Football is the ultimate game of small sample sizes, with players who participate in a full season seeing only 16 chances to display their wares. Because of that, we see all sorts of player variance not only from game to game, but even over the course of multiple seasons.
Because of that, it can be difficult to always trust the numbers. While I whole-heartedly believe we can and should use analytics to acquire meaningful insights to help us win our leagues, sometimes we need to proceed with caution. That’s particularly true on the individual level, where there’s all kind of volatility that must be overcome to understand what’s really happening on the field.
That’s why I advocate analyzing players by type—categorizing them based on certain traits that predict success at each position—especially early in their careers. If we’re analyzing running back Lamar Miller, for example, I believe we can gain more knowledge from the fact that he’s a 224-pound back with 4.4 speed who was drafted in the fourth round out of Miami than we can from his career 4.2 YPC average. It’s not that his NFL stats are meaningless, but just that there’s not much of a resume with which to work, and what we do have is influenced so heavily by randomness that we can’t put much stock into it.
By grouping players into certain categories and then analyzing them individually with that foundation in place, I think we can overcome some of that randomness that accompanies small samples of data. I’m going to try to do that in a five-part series, analyzing quarterbacks, running backs, wide receivers, tight ends and kickers/defenses to show 1) which stats/traits are most important and 2) which player “prototypes” we should seek in fantasy drafts.
It starts today with the signal-callers.
Editor’s Note: Rotoworld’s partner FanDuel is hosting a $100,000 Fantasy Football Contest for Week 1’s games. It’s only $10 to join and first prize is $10,000. Starts Sunday, September 7th at 1pm ET. Here’s the link.
A Hefty Workload
Workload is such an important factor in every player’s fantasy projection that I considered not even mentioning it; without significant opportunities, no player, especially a quarterback, will be able to produce elite fantasy stats.
To show why this is the case, consider the distributions of quarterback efficiency (YPA) and workload (attempts) in 2013.
I charted YPA and passing attempts for the top 30 quarterbacks as a function of the No. 1 overall quarterback in each measure—Nick Foles with 9.1 YPA and Peyton Manning with 659 attempts. You can see that there isn’t nearly as much of a drop in YPA, which really levels out after the first few elite quarterbacks. Meanwhile, the drop in workload is much more linear. Quarterbacks who don’t surpass 80 percent of the top workload in the league just don’t have much of a shot at being an elite fantasy passer.
It’s worth mentioning that rushing attempts are certainly a major factor in quarterback workload, so they shouldn’t be ignored when discussing opportunities.
Age Is Just a Number
I’ve done a lot of work on how age affects fantasy production. Here’s a look at how the typical quarterback progresses throughout his career in terms of fantasy points per attempt.
Most passers peak in their late-20s in terms of efficiency, but they also generally see more opportunities later in their careers, which means that overall fantasy production basically evens out. Quarterbacks are the anti-running-back, capable of producing at an elite level at age 37 as much as age 27.
Note that quarterback efficiency has typically fallen off a cliff, though; it isn’t a steady decline. That suggests that instead of projecting a player like Peyton Manning for a gradual drop in production, it’s probably better to view him in terms of probabilities; as he ages, the probability that he just sort of “loses it” will increase each year. My guess is that when Manning is finished, we’re going to know it pretty quickly.
Mobile, Agile, Hostile
There are all sorts of ways to play quarterback in the NFL and I don’t think one is inherently “better” than another. But certain styles of quarterback—namely the runners—tend to come cheaper. For whatever reason, most fantasy owners tend to value passing attempts more than rushing attempts for quarterbacks. If you’re using a late-round quarterback strategy this year, there are probably worse combinations than Johnny Manziel, Geno Smith and Jake Locker, for example. Maybe running quarterbacks have a higher risk of getting injured, but you can effectively negate that by stocking up on them since they’re so cheap.
Also note that the general perception of mobile quarterbacks being volatile from week to week is probably wrong. Dating back to 2007, mobile quarterbacks (those with over 200 rushing yards in a season) have surpassed their average fantasy points per game 6.9 percent more often than pocket passers. The versatility that accompanies players like Cam Newton and RGIII allows them to record quality numbers even when they aren’t feeling it as passers, which actually makes them more consistent from game to game.
Big Hands, Big School, Big Game
Finally, we want to “go big or go home” at the quarterback position. I explained this in my first article here at RotoWorld—one that showed that hand size is a much better predictor of success than height for quarterbacks. The correlations between hand size and both completion rate and Pro Football Reference’s career approximate value are both much stronger than the same correlations with height.
I did a little more digging and found that big-handed quarterbacks are more accurate in cold weather than their small-handed counterparts. For the record, the average quarterback hand size is 9.6 inches.
It’s worth noting that quarterbacks with big hands also have significantly better touchdown-to-interception ratios in cold weather. The narrative is that quarterbacks need to be tall to see over the offensive line, which is probably perpetuated by the fact that tall quarterbacks do indeed have more success than short ones, on average.
However, there’s a really good chance that the tall quarterbacks succeed not because of their height but because they generally have large hands, which allows them to control the football with ease and deliver it with accuracy. That would explain why we’ve seen huge hands on the majority of short quarterbacks who have thrived, including Drew Brees and Russell Wilson.
The other area where we generally want to “go big” with quarterbacks is their school; overall, BCS quarterbacks have been much more successful than small-school passers.
This is in opposition to most other positions—most notably running back—where I think non-BCS players offer far more value as a whole.
So what’s different about quarterback? It’s difficult to say. It could be experience against top competition, it could be more efficient recruiting so that the best passers go to the biggest schools, or it could be a matter of confidence.
Whatever the reason, though, big-school quarterbacks tend to outperform non-BCS quarterbacks no matter where they’re drafted. Such numbers can be useful when deciding among rookie quarterbacks, or even second/third-year passers who haven’t played much early in their careers. For every Joe Flacco who has succeeded in spite of his small-school origins, there’s a J.P Losman, Kevin Kolb and John Beck who have proven the rule.
