If you’re on Twitter and you follow a decent number of fantasy football minds, you might have seen some serious discussion over the last few days about the role size plays in wide receiver production. If you’ve been on for a while, you might remember that a similar debate raged during last offseason, too. For those who aren’t on Twitter, though, I wanted to bring my take on this discussion to you here, where I can spell out my thoughts with more than 140 characters at a time.
First off, let’s start with a basic acknowledgment. If all else is equal, more size is more better for most positions most of the time. This is trivially easy to demonstrate. The NFL has been around for nearly 100 years now, and here’s a visualization of how average player size has changed over that time. Notice the blob creeping upward and rightward? Players have been getting bigger and taller throughout history. Therefore, bigger and taller is better. QED.
The problem, of course, lies with that pesky little qualifier: “if all else is equal”. You see, we know that when it comes to player size, all else is never equal. This is a gospel fact. We know, for instance, that weight carries a very strong negative correlation to 40-yard dash time. And, insofar as height is positively correlated with weight, we would anticipate that to also negatively correlate with speed. In other words, bigger players are slower than smaller players.
Being big carries other penalties, too. We know that big people have shorter lifespans and higher injury rates. We know that carrying too much size is hard on the knees and shortens athletic careers. We know that taller players have longer limbs, and can easily posit that that might hinder their short-area quickness, as longer legs have trouble taking shorter steps.
The question then becomes whether the positives associated with extra size outweigh the negatives associated with extra size. And for that question, there are three possible answers: yes they do, no they do not, or yes they do… up to a certain point.
Before we continue, however, I should offer a few clarifications. First, when people discuss “size”, they are really discussing one of three things. They either mean height, or they mean weight, or they mean “mass” or “density”. Density is a little bit hard to pin down; many people have different definitions, and frankly, I don’t have great historical data for it. As a result, I will not be addressing it in this article. This means if you believe that, say, BMI is the key to success at receiver, this article won’t be a very effective rebuttal. It’s more aimed at those supporting height- or weight-based analysis.
Because there are several schools of thought on just what matters, I might wind up using the terms “height”, “weight”, or “size” interchangeably. I apologize if this seems to imply a lack of precision. Unless I am talking about specific weights or specific sizes, for the remainder of this article you can assume that any reference to any of those attributes is referring to the concept of size in general.
Evaluating Claims that Size Matters
Alright, disclaimers over. The main assertion of proponents of big receivers is that big players are better at scoring touchdowns. And, indeed, Mike Clay of Pro Football Focus just wrote a very detailed study indicating that there is a very strong positive relationship between a player’s height and the “value” of his targets, in terms of how many touchdowns we would expect them to result in.
This is very important, very compelling data, which comports fully with my earlier assertion that more size is more better for most positions most of the time. I would expect the best receivers in the NFL to be bigger than average. I would also expect the best receivers to be faster than average, and smarter than average, and have better hands than average, and to have better college production than average, and to have been younger as rookies. These are all positive traits, and positive traits correlate with positive production.
If I’m given a choice between some Joe Schmoe who is 5’11” with mediocre hands and a 4.6 forty, and another Joe Schmoe who is 6’4” with mediocre hands and a 4.6 forty, I’d prefer the latter guy. And the NFL clearly agrees with me. I pulled draft data for all receivers in the last 10 drafts and sorted them into 10-player buckets based on their draft position, (slightly more in the case of ties). Here’s where those players were drafted, and what their average height was.
| Draft Position | Average Height |
|---|---|
| Picks 2-8 | 74.3” |
| Picks 9-22 | 72.8” |
| Picks 23-29 | 73.7” |
| Picks 30-36 | 73.3” |
| Picks 39-44 | 73.3” |
| Picks 45-53 | 73.5” |
| Picks 54-61 | 73.0” |
| Picks 63-74 | 72.8” |
| Picks 75-79 | 72.9” |
| Picks 80-83 | 72.3” |
| Picks 84-89 | 72.7” |
| Picks 90-96 | 71.6” |
| Picks 97-104 | 71.3” |
That’s a three-inch drop in average height between the guys at the top of the 1st round and the guys at the end of the 3rd and beginning of the 4th. So it should come as no surprise that taller receivers are more heavily targeted, because taller receivers are more highly-drafted.
(As an aside: it’s interesting that player height drops so dramatically in the second bucket before rebounding again. This might suggest that the NFL grabs any big players with tremendous skills very early, then goes back and grabs any small players who also have tremendous skills in the middle of the first round, before returning to the practice of drafting based on a combination of size and skill. And in case anyone is curious, Tavon Austin actually falls at the very end of the first bucket. If we shifted him from the first to the second, the average height of the top guys rises to 74.9”, while the average height of the second group falls to 72.5”.)
