck from the kicking team’s point of view. Generally when they miss one, it’s because someone screwed up, not because of bad luck. From the other team’s perspective, however, you have very little control over whether or not the kick is good. Sure, some teams are better than others in terms of putting pressure on the kicker. For example, Seattle Seahawks defensive tackle Red Bryant blocked four kicks last year and I’m sure that is reflected negatively in the made field goal figures for their opposition. However, when teams sometimes miss field goals because they screw up, but nobody ever screws up against you, you could be forgiven for thinking the luck is against you.

Let’s say for a minute that it was unlucky. How unlucky?

If we ignore all outside factors and make the leap of faith that there is an 86% chance of ANY field goal less than 50 yards being good. What was the probability of NFL kickers going 22 for 22 against the Jets? I believe it’s 0.86 to the power of 22 … which gives us a one in 27.6 chance. Unlikely but not statistically improbable. Had Janikowski made that 56-yarder giving NFL kickers a 29-for-29 overall record against the Jets, that would have been a lot more improbable: less than a one-in-222 chance. Then again, that didn’t happen.

Conclusions

It creates a paradox any time you try to statistically evaluate luck within something as complex as a football season, because it’s very rare that you find a circumstance uninfluenced by outside factors. However, even if we make a few assumptions, it seems pretty clear that the perceived bad luck suffered by the Jets last year wasn’t that unusual.

There are some things they can do to reduce their exposure to the risk of these things repeating themselves (which in itself is an indication that it didn’t all come down to luck). Luck itself should not carry over from one year to the next, so the good news is that those areas where last year’s performance fell well short of the expected or statistically probable outcome aren’t likely to go as badly again. Dare I say they might even regress to the mean?

You need a lot of luck to win a Super Bowl, as the Giants proved in their improbable run last Winter. However, you need more than just luck – and some would say you make your own luck anyway.