"Scouting" by Number: Does WR Height Tell Us Anything that Dra (1 Viewer)

[AJules]

You often hear dynasty owners who downgrade any WR prospect who is shorter than 6'2" (or some similar number). And the raw data show that tall WRs dominate end-of-season rankings. But that's not surprising: after all, tall WRs also dominate the early rounds of the NFL draft. And any schmuck can base his dynasty rookie rankings off of NFL draft position, right?

The real question is whether WR height has predictive power independent of draft position. My tentative answer: not much, if any.

Methodology:

1. Obtained combine measurements for every WR drafted between 2006 and 2011.

2. Assigned a "career value" to each WR. First, I found the WR36 on a ppg basis for each year from 2006 to 2013 in 1ppr scoring. For all the WRs in the data set who (1) had a higher ppg than WR36 and (2) played 9 or more games that season, I calculated the ppg value over baseline, with a baseline of WR36. I then multiplied the result (ppg - baseline) by 16 to create a seasonal value. A player's career value is the sum of his seasonal values divided by the number of years since (and including) his rookie season.

3. Ranked players by height and separated them into three categories: short (5'8" - 5'10"), normal (5'11" - 6'1"), and tall (6'2"+).

Within each category, I ranked each player by where he was selected in the NFL draft, from earliest to latest. I then divided players into groups based on where they were drafted ("Day 1", "Day 2", "Day 3").

4. Compared career values of similarly drafted player groups with different heights ("short Day 2" v. "tall Day 2", etc.).

Results:

Day 1 (Round 1)

Tall: career value = 59.60; ADP 17.1; 10 drafted

Normal: career value = 25.33; ADP = 22.4; 5 drafted

Short: 0 drafted

This is a little misleading, as almost all top-10 picks in the sample were tall. In fact, if you only compare WRs picked in the 11-32 range, normal WRs performed slightly better than tall ones (36.73 to 30.31).

Day 2 (Rounds 2-3)

Tall: career value = 5.12; ADP 64.15; 20 drafted

Normal: career value = 4.08; ADP 74.81; 26 drafted

Short: career value = 7.66; ADP 60.5; 8 drafted

Tall WRs slightly outperformed normal ones, but the 10-pick gap in ADP may partially explain the difference. The surprising result here is the strong performance by the short group, who significantly outperformed the tall and normal WRs, despite only a modest ADP gap with the tall group. The short WRs' performance may be a product of small sample size.

Day 3 (Rounds 3-4)

Tall: career value = 6.12; ADP 184.00; 30 drafted

Normal: career value = 1.79; ADP 166.68; 34 drafted

Short: career value = 2.89; ADP 152.00; 14 drafted

Here, the tall WRs are clear winners. Despite being drafted later, they significantly outpaced both other groups. Among this group are Brandon Marshall, Mike Williams (TB), Brian Hartline, Stevie Johnson, and Marques Colston.

Conclusions

In the early-to-mid rounds, the data don't show any significant advantage for tall WRs. Their Day 1 advantage comes completely from Calvin, AJG, and Julio, all of whom were top-6 picks. And no normal or short WR was picked that early. (Tavon Austin will be an interesting test case, and the early returns aren't especially promising.) But if you're going to take a late-round flier, these data suggest you should go with a big guy.

 

[Xue]

Not only height, but weight is important as well.

 

[AJules]

Not only height, but weight is important as well.

I'll tease out the numbers on weight alone and try to do some things with BMI. Any magic numbers I should test?ETA: for those who use metrics, how do you control for draft position?

 

[Louche]

FF is primarily about having as many peaks in your roster as possible. It is much better to have a few top tier guys on your roster filled in with some mediocre players rather than having a more consistently good team without any great ones. Mid level production is quite easy to acquire whereas top tier production (WR 1-12) is very difficult to acquire. So what you are seeking when drafting, trading or working the waivers is not a top 36 or top 24 WR, it is a top 12 player (unless you are acquiring players to fill a short term need). And for the very top tier there just aren't a lot of guys that are short. There is a certain physical profile that is strongly overrepresented among the top WRs and size seems to be even more important than speed. I'll borrow a bit from Jonathan Bales' blog here...since 2008 there have been 35 instances of a WR scoring 10 TDS and 1000 yds in the regular season. Only 1 of these 35 were under 6'0" tall. Looking at the top 12 WRs from 2013 there are two players that are 5'10", one that is 6'2", and the remaining 9 players are 6'3" or higher.

http://www.dallascowboys.com/news/article-JonathanBales/Running-the-Numbers-Size-Not-Speed-Matters-Most-for-Wide-Receivers/387e52e0-0409-4d14-8902-38d7f2cc760f

 

[AJules]

