Jimmystats: Fun with Receiver Targeting Data 2020
Not enough of you know about collegefootballdata.com, which pulls an entire season's play by play data. With not too much extra work you can download this and find out all kinds of things that we used to lazily rely on Bill Connelly to provide. Like for example, returning Big Ten receivers who had 20+ targets sorted by yards per target when you include pass interference:
[Discussed after THE JUMP]
--------------------------------
What's This Again?
Targets add another dimension to the general data we can pull on receivers. They're pulled from the play-by-play that gets reported to the NCAA and published on ESPN among other places. For example Jehu Chesson just got a target here:
Darboh did not get a target, but we can also credit the receiver for yards gained from pass interference. That's usually not been much of a game-changer but when Nico Collins is in your chart it starts to get meaningful. Some of that data don't match the stats exactly. That could be statistical calculation errors in reporting, or more likely some play that I missed when trying to clean up the data.
A throwaway doesn't result in a target. Most uncatchable balls look like this in the box score.
Are some scorers completely different than others in what they'll call a target? Absolutely. I even have some ideas to explain some home/road splits in that department, but that's a study for another day.
Catch Rate is simply receptions/targets. It's not a great stat unless you can give it a lot of context. A downfield receiver with a half-decent quarterback who's bringing in 67% of his passes is really good. But a slot receiver who's mostly used near the line of scrimmage should be in the 70s or 80s, and tight ends and running backs will get up there. Also #1 receivers have their catch rates depressed by drawing the opponent's best coverage.
Yards per Target is the best stat we can get out of simple target data. Again it's dependent on a lot of variables, e.g. the quarterback's accuracy, but I find it especially useful in comparing teammates because it will betray their roles. See if you can spot Collins in the next chart by just looking down the YPT column.
Target Rate is the percentage of a team's targets that went to said player. The top two players in target rate in the Big Ten last year were Minnesota's two golden receivers, #1 Tyler Johnson (38%) and #2 Rashod Bateman (30%). Those guys are awesome but that's also a statistic about the Gophers' offense, which was way more inside-run oriented than people who only watch the highlights realize, with passes to the outside guys used as RPOs and line checks.
Any player getting over a quarter of his team's targets in any scenario is doing work. Any Nico Collins getting less than 20% of his team's targets is criminal.
How Did Michigan's Receivers Fare? And What Did PI Do for Nico?
Here are the Michigan guys who got 10 or more targets last year, with returning players bolded:
Base Stats | Include Pass Interference | |||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Player | Rec | Tar | Yds | CRt | YPT | TarRt | TDs | PI | Yds | Tar | Yds | CRt | YPT | |
Ronnie Bell | 47 | 77 | 754 | 61% | 9.8 | 20% | 1 | - | - | 77 | 754 | 61% | 9.8 | |
Nico Collins | 37 | 60 | 729 | 62% | 12.2 | 18% | 7 | 8 | 104 | 68 | 833 | 66% | 12.3 | |
Donovan Peoples-Jones | 34 | 56 | 453 | 61% | 8.1 | 15% | 6 | 2 | 16 | 58 | 469 | 62% | 8.1 | |
Nick Eubanks | 26 | 42 | 254 | 62% | 6.0 | 11% | 4 | - | - | 42 | 254 | 62% | 6.0 | |
Tarik Black | 25 | 42 | 308 | 60% | 7.3 | 12% | 1 | 2 | 24 | 44 | 332 | 61% | 7.5 | |
Sean McKeon | 13 | 28 | 235 | 46% | 8.4 | 7% | 2 | - | - | 28 | 235 | 46% | 8.4 | |
Mike Sainristil | 8 | 15 | 145 | 53% | 9.7 | 4% | 1 | - | - | 15 | 145 | 53% | 9.7 | |
Zach Charbonnet | 8 | 13 | 30 | 62% | 2.3 | 3% | 0 | - | - | 13 | 30 | 62% | 2.3 | |
Giles Jackson | 9 | 10 | 142 | 90% | 14.2 | 3% | 1 | - | - | 10 | 142 | 90% | 14.2 |
Those eight(!) penalties Nico picked up were over a hundred extra yards that didn't make it into the raw data, as well as 4 percentage points of catch rate. It didn't do much to his yards per target because it was already over 12(!!) That was with Shea Patterson's worse-than-average accuracy last year. It's also insanely low usage for a guy of his caliber.
