A Tale of Two Beasts (Part I – Mark Harrison)
Recently, Frank posted this article about the freakishly athletic rookie TE Chris Gragg. He ranked Gragg’s athleticism – via his Physical Score metric – against the wideouts in the 2013 class to see how he stacks up. Reading that article, what caught my eye (other than Gragg’s gaudy measurables) were the second and third names in the table behind Gragg: Mark Harrison and Marcus Davis.
In a class with other athletically gifted receivers you’ve heard of like Da’Rick Rogers, Justin Hunter, Ryan Swope, and *ahem* Cordarrelle Patterson…Harrison and Davis aren’t getting very much love from the draft analyst community despite being quite significantly ahead of the pack on this metric. See for yourself…
|NAME||POS||HT||WT||40 Time||VL||Physical Score|
Scores greater than 2 represent above-average athletes and scores north of 4 put a player in some seriously elite company alongside guys named Andre and Julio. Now, we all know athleticism isn’t the only thing that matters for success in the NFL (see Jackson, Chad). But at the same time, you can’t teach fast. On the surface, both Harrison and Davis project as above average NFL physical talents.
To go beyond just the physical, we need to look at what their college playing stats tell us. Adding some form of normalized college production factor alongside the athletic measurement yields a model that does an even better job projecting NFL success. Call it Market Share, Dominator Rating, Productivity Score, etc. What a player did in college – normalized for his team’s abundance or dearth of raw passing offensive production – matters.
For this analysis, I decided to take a bit of an “upside” or “best case scenario” approach to these guys, since they are both categorized as “downfield threats” and it appears in all likelihood that they will both be available later into the NFL draft as well as your rookie drafts. Fliers, if you will. We know their downside is just another über-athlete who never seems to put it together and fails to becomes a NFL (and therefore fantasy) contributor. But if everything breaks right, what could they be?
To do this, I took a Similarity Score approach. Using data from sports-reference.com I gathered college player seasons going back to 2000 and calculated their Dominator Rating (DR) and Yards Per Reception (YPR). Again, this is an “upside” analysis so I chose YPR over Yards Per Target. I think of YPT as more of an overall efficiency measure that rewards for catch ratio (assuming players normally catch passes for more than zero yards) as well as what you do after the catch. I like YPR in this case because players who are vertically explosive with their receptions don’t get dinged if they also drop more passes than a possession guy.
Taking both DR and YPR into account, I then compared both players’ final college season to the college seasons of players who went on to record a top 60 fantasy finish (WR5 in a 12 team league) in the NFL. It’s important to note, I took the best fantasy finish for each comparable player in all his seasons in the NFL. So any time Braylon Edwards comes up as a match, he gets credit for his 2007 top 3 finish. This is where I feel I’ve taken a bit of a “best case scenario” approach. For every top 60 WR who shows up as a match for these players, there are another 4 or so players who haven’t ever recorded a top 60 finish, but I ignore them for the purposes of measuring the upside.
Like the Similarity Score App, I take the best 15 matches for each player and then look at the average of their stats. Here’s how some recent NFL draftees with elite physical profiles and high rookie draft selections fared with my model:
|Player||Year||School||Rec||Yds||TD||YPR||DR||NFL Best Rank|
|Demaryius Thomas||2009||Georgia Tech||46||1154||8||25.1||0.69||5|
|Thomas Comp Avg||66.1||1236.7||11.9||18.9||0.5||19.5|
|Dez Bryant||2008||Oklahoma State||87||1480||19||17||0.61||3|
|Bryant Comp Avg||67.2||1136.3||11.7||16.9||0.48||15.8|
|Calvin Johnson||2006||Georgia Tech||76||1202||15||15.8||0.55||1|
|Johnson Comp Avg||71.7||1139.3||10.7||15.9||0.45||20.6|
|Meachem Comp Avg||64.4||1123.7||9||17.5||0.42||24.9|
|Golden Tate||2008||Notre Dame||58||1080||10||18.6||0.37||32|
|Tate Comp Avg||59.7||1043.2||9.1||17.7||0.37||19.6|
|Greg Little||2009||North Carolina||62||724||5||11.7||0.34||52|
|Little Comp Avg||79.7||991.6||9.2||12.5||0.31||29.8|
|Green Comp Avg||57.2||858.1||6.1||15.1||0.31||23.9|
|Baldwin Comp Avg||57.3||880.5||6.7||15.4||0.31||16.1|
|Heyward-Bey Comp Avg||53.9||806.2||6.7||15||0.27||36.1|
|Harvin Comp Avg||44||688.9||5||15.7||0.21||28.2|
The table is sorted on DR in descending order. The Avg line below each player is just what the name would imply – the average of the 15 players I found that best match the player. For the NFL Best Rank average, generally anything in the teens and lower is pretty good. Anything over 25 is not a good sign.