FF is primarily about having as many peaks in your roster as possible. It is much better to have a few top tier guys on your roster filled in with some mediocre players rather than having a more consistently good team without any great ones. Mid level production is quite easy to acquire whereas top tier production (WR 1-12) is very difficult to acquire. So what you are seeking when drafting, trading or working the waivers is not a top 36 or top 24 WR, it is a top 12 player (unless you are acquiring players to fill a short term need). And for the very top tier there just aren't a lot of guys that are short. There is a certain physical profile that is strongly overrepresented among the top WRs and size seems to be even more important than speed. I'll borrow a bit from Jonathan Bales' blog here...since 2008 there have been 35 instances of a WR scoring 10 TDS and 1000 yds in the regular season. Only 1 of these 35 were under 6'0" tall. Looking at the top 12 WRs from 2013 there are two players that are 5'10", one that is 6'2", and the remaining 9 players are 6'3" or higher.

http://www.dallascowboys.com/news/article-JonathanBales/Running-the-Numbers-Size-Not-Speed-Matters-Most-for-Wide-Receivers/387e52e0-0409-4d14-8902-38d7f2cc760f

None of that is news. But it also doesn't address the relevant question:

And the raw data show that tall WRs dominate end-of-season rankings. But that's not surprising: after all, tall WRs also dominate the early rounds of the NFL draft. And any schmuck can base his dynasty rookie rankings off of NFL draft position, right?

The real question is whether WR height has predictive power independent of draft position.

 

[EBF]

Not only height, but weight is important as well.

I'll tease out the numbers on weight alone and try to do some things with BMI. Any magic numbers I should test?ETA: for those who use metrics, how do you control for draft position?

I think weight is a lot more important than height.

I roughly break WRs down into different categories based on their dimensions:

28+ BMI = big WR (i.e. Vincent Jackson, Larry Fitzgerald)

27-27.9 BMI = hybrid/tweener WR (i.e. Roddy White, Julio Jones)

26-26.9 BMI = speed/finesse WR (i.e. AJ Green, Percy Harvin)

Anything bigger than 29 is usually a TE or RB. Anything lower than 26 gets into the danger zone of being really really skinny. There are some success stories down there (DeSean Jackson) and some prospects who look like they might have a pulse (Justin Hunter), but it doesn't seem ideal.

I don't necessarily judge players based on their BMI, but all else being equal I prefer a bigger WR. More than anything, it helps me understand their "type" and find NFL parallels. For example, based on the listed measurements, Allen Robinson is going to come in at a mid 26 BMI. That makes sense because his skill set is similar to other WRs in the 26 BMI range like Reggie Wayne, AJ Green, Keenan Allen, and Rueben Randle. Those guys are all slightly different, but they're all quicker-than-fast receivers who excel because of their overall fluidity and suddenness despite lacking freaky strength/speed. Knowing that players of that ilk have been successful helps reassure me about Robinson's chances, whereas if he came out with a strange or unprecedented body type, it would be harder for me to put him in a historical context.

 

[Louche]

FF is primarily about having as many peaks in your roster as possible. It is much better to have a few top tier guys on your roster filled in with some mediocre players rather than having a more consistently good team without any great ones. Mid level production is quite easy to acquire whereas top tier production (WR 1-12) is very difficult to acquire. So what you are seeking when drafting, trading or working the waivers is not a top 36 or top 24 WR, it is a top 12 player (unless you are acquiring players to fill a short term need). And for the very top tier there just aren't a lot of guys that are short. There is a certain physical profile that is strongly overrepresented among the top WRs and size seems to be even more important than speed. I'll borrow a bit from Jonathan Bales' blog here...since 2008 there have been 35 instances of a WR scoring 10 TDS and 1000 yds in the regular season. Only 1 of these 35 were under 6'0" tall. Looking at the top 12 WRs from 2013 there are two players that are 5'10", one that is 6'2", and the remaining 9 players are 6'3" or higher.

http://www.dallascowboys.com/news/article-JonathanBales/Running-the-Numbers-Size-Not-Speed-Matters-Most-for-Wide-Receivers/387e52e0-0409-4d14-8902-38d7f2cc760f

None of that is news. But it also doesn't address the relevant question:

And the raw data show that tall WRs dominate end-of-season rankings. But that's not surprising: after all, tall WRs also dominate the early rounds of the NFL draft. And any schmuck can base his dynasty rookie rankings off of NFL draft position, right?

The real question is whether WR height has predictive power independent of draft position.

Ok. But you're basing your results on average "career value" in each grouping. As long as we're talking fantasy I don't see that as the right number to look at. What's interesting is how many top tier players each grouping has produced. How many "peaks" there are. If a player has put up average production, poor production, or was a total flunk...it doesn't really matter for fantasy as all these players are practically useless in most fantasy formats, but it affects your average "career values". What matters is how many players from each grouping put up top tier production.