Unbalanced schedules also made some of those numbers bigger than they had to be. However I can show you Nico against the rest of the conference receivers who had at least 30 targets (including PI) against Power 5 opponents:
Player | Team | Tar | Rec | Yds | TD | PI | PI-Yds | CRt(T) | YPT(T) | TarRt(T) | TmRk |
---|---|---|---|---|---|---|---|---|---|---|---|
Rashod Bateman | Minnesota | 75 | 50 | 1062 | 10 | 1 | 15 | 67% | 14.2 | 30% | 2 |
JD Spielman | Nebraska | 64 | 43 | 793 | 5 | 1 | 14 | 68% | 12.4 | 26% | 1 |
Nico Collins | Michigan | 54 | 32 | 648 | 6 | 6 | 85 | 63% | 12.2 | 19% | 3 |
Donavan Hale | Indiana | 27 | 19 | 326 | 3 | 4 | 52 | 74% | 12.2 | 8% | 5 |
Tyler Johnson | Minnesota | 95 | 72 | 1107 | 10 | 3 | 27 | 77% | 11.6 | 39% | 1 |
Chris Olave | Ohio State | 61 | 39 | 691 | 9 | - | - | 64% | 11.3 | 20% | 1 |
Jahan Dotson | Penn State | 33 | 20 | 368 | 2 | - | - | 61% | 11.2 | 12% | 3 |
Josh Imatorbhebhe | Illinois | 42 | 23 | 463 | 6 | - | - | 55% | 11.0 | 18% | 1 |
Binjimen Victor | Ohio State | 38 | 25 | 412 | 4 | 1 | 15 | 67% | 10.9 | 13% | 3 |
Jake Ferguson | Wisconsin | 31 | 26 | 334 | 1 | 1 | 11 | 84% | 10.8 | 13% | 2 |
Not to diminish them because they're all dynamite, but Bateman, Johnson, and Spielman played in the West, against a significantly lower quality of defense with the exception of Wisconsin, whom Nico also played. Hale was his team's 5th target. You know how frightening Chris Olave is; against Power 5 competition Collins was a yard per target better, with the same success rate on similar usage.
What besides THROW IT TO NICO is in here?
Of the guys returning only Ronnie Bell's got a significant dataset. I can just continue the above chart because he was just two spots down from Ferguson. Again, this is versus Power 5 Opponents Only:
Rk | Player | Team | Tar | Rec | Yds | TD | PI | PI-Yds | CRt(T) | YPT(T) | TarRt(T) | TmRk |
---|---|---|---|---|---|---|---|---|---|---|---|---|
11 | Ty Fryfogle | Indiana | 49 | 34 | 516 | 3 | 4 | 45 | 72% | 10.6 | 14% | 4 |
12 | Ronnie Bell | Michigan | 62 | 38 | 655 | 1 | 61% | 10.6 | 19% | 1 | ||
13 | KJ Hamler | Penn State | 77 | 51 | 811 | 8 | 66% | 10.5 | 27% | 1 | ||
14 | Austin Mack | Ohio State | 31 | 22 | 316 | 3 | 2 | 26 | 73% | 10.4 | 11% | 5 |
15 | Cody White | Michigan State | 80 | 58 | 812 | 5 | 2 | 30 | 73% | 10.3 | 23% | 1 |
16 | Whop Philyor | Indiana | 87 | 60 | 892 | 4 | 3 | 31 | 70% | 10.3 | 23% | 1 |
17 | Quintez Cephus | Wisconsin | 70 | 47 | 704 | 5 | 2 | 24 | 68% | 10.1 | 29% | 1 |
18 | Ihmir Smith-Marsette | Iowa | 65 | 36 | 642 | 4 | 1 | 15 | 56% | 10.0 | 20% | 2 |
19 | Garrett Wilson | Ohio State | 38 | 25 | 369 | 3 | 1 | 15 | 67% | 9.8 | 13% | 4 |
20 | Dontay Demus Jr. | Maryland | 63 | 37 | 567 | 6 | 3 | 42 | 61% | 9.2 | 27% | 1 |
Playing a similar role to KJ Hamler (slot and flanker), he wasn't Speedy Eaglet, but Bell was Speedy Eaglet's efficiency on less usage. He compares favorably to Indiana's slot bug Whop Philyor and Iowa's return/speed threat Ihmir Smith-Marsette. Again you have to use some context for the catch rate. According to our charting Bell had some extreme bad luck in the way of the kinds of targets that were close enough to get charged a target but not close enough that he was likely to come down with it. I'll steal the chart from HTTV's receivers page so long as you all promise to read the rest:
Everybody got some uncatchables—Black the most—but Bell also had 15 throws that asked him to make a circus catch. He was only 20% on those, but what do you expect a guy to catch?
