Jump to content

FiveThirtyEight - Josh Allen Article


Recommended Posts

  • Replies 42
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

People are gonna ***** all over the article.  I think it's fine. FiveThirtyEight writers are the epitome of stat nerds.  They're not watching football games, analyzing film, or breaking down

just because someone can write doesn't mean they should

This is the paragraph I take issue with:   For quarterbacks, the stats that tend to be stable over time include passing efficiency from a clean pocket, passing efficiency with no play-action

Posted Images

Bills fans know the areas where Josh Allen passes the eye test.  He's never going to be a "traditional" QB.  Which is absolutely fine, given that model is quickly being debunked for good by several current outstanding QB's.  

 

We also know how much he's improved, and is still improving.  Stats prove it, and so do the team's results

 

We also love watching him.  He excites us more than any QB we've ever had on the team, arguably.  To be where he is after ~ 30 starts is incredible.  Don't think for one second that after watching the 2018 draft that you wouldn't be ecstatic with where we are now.

 

Everything else is just noise

  • Like (+1) 1
Link to post
Share on other sites
4 minutes ago, C.Biscuit97 said:

The Jets gave up 31 to a 49ers team that lost their qb, rb, was already playing without their star TE, and starting 2 wrs.  I think the only reason the Jets aren’t even lower is because the Bulls took their foot of the gas and 49ers are really banged up. They are really bad.

 

The 49ers built up a 21-3 lead with Garappolo in the first half.

 

I don't say the Jets have a great D, but it's not garbage. 

 

If you're looking for the bottom of the league, I say look elsewhere, but Time will Tell.

Link to post
Share on other sites
1 hour ago, Logic said:

People are gonna ***** all over the article. 

I think it's fine.

FiveThirtyEight writers are the epitome of stat nerds. 

They're not watching football games, analyzing film, or breaking down individual plays.

Very simply and specifically, they look at numbers -- a data set -- and make conclusions and predictions based off what the numbers are saying. Their conclusions are telling them Allen has improved. Their predictions are telling them his performance may regress to the mean in the coming weeks.

If people are truly that bothered by the predictions of stat nerds, I suggest they learn a new strategy: It's called "stop caring what people you've never heard of think about Josh Allen". I've been trying it out the past couple weeks, and it's working wonders. 

And stop even reading it.

 

Bolded: Was your first clue all the MVP talk going on here already? :)

Edited by CodeMonkey
  • Like (+1) 2
Link to post
Share on other sites
2 hours ago, IgotBILLStopay said:

clickable link

 

https://fivethirtyeight.com/features/is-josh-allen-actually-good-now/

 

Another negative nancy who confuses beliefs and preferences.

 

 

Everyone who has something negative to say about Allen uses the argument that the Dolphins and Jets suck.

 

First, this is early in the year with too small a sample. If the Bills did well, their opponents, by construction, have to suck. Cant use their bad performance when facing Allen to make the argument that Allen is not strong enough since they are weak teams.

 

Second, the Dolphins (who are  supposedly so bad) just walloped the Jaguars, who walloped the Colts ... so who is weak now?

 

 

They would have a small point if Josh just looked average or bad against them.  He didn't though.  He destroyed them.

  • Like (+1) 1
Link to post
Share on other sites
2 hours ago, eball said:

This is the paragraph I take issue with:

 

For quarterbacks, the stats that tend to be stable over time include passing efficiency from a clean pocket, passing efficiency with no play-action, completion percentage over expected and the share of throws that are off target or negatively graded. Each of these aspects of a QB’s performance is indicative of their true skill and is less affected by circumstances and luck. Stats that aren’t predictive of future performance — and that tend to be highly affected by circumstances and luck — include passing under pressure, passing outside the pocket and passing efficiency with play-action.

 

In what world does a QB get to operate without pressure or stay purely in the pocket?  And play-action is specifically designed to make a QB more successful.  The author says that because Josh is performing well in these "non-predictive" areas we shouldn't count on his improvement to continue?  That makes no sense to me.