 

[tdmills]

My theory is that shorter players are more scheme dependent. Taller players are better in PPR and redzone...aka FF.

 

[AJules]

FF is primarily about having as many peaks in your roster as possible. It is much better to have a few top tier guys on your roster filled in with some mediocre players rather than having a more consistently good team without any great ones. Mid level production is quite easy to acquire whereas top tier production (WR 1-12) is very difficult to acquire. So what you are seeking when drafting, trading or working the waivers is not a top 36 or top 24 WR, it is a top 12 player (unless you are acquiring players to fill a short term need). And for the very top tier there just aren't a lot of guys that are short. There is a certain physical profile that is strongly overrepresented among the top WRs and size seems to be even more important than speed. I'll borrow a bit from Jonathan Bales' blog here...since 2008 there have been 35 instances of a WR scoring 10 TDS and 1000 yds in the regular season. Only 1 of these 35 were under 6'0" tall. Looking at the top 12 WRs from 2013 there are two players that are 5'10", one that is 6'2", and the remaining 9 players are 6'3" or higher.

http://www.dallascowboys.com/news/article-JonathanBales/Running-the-Numbers-Size-Not-Speed-Matters-Most-for-Wide-Receivers/387e52e0-0409-4d14-8902-38d7f2cc760f

None of that is news. But it also doesn't address the relevant question:

And the raw data show that tall WRs dominate end-of-season rankings. But that's not surprising: after all, tall WRs also dominate the early rounds of the NFL draft. And any schmuck can base his dynasty rookie rankings off of NFL draft position, right?

The real question is whether WR height has predictive power independent of draft position.

Ok. But you're basing your results on average "career value" in each grouping. As long as we're talking fantasy I don't see that as the right number to look at. What's interesting is how many top tier players each grouping has produced. How many "peaks" there are. If a player has put up average production, poor production, or was a total flunk...it doesn't really matter for fantasy as all these players are practically useless in most fantasy formats, but it affects your average "career values". What matters is how many players from each grouping put up top tier production.

I defined career value based on VBD with a baseline of WR36. So an "average" NFL season wouldn't register. And a WR30 performance won't matter much either. I believe this addresses your concern. Unless maybe you think the baseline should have been higher?

 

[AJules]

My theory is that shorter players are more scheme dependent. Taller players are better in PPR and redzone...aka FF.

No one's disputing that height is an advantage. The question is this: if a dynasty owner is debating between two rookies drafted in the early second of the NFL draft, should he care that one is 5'11" and the other is 6'3"?

 

[BuzzCagney]

My theory is that shorter players are more scheme dependent. Taller players are better in PPR and redzone...aka FF.

No one's disputing that height is an advantage. The question is this: if a dynasty owner is debating between two rookies drafted in the early second of the NFL draft, should he care that one is 5'11" and the other is 6'3"?

it's not complicated: bigger is better.

AT&T TV Commercial - It's Not Complicated "Pool" - YouTube

 

[az_prof]

You often hear dynasty owners who downgrade any WR prospect who is shorter than 6'2" (or some similar number). And the raw data show that tall WRs dominate end-of-season rankings. But that's not surprising: after all, tall WRs also dominate the early rounds of the NFL draft. And any schmuck can base his dynasty rookie rankings off of NFL draft position, right?

The real question is whether WR height has predictive power independent of draft position. My tentative answer: not much, if any.

Methodology:

1. Obtained combine measurements for every WR drafted between 2006 and 2011.

2. Assigned a "career value" to each WR. First, I found the WR36 on a ppg basis for each year from 2006 to 2013 in 1ppr scoring. For all the WRs in the data set who (1) had a higher ppg than WR36 and (2) played 9 or more games that season, I calculated the ppg value over baseline, with a baseline of WR36. I then multiplied the result (ppg - baseline) by 16 to create a seasonal value. A player's career value is the sum of his seasonal values divided by the number of years since (and including) his rookie season.

3. Ranked players by height and separated them into three categories: short (5'8" - 5'10"), normal (5'11" - 6'1"), and tall (6'2"+).

Within each category, I ranked each player by where he was selected in the NFL draft, from earliest to latest. I then divided players into groups based on where they were drafted ("Day 1", "Day 2", "Day 3").

4. Compared career values of similarly drafted player groups with different heights ("short Day 2" v. "tall Day 2", etc.).

Results:

Day 1 (Round 1)

Tall: career value = 59.60; ADP 17.1; 10 drafted

Normal: career value = 25.33; ADP = 22.4; 5 drafted

Short: 0 drafted

This is a little misleading, as almost all top-10 picks in the sample were tall. In fact, if you only compare WRs picked in the 11-32 range, normal WRs performed slightly better than tall ones (36.73 to 30.31).

