A week after big upsets of Stanford and Ohio State, the USC and Virginia Tech themselves fell to less-talented opponents. Georgia also fell to South Carolina after pummeling Clemson in the opener.
27 May 2014
by Rivers McCown
So, it was rather peculiar when the Texans retained him with a three-year, $11.25 million contract. The Texans only guaranteed $4.5 million in the deal, but the fact that he had such poor advanced statistics stood out. He was the only Texans free agent of note to re-sign. So, why him? Why not someone like Antonio Smith, who had proven to be a worthwhile piece already?
When I went over Andre Johnson's season, quarterback play was a big factor in his eroded DVOA and DYAR figures. That shows up again on the tape here.
The question then, is how different Graham is on tape compared to the statistics. Graham is definitely not the 50th-best tight end in the NFL. But how much better is he?
I looked at where Graham lined up before each snap, and then gave an educated guess on whether he was facing man or zone coverage. This can be hard to tell at times, and I still don't think my eye for film matches up with the best in the industry, but I did my best. I don't come into this looking to make sweeping conclusions based on the data we use. In fact, I don't come into this column looking to have a final say on anything.
The target totals listed below don't quite match the total listed on our tight end stats page. This is because we did not consider certain plays, including:
Here are my observations:
"Behind the line of scrimmage" includes passes that traveled less than one yard. "Short" indicates passes that traveled between 1-15 yards beyond the line of scrimmage. "Deep" indicates passes that traveled 16 or more yards beyond the line of scrimmage. Red numbers show a Success Rate of 50 percent or higher. Black numbers show a Success Rate between 40-49 percent, and blue numbers show a Success Rate of 39 percent or lower.
1. How much "there" is really there?
Out of 90 total targets intended for Graham, I threw out 12 of them.
Graham was the target on two Hail Mary passes, five clear throwaways, defenders tipped two balls at the line of scrimmage, one miscommunication, one play where a defender tripped him, and one where a defender hit Case Keenum in his throwing motion. And a partridge in a pear tree.
You don't need to be a math major to estimate the effect this had on his DVOA. It drops an anvil on it when you consider the sample size. Eventually we're going to get data fast enough after Sunday games to filter this stuff out of DVOA, but for last year, 13.3 percent of his targets were uncatchable and counted against him. Because his quarterbacks played like ass.
Unfortunately, that might not change next year. Ryan Fitzpatrick did not inspire much confidence last year, and has inconsistent accuracy. Tom Savage is a fourth-round rookie drafted on tools. Two of the quarterbacks that threw passes at him are still here. Bill O'Brien's system should change some of the problems this team had last season. Can they hide garbage quarterback play?
"Slot" plays include all plays in which a receiver was inside of another receiver, regardless of how tight to the line he was. (For tight ends, if they are lined up at the LOS, they are counted as inline.) Some plays in the charting project include two or three "slot receivers" by this definition.
2. Responsibility shift
Let's split up Graham's season into two distinct sections.
Weeks 1-5: 15 DYAR, 2.9% DVOA, 22 targets, 71 percent catch rate
Weeks 6-14: -99 DYAR, -29.2% DVOA, 68 targets, 50 percent catch rate
You might think I've selected these to showcase how Graham did with Schaub and Keenum. Schaub fell to injury in Week 6, after all.
But no, another injury that happened in Week 5 is what I'd like you to notice here. The loss of Owen Daniels to IR. That elevated Graham up the depth chart, and it forced him to become a major target for a team that needed someone to step up.
Does that explain the DVOA discrepancy? No. All 12 targets I threw out happened after Graham became the No. 1 tight end. Let's estimate that his "true" DVOA was something closer to -10%. That's still not a desirable outcome for your No. 1 tight end, particularly in a Gary Kubiak offense. As you can see in the first chart, the Texans weren't exactly asking him to stretch the field. So, why exactly did Graham struggle in the new role?
Obviously we don't have the actual play calls. These charts are educated guesses and should not be considered gospel.
3. The eye test
What stands out about Graham's game is that he has no plus attribute to rely on.
His hands are solid. He's decent at contested receptions as long as the ball is there on time. (It often wasn't in 2013.) He gets off the snap well and can threaten the seam. He can play on or off the line. He was able to make a few catches last year after chip blocks.
Graham just lacks a real calling card as a player. There isn't a route he's particularly great at. He doesn't eat up zone coverage. (If he has a weakness as a receiver, it's that he had problems reading the right zone release.) He's not at all imposing as a physical specimen.
He doesn't have glaring flaws, like I've seen from a few players I've researched. He's not much of a blocker, but that's out of the bounds of what I'm looking at in this series.
4. JAG is (still) a re-run tonight
The Texans plan to use Graham as more of a "move" tight end this year. That'll at least lessen the blocking inadequacy, but I don't see that working as well as they seem to expect.
Much like Coby Fleener, Graham isn't a problem on a good offense. He's got enough talent to win when he isn't game planned against. The difference between the two is one of implied expectations versus explicit expectations. Fleener was a second-round pick. Scouts supposed he would be a quality cog in the passing game. The Colts have enough receiving talent to keep him from being a problem if he isn't.
The Texans are now paying Graham like a starting tight end. Only there is no Dwayne Allen on the Houston depth chart. The No. 3 receiver is whichever of DeVier Posey or Keshawn Martin you prefer. The No. 2 tight end is Ryan Griffin, with C.J. Fiedorowicz lurking. Neither of those guys offer more than Graham does in the passing game right now.
It's my belief that, based on the information we have, the Texans are wishcasting with Graham. To me, he grades out as the Joel Dreessen of the 2010s. He won't hold the team back or anything -- he's just a bit stretched as a No. 1 tight end.
5 comments, Last at 28 May 2014, 12:22pm by Rivers McCown