Dowdle is a High Variance Rusher

Actually...I just did the math.

The difference in their YPCs last year is actually a lot tighter than the 0.1 YPC reported on PFR (because Dowdle got rounded up and Warren got rounded down).

Dowdle's YPC is 1076/236 = 4.559 YPC.
Warren's YPC is 958/211 = 4.540 YPC.

The difference is about 0.02 YPC. Assumes I didn't screw up the math here.

So if Dowdle had 4.7 yards less across his 236 carries, he'd have Warren's YPC.

And if Warren had 4.2 yards more across his 211 carries, he'd have Dowdle's YPC.

So while it is technically true that Dowdle had a higher YPC in 2025, I think this tells us the difference is negligible.

Warren (54.5%) had a significantly higher success rate than Dowdle (47.9%). But dowdle did have a good success rate in 2024 (53.6%).

Hopefully the drop was more schematic and coming back to MM helps.

 

Hope Dowdle doesn't make me miss Gainwell. Gainwell was one of the few players who showed up every week in 2025. Dude was a true Steeler. Sucks that Bucs have him now. Wouldn't surprise me if Gainwell ripped off 150+ yards, ground and air.... on October 18th at Tampa.

 

Maybe this is being silly, but can you also post this with a log scale on the y-axis?

I think that would help us see the long tail better.

It will flatten out the differences around the mode though.

 

I really liked Gainwell's performance last year too.

And I agree re: consistent performance.

But I hope that it's Pittman that helps us forget about Gainwell (and the TEs). I don't think having a RB act as WR2 was a good way to run the offense last year.

 

It's available if you want to have a go at it:

https://nflreadpy.nflverse.com/

 

The last time our team MVP left and went to another team the following season was in 1970. Roy Jefferson was the 1969 MVP but being a vocal voice for the NFL players union had Chuck Noll ship him to the Colts after the season.

 

Aargh, you left me hanging. How did he do the following year?

 

Awesome!

Thanks

 

You are correct but that doesn’t fit a narrative so it may be dismissed

We are talking 3/4” different

The only place that would be significant is in KtCP’s hands

 

This just made me feel old

A cut and paste from your site:

A Python package for downloading NFL data from nflverse repositories. This is a Python port of the popular R package nflreadr, designed to provide easy access to NFL data with caching, progress tracking, and modern Python conventions.

 

I like the pretty colors

 

Only if the Steelers like him best

 

The second video is football in shorts so meh

What really jumped out at me in the first video is the downfield people who kept blocking no matter what

George please see above

 

In shorts

 

44 receptions, 749 yards and 7 TD's. And the Colts won the Superbowl that season. So Jefferson went from the 1-13 Steelers to winning it all with the Colts the next year.
Just to add we sent Jefferson to the Colts for Willie Richardson and a fourth round pick. Richardson didn't do much with the Steelers but the fourth round pick turned out to be Dwight White.

 

I know you like success rate, but it is subjective and partially dependent on more factors beyond the control of the back. Dowdle has taken on larger workloads than Warren the last two years, which makes maintaining the higher average per carry more difficult.

 

I think they are slightly different types of runners .. 1 is more of a inside guy who just hits the hole quicker and the other has a bit more wiggle and can make guys miss more in space...After they get going both of them have power to break tackles..IMO I just prefer 1 over the other for short yardage and inside work...

 

FWIW, I like the combination of YPC and success rates. By themselves, I don't like either as well as the combo. If I could only take one, I'd probably take YPC because it tells us something about explosive plays. But it "flattens" the importance of being good at short yardage.

The orange part of your comment is exactly why I like the combination of YPC and success rate. Taking on a high workload often means that a player gets more short yardage work. So we don't know if the drop we might see in YPC is because he was delivering fewer explosive plays, or if he got more work on 3rd and short (or whatever). If you had a guy who only got carries on 3rd and 3 or less and got to the 1st down marker (but never further) 80% of the time, he'd have a YPC of something like 2.5 (which looks terrible). But he'd have a success rate of 80%. Someone who just looked at his YPC would say "that guy sucks". But he has a role and he's very good at it.