I guess Lamar Jackson has a 0.0 completion percentage by this logic as the Ravens almost never throw without play action or keep him in the pocket. So, Lamar sucks too?

 

I love 538’s election models but this article has a lot of issues because it expects teams to always play to a Tom Brady style offense, which won’t always maximize the strengths of their QBs.

Link to post
Share on other sites
2 hours ago, Tortured Soul said:

Does anyone know the source of this statement in note 1? 

 

"Analytics staffers on NFL teams ranked the Bills among the top five most analytically advanced in the NFL."

i read this yesterday.. I'm going to assume this is where they got it from:

 

https://www.espn.com/nfl/story/_/id/29939438/2020-nfl-analytics-survey-which-teams-most-least-analytically-inclined

 

image.thumb.png.b77983a05f21a5aea7a06667633b838d.png

Edited by Nineforty
add image.
Link to post
Share on other sites

Color me unimpressed by this article.

 

The author says that some metrics are highly variable while others are more stable and predictive - but offers no evidence or analysis.  He goes on to note that "Allen's numbers are great in the unstable, less predictive metrics and middle in the road in their stable, more predictive counterparts."  He concludes that Allen is likely to return to his norm of 2018 and 2019.  

 

This analysis precludes the whole idea of progress.  For example, he argues that passing efficiency from a clean pockets is both stable and predictive.  I'm sure if you're looking at a QB in mid-career that's true.  But don't tell me that young QBs don't often get better at passing efficiency from a clean pocket over time.   The author, btw, never mentions whether or not Allen has made progress in the predictive metrics from 2018 to now.  He doesn't look to see if there's any upward trend.  If these metrics are truly predictive, wouldn't we expect an upward trend to continue?   

 

Furthermore, he makes the ridiculous argument that Allen achieved his statistical heights this year against weak opponents.  How - statistically speaking - do we know the Jets and Fins have weak defenses at this point in the season?  Not counting the Bills game, each had only played one opponent.  What conclusions can you base off of that?  Both the Jets and Fins defenses may have had bad statistical performances against the Bills simply because Allen and the Bills offense are that good.  We just don't know yet.  

 

Fitz, Drew Brees, Alex Smith, Rich Gannon, Jim Plunkett, Len Dawson and many other QBs improved over time.  Steve Deberg had a 40.0 passer rating in his first year as a starter.  He only got better after that with a peak rating of 96.3 as a 12 year veteran.  Early career stats often are NOT predictive of future performance.  Players sometimes just get better. 

 

They eye test seems to say that Allen is indeed getting better with his touch, deep ball, mechanics, and ability to find the open receiver.  The signs are certainly hopeful but I don't think two games is enough to rush to any conclusions, pro or con.  

 

 

  • Like (+1) 2
Link to post
Share on other sites
23 minutes ago, BisonMan said:

I guess Lamar Jackson has a 0.0 completion percentage by this logic as the Ravens almost never throw without play action or keep him in the pocket. So, Lamar sucks too?

 

I love 538’s election models but this article has a lot of issues because it expects teams to always play to a Tom Brady style offense, which won’t always maximize the strengths of their QBs.

 

 

"Each of these aspects of a QB’s performance is indicative of their true skill and is less affected by circumstances and luck.

 

I have a big problem with this.  Performance when under pressure and outside the pocket is more affected by circumstances and luck and performance when protection is clean is more meaningful?  You brought up Brady and he would be a great example.  He was throwing balls away, daring officials to make the grounding call, when he was facing pressure.  Being unwilling to take a hit to make a play is luck or "circumstances?"  Some of these QBs can't escape pressure and making plays on the move and outside the pocket is not even an option for them.  It's a club that is not in their bag.  I don't know how you can evaluate some QBs for parts of their game that do not exist for many of them.

