I have mentioned this in several comments and got quite a bit of doubt and some scorn, so I'm laying the full argument out in a Fanpost.
First, we need to understand the concept of a counting stat. A counting stat is a statistic that is incremented every time something happens. In the case of Jason Witten, the primary argument that 2012 was a great year is that he caught more passes than any tight end in history in one season. Total receptions is a counting stat. Every time he caught a pass, the stat was incremented, and this particular stat was incremented 110 times.
Here are his counting stats for 2012:
Targets: 150 (8th overall, 1st as TE)
Receptions: 110 (5th overall, 1st as TE)
Yards: 1036 (20th overall, 1st as TE)
Touchdowns: 3 (T-81st overall, T-23rd as TE)
First Downs: 56 (14th overall, 3rd TE)
Great stuff, right? No sign of a decline, right? The only mediocre stat here is touchdowns. What the heck am I thinking?
The problem is, and I've said this before, and I'll say it again: *counting stats lie*.
Why? Because counting stats are as much of a function of opportunity as they are of production. Jason Witten was targeted 150 times. 150!!!!
The next closest TE was Jimmy Graham with 135. That's a big difference. Had Witten been targeted 135 times he would have ended up with 99 receptions for something like 925 yards. That would be a very good year, but not an historic one. Please note that 135 would still make him tied with the lead in targets at TE.
So the second thing to look at are formula stats. Basically, with these you use the counting stats and create a formula with them to better understand efficiency and production. Unfortunately, this is where we can see the huge dropoff from Jason Witten in 2012.
Here are his yards per reception and his yards per attempt for his career (the NFL did not track targets in 2003, 2004, 2005, hence the xxx on those lines):
2012: 9.4 / 6.93
2011: 11.9 / 8.05
2010: 10.7 / 7.83
2009: 11.0 / 8.24
2008: 11.8 / 7.87
2007: 11.9 / 8.06
2006: 11.8 / 8.20
2005: 11.5 / xxx
2004: 11.3 / xxx
2003: 9.9 / xxx
Two things to note. First, please note how remarkably consistent he was from 2004 to 2011. These are fantastic and fantastically consistent numbers. This is why we all absolutely think he's one of the best TEs ever to play. By the way, I totally agree with that assessment. He *is* a Hall of Famer.
However, the second thing to note is the dropoff for 2012. His yards per catch is by far the lowest of his career, His yards per attempt is by far the lowest of his career. These things are, as Nishka would say, *facts*.
Here is another fact. Jason Witten was 30 going into 2012.
Here is a third fact. Tight Ends have a smaller peak than many other positions. Jonathan Bales has studied this quite a bit and here is a graph showing his research. What this chart shows is that TEs tend to take a while to become good and tend to drop off quickly.
So, I'm going to blow your mind with a stupid but interesting comparison. Which would you prefer:
TE A: 150 targets, 110 receptions, 1039 yards, 3 TDs, 56 first downs, 9.4 yards per catch, 6.94 yards per target
TE B: 150 targets, 92 receptions, 1045 yards, 7 TDs, 58 first downs, 11.4 yards per catch, 6.96 yards per target
Basically you're trading catch percentage for touchdowns. TE A is Jason Witten, of course. TE B is Martellus Bennett's year pro-rated to 150 targets. Obviously Marty B has mediocre hands, but would have had essentially the same number of yards, first downs, and scored more TDs.
I know, that's a stupid little exercise but think about this for a second. Marty B was almost exactly as efficient as Jason Witten was last year. Does this not sound like a dropoff from Jason Witten?
*Could* the dropoff be due to the lacerated spleen? Possibly, though if it did affect him, it was only in the first three games that there's any real sign of that and this is only a small portion of his total stats.
*Could* the dropoff be due to the offensive line. Sure, it could be. Hard to quantify.
*Could* he have the best year of his career in 2013? Sure. In 2014? Sure. Heck, he *could* have the best year of his career in 2019. That definitely is possible. But it is not probable.
So which do you want the Dallas Cowboys front office to do? *Hope* that Jason Witten's dropoff in efficiency in 2012 was an aberration because of the spleen and the offensive line and ignore it? That he is immune to the normal dropoff in efficiency that TEs experience at exactly this point in their career? That he is more like Tony Gonzalez (who has had some dropoff, but not much) as opposed to Antonio Gates (who has seen quite a bit of dropoff since his age 30 year).