I've published all these data on a google sheet if you'd like to poke around yourself.
So what you're saying is, throw to Nico always? Cuz they need to at least do it far more frequently than they have the past two seasons
Even if the starting QB becomes like Matthew Stafford when he had Calvin Johnson that might not be such a bad thing.
DPJ but not Nico leaving for the NFL is the great mystery of the last season to me. DPJ was clearly not ready and would have at least gone where he did (6th round was it?) next year. Nico however did not get the hype he deserved somehow, I thought he would be gone.
I think that decision was based on his utilization within the offense and relationship with the staff, moreso than production.
Without having any insider info or perspective, I got the impression DPJ was not happy with his usage in the offense and had a luke warm relationship with the staff.
I will say, I think he was criminally under used in the offense as a dynamic player that was shoehorned into your basic outside receiver role. Not putting it all on Harbaugh and the staff, I thought at times he underperformed given his level of talent, but he had the skills to be a focal point of the offense and for whatever reason he didn't see as much opportunity as he probably expected at Michigan. I can understand why he decided to move on.
DPJ was always a better athlete than receiver. He and Shea didn’t seem to click.
DPJ was always a better athlete than receiver. He and Shea didn’t seem to click.
So Patterson threw 129 passes that were uncatchable or needed a circus catch for a completion! Wow. 75 the rec had no chance, I know a few were throw away passes but not many. SMH..
That's 6 uncatchable balls and 4 more requiring a circus catch each game. And he wasn't even a running threat most of the year. Did he have dirt on Harbaugh or something? Its gonna feel weird to have competent qb play again one day
McCaffrey’s severe concussion ended alternatives to Patterson. Milton just a red shirt freshman with little experience. Harbaugh forced to stay with & support Patterson.
That data screams we had way too many mouths to feed last year. That WR room must have been a mess, attitude-wise. Gattis is a hard-ass compared to Pep and he probably pissed off some guys. Heck, Black transferred and DPJ went to the NFL way too early. Those guys probably never bought into the new offense ... look at the lackluster catch-rates, especially DPJ's performance in the OSU game. In hindsight, his head and heart was probably already gone at that point.
But this is the year to take those 100 DPJ/Black targets and re-distribute them mostly to Nico to get him up to ~120 targets. Bell at 80 targets feels right, but we need to see way more of Giles Jackson with the ball in his hands ... that true speed in space.
I think you hit the nail on the head assessing this particular intangible. We may actually see some addition by subtraction in terms of production now that there is a clear pecking order in the WR room:
1. Nico
2. Bell
3. The sophomore or freshman that earns the most targets (and there are some good ones)
On paper it's great to have talent in the same room like Nico/DPJ/Black. The reality more times than not however, is each of those guys believe they're primary target number one, and there isn't enough footballs to go around to keep them all engaged and playing with enthusiasm.
It's true of any organization, when the personnel accepts and understands their role, they are much more likely to give their full effort and contribute to morale rather than undermine it.
Not a shot a Black and DPJ at all, they seemed like good teammates and they're talented players. Sometimes the dynamic of a team can suffer however when there are conflicting egos in the room.
