Three takeaways: Michigan rides rushing attack, defense to 24-15 win
- By Brock Heilig
- The Den
- 0 Replies
Maize&BlueReview - Three takeaways: Michigan rides rushing attack, defense to 24-15 win
Three takeaways from Michigan's impressive win in Happy Valley.
michigan.rivals.com
Explosive runs
- By TimSullivan
- The Den
- 13 Replies
I know a lack of explosive runs has been a pretty big talking point around here. Thought I'd mess around with some play-by-play data to see what it* says. Here's the number of runs by yardage gained for U-M's top two backs (for clarity's sake, the "20" bucket includes all runs of 20 or greater. You can basically just consider that far right edge "true explosives" more than 20-yard runs on their own merits).
Blake Corum:
Not a ton different from last year. He's getting tackled for loss less (we will also see this with Edwards), but he's also ripping off fewer medium-length chunks in that 5-8ish range. Beyond that, there's a little blip in runs for approximately first down yardage, but the rest of the way, it's mostly the same as last year with - 2023 actually better at some distances - until you get to the "true explosives," which are basically halved from last year.
As for Donovan Edwards...
Same story in terms of fewer tackles for loss, but an even steeper dropoff between "rushes for a small gain" and "rushes for a chunk." He was actually quite impressive at those chunk runs last year, and has been what feels (and I guess I could theoretically compare him to national averages to confirm this) specifically bad at it; that matches the eye test, at least. The great issue for Edwards this year - and I'm well aware I'm not breaking new ground here, just confirming it - is that his explosives of any type have fallen off a cliff, and he has literally zero runs longer than 14 yards. On the bright side... 20% of his true explosives came in the final regular-season game last year, so the very nature of rare events is that one or two of them changes the statistical picture in a major way.
Somebody who's way better than I am at calculus (which is to say: at all good at calculus) can decide whether the meaningful area under the curve is improved without the negative runs from a year ago, and whether the improvement there at low yardage marks makes up for a lack of explosives from both players. I will say that a slightly better - against all odds - offensive line than last year's at getting that initial push is a factor here, I haven't watched in enough depth to know if the issues at ripping off the chunks are in part their responsibility, or just the fault of the backs (or simple statistical randomness, though at least in Edwards's case, that seems unlikely to explain the whole thing).
I also don't know if Edwards's lack of explosive running is made up for by the fact that he's been a more frequently-used receiver (already 33% more receptions in nine games than he had in 11 last year), though even that might be mitigated by the fact that he's not averaging a whole lot on catches in comparison to last year, either.
* Use "data" as a singular noun to trigger your local statistician challenge
You may now return to your regularly-scheduled discussion of ShondaVerse television drama "Scandal" (I assume that's the scandal I keep hearing about, right?).
Blake Corum:
Not a ton different from last year. He's getting tackled for loss less (we will also see this with Edwards), but he's also ripping off fewer medium-length chunks in that 5-8ish range. Beyond that, there's a little blip in runs for approximately first down yardage, but the rest of the way, it's mostly the same as last year with - 2023 actually better at some distances - until you get to the "true explosives," which are basically halved from last year.
As for Donovan Edwards...
Same story in terms of fewer tackles for loss, but an even steeper dropoff between "rushes for a small gain" and "rushes for a chunk." He was actually quite impressive at those chunk runs last year, and has been what feels (and I guess I could theoretically compare him to national averages to confirm this) specifically bad at it; that matches the eye test, at least. The great issue for Edwards this year - and I'm well aware I'm not breaking new ground here, just confirming it - is that his explosives of any type have fallen off a cliff, and he has literally zero runs longer than 14 yards. On the bright side... 20% of his true explosives came in the final regular-season game last year, so the very nature of rare events is that one or two of them changes the statistical picture in a major way.
Somebody who's way better than I am at calculus (which is to say: at all good at calculus) can decide whether the meaningful area under the curve is improved without the negative runs from a year ago, and whether the improvement there at low yardage marks makes up for a lack of explosives from both players. I will say that a slightly better - against all odds - offensive line than last year's at getting that initial push is a factor here, I haven't watched in enough depth to know if the issues at ripping off the chunks are in part their responsibility, or just the fault of the backs (or simple statistical randomness, though at least in Edwards's case, that seems unlikely to explain the whole thing).
I also don't know if Edwards's lack of explosive running is made up for by the fact that he's been a more frequently-used receiver (already 33% more receptions in nine games than he had in 11 last year), though even that might be mitigated by the fact that he's not averaging a whole lot on catches in comparison to last year, either.
* Use "data" as a singular noun to trigger your local statistician challenge
You may now return to your regularly-scheduled discussion of ShondaVerse television drama "Scandal" (I assume that's the scandal I keep hearing about, right?).