 

Most all NFL starting caliber QBs perform well when not pressured or else they would not have the job.  So how could that be any meaningful way to evaluate their performances?  This author used their ordinal ranks after only two games to to say that Josh was middle of the pack in the "meaningful" measures: clean pocket QBR, no play action, completion% over expected and poor throw % even though the clean pocket QBR is a solid 79 and the bad throw % is a stellar 16% and SIS has him 4th in on target throw %.   

  • Like (+1) 3
Link to post
Share on other sites
7 hours ago, BeerLeagueHockey said:

Summary: We think players don't improve in the NFL, and he'll quickly return to his past performance level.

 

Thanks. Nate Pyrite's predictions are so seldom correct, I was hoping it was something like that.

 

image.png.8cd3941aa389bc46533f74db44cd9acc.png

  • Haha (+1) 1
Link to post
Share on other sites

My question is, that 79 QBR when not under pressure (19th rank in the league), wouldn't that include his two runs where he fumbled and lost the ball? I would imagine both those turnovers would have brought his QBR down quite a bit. If that's the case, I'm definitely not worried. 

 

But the other flaw here is where do Mahomes, Rodgers, Jackson and Wilson rank in QBR when not pressured compared to their overall QBR's? Do they also have decreases in ranking from their overall QBR? All four of these QB's and Allen excel in avoiding pressure and creating secondary reaction type plays. Guys like say Phillip Rivers, Drew Brees, Tom Brady on the other hand might have higher QBR ranks when not under pressure but lower QBR ranks overall. I'm sure the stats is out there somewhere but until we have it, for all we know guys like Cousins (well probably not this year) or Goff might have the highest QBR "not under pressure". And if that is the case, I don't really see the value. Just give me total QBR because a QB is going to be under pressure at some point of every game.

Link to post
Share on other sites
7 hours ago, JESSEFEFFER said:

Most all NFL starting caliber QBs perform well when not pressured or else they would not have the job.  So how could that be any meaningful way to evaluate their performances?  This author used their ordinal ranks after only two games to to say that Josh was middle of the pack in the "meaningful" measures: clean pocket QBR, no play action, completion% over expected and poor throw % even though the clean pocket QBR is a solid 79 and the bad throw % is a stellar 16% and SIS has him 4th in on target throw %.   

 

Very true. 

 

79QBR  is 19th ranked. What is is 32nd rank? 60 QBR? Basically it's likely a tight range of good QB play between 60 and 95. Which would be expected as you pointed out: NFL starting QB's should play well not under pressure.

 

 

7 hours ago, hondo in seattle said:

This analysis precludes the whole idea of progress.  For example, he argues that passing efficiency from a clean pockets is both stable and predictive.  I'm sure if you're looking at a QB in mid-career that's true.  But don't tell me that young QBs don't often get better at passing efficiency from a clean pocket over time.   The author, btw, never mentions whether or not Allen has made progress in the predictive metrics from 2018 to now.  He doesn't look to see if there's any upward trend.  If these metrics are truly predictive, wouldn't we expect an upward trend to continue?   

 

 

 

This is another excellent point. A fair comparison would be going back in history 10-20 years and seeing where all the 3rd year QB's ranked in these metrics. 

Link to post
Share on other sites
7 hours ago, hondo in seattle said:

Furthermore, he makes the ridiculous argument that Allen achieved his statistical heights this year against weak opponents.  How - statistically speaking - do we know the Jets and Fins have weak defenses at this point in the season?  Not counting the Bills game, each had only played one opponent.  What conclusions can you base off of that?  Both the Jets and Fins defenses may have had bad statistical performances against the Bills simply because Allen and the Bills offense are that good.  We just don't know yet.  

 

Especially that Dolphins defense. In fact the Dolphins overall as a team actually might even be good, when we look at their 3 games now and the teams they have played and how they played them. 

 

 

Link to post
Share on other sites

Article isn't unfair, especially when you consider the author, Hermsmeyer.  Guy has never been an Allen fan, he's extremely critical.  It's only week 2, jury is still out, I think that's fair enough to say.  But it's also fair to think that Allen may be a bit better than most predicted predraft, including myself.  Still a long season. 

 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...