We won't see real RBs with that stat line, but look at Mike Alstatt's last 3 years in the league. His YPCs here are brutal: 3.4, 2.4, 2.9. At first glance, this is a guy who shouldn't be in the league. But his success rates in those years were 56.7%, 55.9%, and 48.3%. So it's really that this is a guy who's usage was mostly short yardage carries. And he was good at them (falling off the year before he retired).

The use case argument was something I'd often see on the board I came from in favor of Najee. And I think it was probably more true for years 1-3. He got a lot of work. More or less kept up with the chains (a little behind, but high 40s succ%). Basically never delivered game breaking runs (low YPC). But he got a lot worse in in 2024. His YPC didn't really change much. But we saw that his success rate cratered. Maybe it's confirmation bias, but this is what I think I saw during games. Najee getting those short yardage carried, but getting stuffed more often than it felt like he did in the first 3 years.

I know that you don't like that they "arbitrarily" set definitions for keep up with the chains means depending on the down. But I don't think I've ever seen you argue for other definitions of what keeping up with the chains should look like. I wonder if anyone has ever written publicly about the effect of choosing different cutoffs for success on different downs?

I would imagine that you'd more or less agree with the success criteria on 3rd and 4th down? I think that most places will also count something like 5-6 yards as a successful run. So the old Cowher "run a draw on 3rd and 16" could still count as a success. I agree that it's not perfect. But having both metrics is much better than just having YPC. Because all YPC really tells us is how long / frequent were the long runs. Because all half-decent or better backs in the NFL are going to have the same "mode" of 3-4 yards.

Personally, I think a run on something like 2nd or 3rd and 8 should just be thrown out of the dataset. These are just wasted plays...and I hated how often Canada would call them. It's like "we didn't win on 1st down, now we're just going to punt". I think this is somewhat accounted for by having some max threshold (I think it's usually 5-6 yards)...but it's a new enough metric that I don't think it's standardized everywhere.

The thing I worry about re: Dowdle is that he had a similar workload in 2024 and 2024, but his succ% dropped by 6%. It's still a pretty new metric, so I don't think we know how consistent guys are year to year: Bell was all over the place; Warren is fairly consistent; Najee was consistent, fell off in year 4 and had a massive injury at the beginning of year 5. So I think that succ% + YPC is good at telling us what happened in a given season. But I don't know if it's predictive. The thing that I'm hopeful for with Dowdle is that (IIRC) his high succ% year was with McCarthy. Then MM got fired and Dowdle went to CAR and his succ% cratered. So hopefully this difference was about scheme / play calling more than ability. With that many variables changing and the difference in success rate, I think it's pretty interesting that his carries, YPC and total yardage all stayed basically the same.

Either way, I think Dowdle being in the fold will make us much more resiliant to injuries at RB (which I think are frequent). No more Fitz Toussant type guys if our RB1 went down. I wanted to see Johnson get a chance. But (1) the team knows a lot more about him than I do and (2) I think new regimes (or at least 1/2 new regimes) are more likely to give up on medium-pedigree guys. I think Johnson will look back on his NFL career and always question how life would have been different if he didn't fumble that kickoff (or if Gainwell didn't have what I imagine will be by far his best year in the league).

 

I agree.

It seems like they have different styles.

And deliver essentially the same results on average over a season.

I'm sure you're right that different guys will be better in different roles. Hopefully the coaching staff can figure out how to get the best out of both of them at the same time.

 

The guy is certainly capable of highlight-reel plays. Too bad his max is normally under 20 touches per game.

 

I prefer total yards rushing and yards per carry, as both measure what the back actually did without any subjectivity.

Of course, some folks here don't like that because it works against Warren. As much as I like Warren, I prefer more accurate evaluations.