Or would you prefer that they be proactive and get Witten's replacement ready to go? Many BtBers screamed bloody murder because we didn't get a successor to Ratliff. Many have been begging the Boys to get a QB because Tony's not getting any younger. I'm also worried about Ware, who's been hit with nagging injuries and getting up there himself. All of you are right in wanting the front office to have a plan for their inevitable decline. Why is Witten any different?
Look, we are *all* rooting for him. Believe me, I *hope* he ages like Tony Gonzalez. However, the combination of the fact that he did drop off in his age 30 year, the fact that he is now 31, and the fact that TEs tend to drop off quickly from 30 on down suggest to me that he is on the downward slope of his career.
I wish our Boys didn't age. Imagine a DLine of Martin, White, Lilly, and Ware with Haley and Rat and Too Tall and Andrie on the bench. I'd start TD over Emmitt. Roger over Troy. And yes, I'd start Witten over Novacek and Cosbie and the perennially underrated Billy Joe.
But sadly, though I can but dream of Ware stunting around Lilly, that is just fantasy.
OCC made a comment below and it is not letting me reply in that thread. Besides, this is information that should have been in the original post so I am adding it as an addendum.
Basically, OCC says that the dropoff is a sign of a difference in offensive philosophy, not in Witten's dropoff. He cites a post by Rabble that has a lot of great info, but I don't think Rabble's point explains the whole of Witten's dropoff.
I know, heresy that I'm debating with both the OCC and the Rabble :D
He's absolutely right that we can't take Witten's play in isolation. Also, he may absolutely be able to trump the statistics that I have with better ones. I do not have play by play data for Witten for his career. There are often more and better stats, but I have to go with the ones I have.
What I do have is the percent of the passes that were thrown to Witten that were more than 15 yards in the air and his yards after catch. I also have Tony's overall Deep% plus his normal stats. I will list these out since 2006, which is when most of these stats were tracked.
Here they are: (Witten's Deep% / Witten's YAC per catch / Tony's YPA / Tony's YPC / Tony's Overall Deep%)
2012: 12.9% / 2.8 / 7.6 / 11.5 / 19.3%
2011: 15.4% / 4.5 / 8.0 / 12.1 / 19.3%
2010: 9.4% / 4.2 / 7.5 / 10.8 / 15.5%
2009: 11.3% / 4.3 / 8.2 / 12.9 / 15.5%
2008: 17.9% / 4.4 / 7.7 / 12.5 / 21.3%
2007: 13.8% / 4.1 / 8.1 / 12.6 / 19.6%
2006: 17.1% / 4.1 / 8.6 / 13.2 / 27.3%
Looking at these numbers we can see three things. First, other than 2009 and 2010, Tony throws deep about 20% of the time. 19.3, 19.3, 21.3, 19.6, and then the outlier of 27.3.
Second, they might have thrown a little less to Witten than their normal average for his career, but some quick and dirty calculations before work puts this at about 1%. Significant, but not huge.
Third, they attempted exactly the same percentage of deep passes in 2011 as they did in 2012. What I think is that they didn't change their offensive philosophy for the year as Rabble contends, rather they simply executed worse than in 2011. Clearly they completed less as a team, as shown by the drop in YPA and YPC.
I think, by the way, that this is true of the entire team. The problem here is that we can't know how much of the dropoff was caused by the OLine, by Romo, and by Witten himself from these numbers.
Fourth, and this is the problem. The number of deep attempts has varied significantly over the past 7 years. Romo's overall productivity has varied significantly. The team's game plan has clearly gone up and down. Many are the variables that have changed.
What hasn't varied significantly are Jason's yards per catch, yards per attempt, and YAC per catch. Whether they threw deep to him at his average rate (which 2007 essentially happens to be the average), or his highest rate (2008), or his lowest rate with Kitna (2010), his average production per catch and target was essentially the same.
Whether he caught the ball deep or short, his YAC was by far the worst of his career. By the way, you add 1.2 yards per reception of YAC and put it at 4.0, which would still be his worst since 2006, and you add 132 overall yards, which would put his YPC at 10.6 and YPA at 7.8. Low by his standards, but terribly out of the norm.
What OCC's post prompted me to see is that the primary dropoff is not in depth of targets and shifting of offensive philosophy, it is a dropoff in YAC per reception.
Could this be influenced by the OLine's struggles? Yes, but he has been remarkably consistent up to this point. 2012 is the first year his production changed. Why 2012 and why not any other year? Could it be a coincidence that his production changed in his year 30 campaign? Absolutely. But could it not? Sadly yes.
Again, OCC, if you have better numbers, please let me know. I'm perfectly willing to be proven wrong. You certainly inspired me to dig deeper and that's always a good thing.