Friday, April 25, 2014

Jordan Matthews: Stealth Midget

I've really struggled a fair bit with how I feel about Jordan Matthews.  As a prospect, he is simply beyond weird.  Still, for the most part, I have had generally positive things to say about him.  I've tried to temper my tendency to speak negatively about every single prospect that crosses my path.  He easily met the computer's criteria when it came to statistical production.  When it came to his athletic ability, he also managed to meet the computer's standards for a 'Big' wide receiver, though he wasn't exactly off the charts in this area.

Despite all of that, there are some things which make me very nervous about Matthews.  Part of the problem, I can't deny, is that I figure there is a good chance of looking like an idiot if I bet too strongly against somebody who dominated the SEC the way Matthews did.  This is why I let the computer do most of the thinking for me.  If I am wrong about a prospect, I can blame a wonky hard drive.  In this case, as I said, the computer thinks Matthews falls within the acceptable range of what it would expect to see in a successful wide receiver prospect.  It's not an insignificant accomplishment for a receiver to get the computer's blessing, and I don't treat the computer's recommendations lightly.  Still, I do get the feeling that Matthews might be a bit of an optical illusion.

When you hear that Matthews is just a tad over 6'3"tall, and weighs in at 212 pounds, you naturally think "Hey, that's a pretty big guy!"  Because of the wight-class he falls into, the computer judges him against other 'Big' wide receivers, who are over 200 pounds.  No, he's not quite as big as Andre Johnson, Larry Fitzgerald, or Julio Jones, but he does have nice physical proportions, which leads you to look at him as a potentially intimidating #1 wide receiver.  This creates a somewhat odd problem, because even though the computer says he meets all of the criteria of a 'Big' receiver, he doesn't really move the needle too far when it comes to his athletic ability, at least in comparison to some of his bulky brethren. 

When I was discussing Jordan Matthews, I suggested that he might be more like a 'Small' receiver in a 'Big' receivers body.  So, let's think about this for a second.  Is it just the physical dimensions of a receiver that should get us excited about their potential?  Or, is it the power that generally resides in these larger frames, which really makes a difference?

There are all sorts of physical traits that you can look at with a receiver.  First of all, you want him to actually be able to catch the ball.  That seems pretty obvious, though teams do ignore this quite frequently (Stephen Hill!).  Some measure of speed and quickness is also nice.  We can look at the player's 10-yard split and 40-time to get some sense of that, and Matthews appears to do quite well in this area.  There's also the issue of agility, and the ability of a receiver to contort his body in seemingly unnatural ways, or have the elusiveness to get open.  With larger prospects, we also start to dream of something even more unusual.  We imagine a kind of physical dominance, and the ability to simply overwhelm an opposing cornerback.  For this, I think we want plain and simple old-fashioned power.

As I've said before, when the computer is weighing the 'Big' receivers, it puts more of an emphasis on the player's Kangaroo Score than it does for the 'Small' prospects.  The Kangaroo Score is my measure of lower body power, based on the player's vertical jump and broad jump, in relation to the player's weight.  Some people shift more of their attention to a player's bench press results, but I just suspect that lower body power is the more important factor.  I'm not suggesting that this is the be-all-end-all of physical traits, but I do think it has some significant bearing on a player's ability to plow through an opponent, and probably increases the difficulty with which a player is going to be tackled or pushed around.

When the computer comes up with the Kangaroo Score for a player, it basically averages the results from the vertical jump and broad jump, while factoring in a player's weight.  Normally this isn't a huge problem.  The two scores generally mirror each other, to some degree.  If you take a player's vertical jump, and multiply it by 3.5, you'll probably wind up pretty damn close to what their broad jump is.  This isn't always the case, however.

When I previously discussed Jordan Matthews, I expressed some concern about his rather mediocre Kangaroo Score 0.232.  This result is given in the form of how many standard deviations he was above the average result for a wide receiver, and while his result is technically above average, it is still really quite pedestrian.  Regardless, I suggested that players such as A.J. Green, Jordy Nelson and Sidney Rice had also produced rather pedestrian Kangaroo Scores, yet managed to go on to have a significant degree of success.  This is true, but it is also a bit misleading, as these 3 players had rather peculiar results that might require deeper examination.  So, let's look at their Kangaroo Scores, and break things down  a bit more thoroughly.