And WR have the biggest egos of all...TO, Ocho Cinco, Randy Moss, OBJ, Antonio Brown.....
LSU had more talent than we did in the receiver world, and it didn't seem to bother them at all.
Can't argue LSU has elite pass catchers (they always do), they also had a passer throw for the most prolific season in college football history. Michigan had... not that.
My point is only that when you have three high-profile players who think they should be the focal point of the passing game, paired with lackluster QB performance, you're going to get some consternation in the locker room which can impact performance on the field. When players accept and understand their role on the team, you will receive better effort and buy-in towards the team's common goal.
Seth, I am still working through HTTV so this information may already be included, but do you see anything schematically that Michigan can do better to get Nico more targets? I understand that last year's QB play left a lot to be desired from an accuracy and willingness to stay in the pocket standpoint but what can be done to mitigate that?
Cheers for more arm punts to Collins in our future.
Schematically they did what they were supposed to except they used the other guys as RPO targets because Nico was more likely to draw a double. It's more that they chose not to throw it to him when he was singled or wide open even.
The image I have in my head is of Patterson settling back, pocket clean as clean as your favorite breakfast bowl out of the dishwasher, staring at Collins running downfield with a panicked CB scrambling to give chase, and he pulls it down and dumps it off. You think aw man the safety must have gone back with them.
Then replay shows the safety on the other side of the field and you ask the gods why.
Thanks, that painfully tracks with my observations from last year. Patterson's hesitance to throw into single coverage in games where the defense was getting torched was a mystery to me. Running from ghosts and ignoring open NFL wide outs are hopefully not contagious traits.
Sainristil - 3 for 3 on circus caches! I be like damn.
Why is yards per target a better metric than yards per reception?
I get that actually catching a ball gets picked up in the former, but it seems like QB inaccuracy would offset that. You could just use yards per reception and then look at drops. Or maybe build an adjusted yards per reception metric with drops built on? Something like YPR * (1 - Drop Rate)? Drop Rate could be drops / receptions. If that were 10% - which I'm guessing is pretty high? -, you'd take more than yard off someone's YPR, which seems significant. And if the guy drops a lot of balls but still has a high adjusted YPR, then it means he's still worth throwing to, because the drops are offset by the efficiency.
That would take the QB out of it to a further extent. But maybe drop stats aren't readily available, though I'm pretty sure most teams track that (whether they publish it is probably the question).
Looking at receivers across different teams I agree ...but if I was an OC looking at which receivers needed more PT, yards per target would mean more (and in that case the QB accuracy is relatively uniform across all receivers)
" he wasn't Speedy Eaglet, but Bell was Speedy Eaglet's efficiency on less usage "
I'm not sure how you're defining "efficiency"... but the only way this statement makes sense is if you're only looking at yards per target, and ignoring everything else.
Efficiency is usually defined as output/input for some system. Or put in different terms, production/usage. It seems an interesting measure for discussing individual player quality. So I can think of a number of measures of efficiency:
- Receiving Yards/Target
- Receiving Yards/Catch
- Total Yards/touch (for including runs and returns)
- Receiving Touchdowns/target
- Receiving Touchdowns/catch
- Total Touchdowns/touch
Those are some that I can think of off of the top of my head. The first one, (yards/target) is a reasonable comparison, as it includes things like drops (If a receiver drops it often, I wouldn't think of them as terribly efficient). A potential downside is that it values yards, which doesn't directly show up on the scoreboard.
SP speed ACE
Another excellent post, but I can't view the first chart. Don't know if anyone else is having that problem
Some interesting items for this season (if it happens):
Collins is a bonafide #1 and should get a quarter or more of Michigan's targets. Bell is a guy that might be best operating out of the slot. Jackson is a home run threat and he and Sainristil bring some versatility to Michigan's spread game.
Eubanks is interesting. More of a medium range guy than a downfield threat as a TE.
I really hope letting WRs run wide open down the field and not have the QB not see him or miss him by 10 feet is over. I hope that was just a Shea thing. I mean, when the crowd goes “ooohhhh” because the guy is so open, and he doesn’t throw to him it’s beyond frustrating.
Comments