For a variety of reasons, I was curious to know what percentage of receiving yards was gained by each class of players. As it turns out, second-year players gain the most receiving yards of any class. Year two players have the advantage of added experience over rookies, and also are less likely to be out of the league even if they aren’t very good, relative to older players. Even bad players usually make it to the field in year two.
One reason to study this data is to analyze receiver production versus draft class. Because of passing inflation, you can’t simply compare the receiving yards gained by the 1973 class to the receiving yards gained by the 2003 class. But what you *could* do is measure the percentage of yards gained by each class, which should control for era. For example, the famed 2014 rookie class (with Odell Beckham, Sammy Watkins, Mike Evans, Allen Robinson, et al.) was responsible for 13% of NFL receiving yards in 2014 and then 19% of receiving yards last year. Those numbers are both really, really good.
But before we get to individual classes, I want to take a look from the 30,000 foot view. Are there changes in this over time? And if so, are they due to random variation, or something meaningful? Take a look:1
There are some bumps in the data that are probably due to random variation — for example, I don’t have a good explanation for why 5th-year players dominated in the ’60s (particularly in ’60, ’61, and ’62, and then again in ’68 and ’69). Others may be meaningful: rookies and 2nd-year players were great in the ’50s, which may be a sign that the college game wasn’t far behind the pro game. Also, older players declined more in the ’50s, which likely corresponds to worse training, less year-round commitment, and players simply retiring earlier.
On the other hand, rookies and 2nd-year players were straight out bad in the ’90s, which may be a sign that NFL offenses were simply too complicated during that era. Or it could be a sign that the college game simply wasn’t producing NFL-ready wide receivers. Or maybe it’s nothing; who knows.
Let’s take a look at the same chart, but including the average of every year since 1950:
So if you wanted to study this data further, would you adjust for era, or would you use the black average line for all eras when determining what is the expectation? And if you do adjust for era, how precise would you be? I broke things down by decade, but that doesn’t mean it’s right, just simple. And finally, what would you like to see done with this data?
- Note that AFL data was excluded, as was the 1987 season. [↩]