NAME         Vert Kangaroo    Broad Kangaroo      Total Kangaroo
Jordan Matthews 0.160 0.340 0.232
Sammy Watkins -0.270 0.821 0.166
AJ Green -0.141 0.821 0.243
Jordy Nelson -0.816 0.865 -0.143
Sidney Rice 0.617 -0.346 0.231

There's something I should mention here, before we go any further.  The 'average' of the Vertical Kangaroo Score and the Broad Jump Kangaroo Score, obviously isn't weighted in a 50/50 manner.  The Total Kangaroo Score gives a slight advantage to the Vertical Kangaroo Score, with more of a 60/40 split.  There are reasons for this, which I'll try to get into as we go along.

Either way, all 5 of these players have Total Kangaroo Scores that would basically fall in the somewhat 'average' range.  When we look a bit deeper, however, we see that 4 out of these 5 players did manage to rise at least 0.500 standard deviations above the average, in one of the two jumps.  Watkins, Green and Nelson, are rising above the 0.500 mark somewhat significantly, in at least one of the two jumps, so there is at least some suggestion that power is lurking within them.  Matthews, on the other hand, seemed to demonstrate merely average power, regardless of the jump.

Now, what could all of this actually mean?  There was an article I linked to, at one time, that discussed some of the potential differences between a vertical jump and a broad jump, in relation to speed and quickness.  I'll post it again, here, though I don't want anyone to think that I swear by any of the theories expressed in this article.  Basically, the writer suggests that this is all about muscle development, and says that players with an above average vertical jump might have better quickness, while players with a better broad jump might have better deep speed.  I wouldn't say that my own conclusions on this subject are entirely set in stone, but based on a number of players that I have gone back to examine, there might be some merit to this idea.  Still, I wouldn't overreact to any of this, as the results only seem to gain some debatable value when there are fairly extreme variations in the two results.

The potentially more important issue that I want to address is a bit different.  Instead of discussing how these things relate to speed or quickness, let's just focus on the value of having some sort of demonstrable power, that is above the average expected result.  We're talking about power for the purpose of violence and intimidation.  While the majority of the players in the above mentioned group had wildly varying results between their vertical jump and their broad jump, at one time or another, they did demonstrate significant power of some sort.  This isn't necessarily a knockout blow to Matthews, it just raises the question of whether we should have certain expectations of him performing like a conventional 'Big' wide receiver.

Now, this has all been comparing Matthews to 4 particular/peculiar players, who produced results that might have needed a deeper examination.  Perhaps, some of these players would have produced less divergent results in their two jumps, if given another chance.  It is impossible to say, though many players do correct these inconsistencies/irregularities when they perform at their pro days.  Either way, we will never know the answers to some of these questions.

Instead, let's ask a question about a broader selection of wide receivers.  While contemplating this year's wide receiver prospects, I reexamined some of the prospects from the last 15 years.  Among the 'Big' receivers, I set aside 40 players who I felt had managed to become statistically significant in their NFL careers.  'Statistically significant' is obviously rather vague, and my opinions on this are probably a bit biased.  Nonetheless, I suspect we would all agree on most of the players I included in this examination.  They tended to be the obvious sorts of players, that have become well known to us all.

Among these 40 'Big' receivers, there were only five for whom no data/insufficient date related to the vertical jump or broad jump was available.  These five players were Hakeem Nicks, Michael Crabtree, Eric Decker, Roy Williams, and Demaryius Thomas.  There's no conclusion we can come to about these players, but I'll leave them in the calculations anyway.  They represent 12.5% of the 40 players I will be mentioning, and will count as players without exceptional Kanagroo Scores, though I have to suspect that some of them would have actually done quite well in this area.

21 of the 40 (52.5%) players managed to produce both vertical jumps as well as broad jumps that were approximately 0.500 standard deviations above average, or better   The players in this group included the majority of the Calvin Johnson, Larry Fitzgerald, Andre Johnson, Roddy White, Julio Jones, Marques Colston, Vincent Jackson variety of receiver, and most were producing results well above a mere 0.500.  The median Total Kangaroo Score among these 21 players was a fairly stunning 1.095

29 out of these 40 (72.5%) 'Big' receivers managed to rise above the 0.500 standard deviation threshold in at least one of the two jumps.  This somewhat broader group might include more enigmatic players, but at one time or another they did display some power.  This group obviously includes the 3 players that I used in the comparison to Jordan Matthews, as well as others such as, Dwayne Bowe, Pierre Garcon, David Boston, Jerricho Cothcery, James Jones as well as a few others.  It's still a rather respectable group.

