Great work.
So, is it possible that ~50% of TJ's targets have resulted in either a drop or a PBU? I still have a hard time believing that.
I'd bet that the biggest part of that number is going to be poorly-thrown balls, rather than drops and PBUs. But it certainly appears that there are also a number a throwaways in that number that most of us would agree are not targets. If TJ's getting credited with just one extra target per game, he jumps from 51.7% to 60.8%. For what it's worth, NFL WRs (for whom target data are much more readily available) range between 45-75% catch rate and my napkin math shows that number to be heavily skewed by QB accuracy.
I broke out the numbers a bit on the spreadsheet just to play around with them a little. Assuming that the two top performers on each team tend to be the two outside WRs and the next two tend to be TEs and slot WRs, I wondered if maybe we'd see a statistical difference in catch rate among the four groups. The results are mixed:
63.59583952
62.37904571
62.59317822
63.99089098
Obviously, there are a number of highly relevant confounders that we can't really control for there: quality of defenders, relative ability of each category of receiver, etc. Focusing on Yards Per Target can serve as a bit of a dirty control for route length, though, as YPT = Catch % * YPC. So, to the extent that YPC is a suitable proxy for route length, YPT
is Catch % adjusted for route length. Those numbers:
8.408282277
7.996712194
7.50051103
7.673375967
Among top-25 schools:
9.169130466
9.537816189
8.219752255
8.964255767
TJ actually does slightly worse once YPC is controlled for (~75% of mean under YPT vs. ~81% of mean under Catch %). DaVaris and Chris Brown do substantially better (Brown obviously gets the benefit of a single long catch on an exceedingly low number of targets).
At the end of the day, I think what we're really seeing here is TJ suffering from a QB who hasn't been super-accurate when throwing to him. Makes DaVaris's YPT all the more impressive. The Passing S&P+ masks the issue by ranking us 10th. It's important to keep in mind, though, that this is a defense-adjusted value, where YPT and Catch % aren't. So, while it's the correct stat to use in the overall model, it's not a good reference point to perform raw comparisons of Catch % or YPT. For that, unadjusted Passing S&P would be far better (although it's really just not an appropriate statistic for this particular purpose; even raw passer rating data would be more useful). ND drops from 10th in adjusted S&P to 65th in unadjusted.