Day 2 (Rounds 2-3)

Tall: career value = 5.12; ADP 64.15; 20 drafted

Normal: career value = 4.08; ADP 74.81; 26 drafted

Short: career value = 7.66; ADP 60.5; 8 drafted

Tall WRs slightly outperformed normal ones, but the 10-pick gap in ADP may partially explain the difference. The surprising result here is the strong performance by the short group, who significantly outperformed the tall and normal WRs, despite only a modest ADP gap with the tall group. The short WRs' performance may be a product of small sample size.

Day 3 (Rounds 3-4)

Tall: career value = 6.12; ADP 184.00; 30 drafted

Normal: career value = 1.79; ADP 166.68; 34 drafted

Short: career value = 2.89; ADP 152.00; 14 drafted

Here, the tall WRs are clear winners. Despite being drafted later, they significantly outpaced both other groups. Among this group are Brandon Marshall, Mike Williams (TB), Brian Hartline, Stevie Johnson, and Marques Colston.

Conclusions

In the early-to-mid rounds, the data don't show any significant advantage for tall WRs. Their Day 1 advantage comes completely from Calvin, AJG, and Julio, all of whom were top-6 picks. And no normal or short WR was picked that early. (Tavon Austin will be an interesting test case, and the early returns aren't especially promising.) But if you're going to take a late-round flier, these data suggest you should go with a big guy.

This is intriguing, but it misses the boat for this reason: with rare exceptions, most of us are not interested in whether or not a short WR4 is better or worse than a tall WR4. I am interested in the elite players. Looking only at elite players, top 20 WR, how many are taller than average? Shorter than average?

 

[coolnerd]

My theory is that shorter players are more scheme dependent. Taller players are better in PPR and redzone...aka FF.

No one's disputing that height is an advantage. The question is this: if a dynasty owner is debating between two rookies drafted in the early second of the NFL draft, should he care that one is 5'11" and the other is 6'3"?

it would depend on situation and his judge of football skill over raw metrics. If you are going to make a blind percentage play then taking the bigger, faster, stronger guy is an easy call.. But if you trust the teams (or your own ability) to analyze the combination of football skill and situation, then there is no one size fits every situation answer. Remember, even in the 2nd round (NFL) getting a FF player who is even "solid" is less than 35% , much less anything approaching elite., so you probably going to be wrong either way.

With a smaller WR, also check whether the team is drafting him in this range because he has above average return skills and/or because they want the big play but non dependable aspect to this player's game.

 

[AJules]

My theory is that shorter players are more scheme dependent. Taller players are better in PPR and redzone...aka FF.

No one's disputing that height is an advantage. The question is this: if a dynasty owner is debating between two rookies drafted in the early second of the NFL draft, should he care that one is 5'11" and the other is 6'3"?

it's not complicated: bigger is better.AT&T TV Commercial - It's Not Complicated "Pool" - YouTube

 

[AJules]

This is intriguing, but it misses the boat for this reason: with rare exceptions, most of us are not interested in whether or not a short WR4 is better or worse than a tall WR4. I am interested in the elite players. Looking only at elite players, top 20 WR, how many are taller than average? Shorter than average?

I counted a #4 WR season as exactly 0 points.

A mid-#3 WR (WR30) season was worth about 15 points.

A mid-#2 WR (WR18) season was worth about 65 points.

A mid-#1 WR (WR6) season was worth about 120 points -- eight times as much as a #3 WR season.

So explain exactly how this "misses the boat".

Note: some of you may have noted that I made this post earlier as my "alter ego". I'm switching usernames so that I'm the same person in multiple FF forums. My old username doesn't have much of a reputation (good or bad) anyway.

 

[tdmills]

My theory is that shorter players are more scheme dependent. Taller players are better in PPR and redzone...aka FF.

No one's disputing that height is an advantage. The question is this: if a dynasty owner is debating between two rookies drafted in the early second of the NFL draft, should he care that one is 5'11" and the other is 6'3"?

I did answer the question...taller players are better in FF.

 

[matuski]

Why would you arbitrarily single out a single measurement like this?

A player's height is not a determining factor of anything independently. Ceteris paribus, you would always take the taller player... but I don't think that makes for much debate, and obviously that is not a realistic assumption.

 

[Concept Coop]

You often hear...

I like the work and agree with your conclusion(s). I think a few posters may have missed your aim.

A lot of people are going to take Mike Evans over Marqise Lee, based on height (I find myself leaning this way too). This simply shows that if the NFL grades them as equal prospects, despite their difference in height, odds are (if we trust the study and sample size), the production will match, despite the height difference.