It’s a little disconcerting to find Megatron’s average Best Rank lower than Golden Tate’s, despite the ridiculous .55 DR. But if I were controlling for Physical Score here, that would easily give Mega the edge in rankings. It’s worth noting that the DRs of Demaryius, Dez, and Calvin are so off-the-charts good, that the results of a comp analysis don’t actually look all that similar to those players. Further proof that outlier DRs are something to pay attention to.
I don’t love that Meachem is a few spots ahead of A.J. Green, but when it comes down to it, Green just didn’t record dominant stats from a DR perspective. I think my model gets Percy a little “wrong” because his DR was relatively low on the normal receiving metrics, but he also accounted for a good share of Florida’s rushing yards as well. We may have to adjust our normal models for these hybrid-type players like Harvin, Cobb, and now potentially Tavon Austin and *gasp* Cordarrelle Patterson. Lastly, I am encouraged by the fact that it red-flags Greg Little for his low YPR, despite his relatively solid DR, slapping him with almost a 30 for average best finish. It’s certainly not perfect, but what model is?
On to Mark Harrison’s comps. I’ll take a look at Harrison first, since he seems to be the better physical talent. What story do his numbers at Rutgers tell us?
|Player||Year||School||Rec||Yds||TD||YPR||DR||NFL Best Rank|
|Harrison Comp Avg||61.7||811.9||7.5||13.3||0.24||35.8|
|Brian Hartline||2007||Ohio State||52||694||6||13.3||0.25||35|
|Josh Morgan||2006||Virginia Tech||33||448||4||13.6||0.25||52|
|Early Doucet||2006||Louisiana State||59||772||8||13.1||0.25||45|
|Titus Young||2009||Boise State||79||1041||10||13.2||0.26||46|
|Josh Morgan||2007||Virginia Tech||46||552||5||12||0.25||52|
|Keary Colbert||2003||Southern California||69||1013||9||14.7||0.25||37|
|Anthony Gonzalez||2006||Ohio State||51||734||8||14.4||0.26||45|
Well, that’s just awful. Worse than all but DHB in the table above. All hope may not lost, however. Perhaps we could attribute Harrison’s bad YPR and DR to poor offensive/quarterback play at Rutgers? When measured against all college seasons from 2000 to 2012 here’s how it ranked:
|Attempts Per Game||30.2||43rd|
|Yards Per Game||208.5||42nd|
|Yards Per Attempt||6.9||45th|
No luck there. The Rutgers passing offense was only slightly worse than the historical average, and by no means was it historically bad.
From the scouting department, the SeahawksDraftBlog.com extolls his measurables and ventures some guesses about why his production may not match up with his potential. Harrison does get some token good praise for being “powerful” and a “Big-play and red-zone threat” here at NFL.com which does seem to be borne out by his Red Zone TD Rate:
According to withthefirstpick.com “Harrison has shown that he is more than capable of shedding a tackler and taking it the distance.” But even with the praise, you do still get some caution in these reports about his lack of production, especially given his physical attributes.
Even my favorite scouting source, Matt Waldman’s Rookie Scouting Portfolio, rates Harrison higher than you might think, and identifies him in his “underrated” section. Still, the stats paint a pretty bleak picture so it’s hard to endorse Harrison as a great flier pick, especially when you contrast his stats against the guy right next to him on that Physical Score list, Marcus Davis. More on him in Part II…