Now, we get to the 6 players out of the 40 (15%) I examined, who failed to produce a result over 0.500 in either of their 2 jumps.  These six players were Justin Blackmon, Cecil Shorts, Brandon LaFell, Anquan Boldin, Austin Collie, and Steve Johnson.  There are some good players in this group, but with the exception of Anquan Boldin, who is a well established outlier, it is very difficult to say that many of the others would be mistaken for 1st tier wide receivers.  Some, like Shorts and Collie (one of the more peculiar inclusions in this list, and someone I should have probably excluded), clearly are more comparable to what we think of as small receivers, and only make this list because their weight, being over 200#, makes them technically eligible, while being obvious misfits for 'Big' receiver comparisons.

Like I said though, there are still those 5 rather significant receivers for whom we have either insignificant results, or no data at all.  You can use your own judgement as to whether they should bolster the results of one of these groups, or another.

At least to some extent, I think you can probably see why I have some concerns about Jordan Matthews, and his fit as a conventional 'Big' receiver.  None of this is meant to imply that he can't do well, or that I don't think he will do well.  I'm just saying that if you think he is somebody who will perform like many of the larger receivers who have come before him, I would tend to hedge my bets on that.  The possibility that he could perform well in a role similar to what you see smaller receivers doing, on the other hand, is quite possible, and probably a better fit for his particular skills, namely his above average speed.  His statistical production, also, can't easily be ignored.  Still, when I have watched him play, it was this speed that caught my eye, and not any sort of abundance of power or physicality.  Like I said, I think Jordan Matthews is kind of an optical illusion.

5 comments:

  1. I really enjoy your work. Always an awesome read. I'm not sure if you have the official combine numbers, but I have them in a spreadsheet:

    https://docs.google.com/spreadsheet/pub?key=0AotIVo4gpWF8dEJqRmtrSjktMm5nMFRjOTNielVJV1E&output=html

    If the number is in bold, I listed the respective Pro Day numbers. In a few cases with the QB's (Mett, Murray), I listed their Junior Pro Day numbers. All others are the lowest official electronic time from the Combine. The hand numbers were all over the place. If you'd like the official copy, shoot me an email and I'll send it to you.

    I also have a spreadsheet for all the lesser known guys. You might be able to find a nugget or two or some numbers you can't locate. Here is the link:

    https://docs.google.com/spreadsheets/d/1cx-zsXhwoOBZcMH1azWPoHE524JDQvDDSZEnhY90-LY/pubhtml

    Again, I love your work, and wanted to show some appreciation for your efforts.

    ReplyDelete
  2. I failed to mention....the 10 yard splits the NFL Network reported during the Combine are significantly different than the official times. I thought you'd be interested to having the actual numbers.

    ReplyDelete
    Replies
    1. This really is a small world. I've run across your work/name numerous times over the last couple of years, while scouring the internet for combine data. Sometimes I really start to suspect that there are only 15-20 people who write about the draft, and we all just spend our time reading each other's work, since nobody in their right mind would find this stuff terribly interesting, except for us lunatics.

      Either way, I'm pleased and honored to have you pop in here, and wanted to thank you for the additional data to look through. I did have two questions for you, though maybe I should put this in an email instead. First, where are you posting most of your work nowadays, as I seem to just see you pop up in odd places from time to time? Secondly, were you the guy who used to maintain a rather excellent spreadsheet on college wide receiver drop rates/catch rates? I could have sworn that was you, and if so it was excellent, but I haven't seen it around for a while.

      Feel free to email me anytime, as I'm sure I'll probably end up bugging you with my own questions at some point.

      Delete
  3. You make a great point. I really think there are only a few of us that enjoy this stuff. I honestly don't understand it either. I think there are a few teams that really use it....the Seahawks for one, but it's going to take the league by storm at some point.

    I've been following your work since you started the blog, and I've loved every piece you did. Even though your thoughts are aligned with mine almost every time, the way you present it is genius. As for me, I don't write for anyone. My skills absolutely stink in that department. I didn't do the spreadsheet for drop and catch rates either. I know a guys that frequents Footballguys.com did though. I probably have it somewhere if you need it....just let me know.

    Again, I really enjoy your stuff. I hope you keep at it......

    ReplyDelete
    Replies
    1. I think you're right about the Seahawks. Their personnel decisions in recent years have been fascinating to me, as well as their connection to Scot McCloughan, who I get the sense is into this sort of stuff. I thought it was really unfortunate to see him resign the other day.

      Delete