So anyway, Mike Clay’s analysis tells us that we should expect the guys getting the most valuable targets to be bigger… but draft position already told us that, anyway. The better question, in my mind, is what happens if we control for receiver quality. Instead of comparing the entire range of NFL wide receivers— the tall guys from the top of round 1 against the short guys from the middle of round 3— what happens if we look at the height of just those receivers who have proven themselves to be pretty good?
Mike Clay focuses his analysis on OTD, a proprietary stat he uses that calculates how many touchdowns a player should be expected to score based on where his targets are coming on the field. Clay’s database is much better than mine, but the discussion essentially boils down to a receiver’s touchdown-scoring ability. To that end, I looked at how effective receivers were when their offense was inside the 10 yard line.
Pro-Football-Reference provided searchable play-by-play information going back to 1998, so I pulled up the 50 most-targeted receivers inside the 10-yard line over that span. Because of ties, I actually wound up with 52 receivers, all of whom received at least 39 targets when their offense was within 10 yards of the end zone. These 52 receivers represented the most prolific end-zone weapons of the last decade and a half.
Once I had my list, I found heights for all 52 players. I found that the average height of the WRs in the sample was just 73.3”, or just a shade over 6’1”. In other words, the 52 most prolific end-zone receivers of the last 16 years were not actually tall, on the aggregate. I weighted each of the 52 receivers’ height by the number of touchdowns he scored and the group average jumped all the way to 73.4”. In other words, the taller receivers in the sample weren’t really scoring any more touchdowns than the shorter receivers- or, at least, not enough more to do anything to move the group average.
The correlation between height and end-zone targets in the sample was just 0.090. That’s very, very weak, but at least it’s a positive relationship. Taller receivers in the sample received, on average, ever-so-slightly more end-zone targets.
The correlation between height and total touchdowns on those targets, on the other hand, was just 0.026. That’s virtually indistinguishable from zero, which means there was basically no relationship between how tall a receiver was and how many touchdowns he scored inside the 10-yard-line.
Based on those two correlations, this last bit won’t come as a surprise, but the correlation between player height and touchdown%, (or touchdowns scored divided by total targets), was -0.133. That’s still a very weak correlation, though stronger than the first two. A negative correlation actually tells us that the shorter receivers were more efficient at converting targets into touchdowns at the goal line.
Indeed, if we slice the data another way, we see something similar. Of the 52 receivers, 19 were 6’3” or taller, 14 were either 6’1” or 6’2”, and 19 were 6’0” or shorter. Those cutoffs divide the data into rough thirds.
The tallest third— the receivers who stood 6’3” or taller— scored a touchdown on 33.5% of their targets within 10 yards of the end zone. The middle third scored a touchdown on 34.7% of their targets within 10 yards of the end zone. And the shortest third— the receivers who were 6’0” or shorter— scored a touchdown on 35.8% of their targets within 10 yards of the end zone.
In short, the most prolific goal-line receivers of the last 16 years were, on average, not very tall. And within that subset of elite goal-line threats, the taller receivers were more likely to get targets, while the shorter receivers were more likely to convert the targets they got. The end result was that there was virtually no correlation between player height and touchdowns scored in that group.
There are a lot of good theories for why the data looks this way. Personally, I lay a lot of the blame at the feet of the fade pass. Tall receivers are much more likely to be targeted on fade routes than short receivers. This inflates their total number of targets. Fade passes, though, are low-percentage plays, which hurts the tall cohort’s success rates.
Either way, there’s a lot of reason to conclude that, if we already know a receiver is good, his height doesn’t matter nearly as much. Speed is important for receivers, but once we know that Anquan Boldin and Larry Fitzgerald are good, it doesn’t really matter that they only ran a 4.71 and 4.63 forty-yard dash, respectively. Similarly, height is a useful attribute, but the fact that Marvin Harrison and Greg Jennings were both 6’0” or shorter hardly stopped them from being two of the top five receivers in touchdown rate.
Or, to look at it from another angle, if the only thing we knew about a receiver was whether he was short or tall, we should prefer the tall receiver. And if the only thing we knew about a receiver was whether he was fast or slow, we should prefer the fast receiver. Which I’m sure will be useful information if I ever find myself choosing between two different players who I only know one thing about, a scenario that I have yet to find myself in.
I’ll have a lot more to say on player size in the next few days, but that’s the analogy that I’d like to end on today. Height is important for receivers, much like speed is important for receivers. At the same time, height is just one tiny piece of the puzzle, just like speed is just one tiny piece of the puzzle. I see no particular reason why size should be advantaged over every other positive attribute when evaluating receivers in the NFL or in fantasy.