[Vague Title] It continues... Josh Allen...

[Sky Diver]

6 minutes ago, C.Biscuit97 said:

Do you even read what you post?  You were upset with people who are finding faults with Allen when he hasn’t played yet, but that hasn’t stopped you from doing the same thing with Rosen.  Be consistent and stop holding a grudge against a guy for saying Bama doesn’t challenge its football players academically. 

 

And the Brady hate is out of jealousy and for the most part, not serious.  We “hate” him but would love to have him. As much as it’s weird how he kisses his son.

 

Allen plays for the Bills and Rosen doesn't. There is a time honored tradition of poking fun at opposing players and Rosen is an easy mark

[oldmanfan]

23 minutes ago, TheElectricCompany said:

Sure, let's go with that. Glad we cleared that up. 

What is a multivariate analysis?

[Sky Diver]

5 minutes ago, jrober38 said:

 

The only meaningful part of QBASE is who had elite scores, and who had negative scores. There are too many hits and misses in between. 

 

With that said, players with a negative score (like Josh Allen) literally never go on to become franchise QBs in the NFL. 

 

QBASE gave Wentz a 61.9% chance of being a bust? How well did that work out? We need some analytics on the analytics.

 

Edited July 25, 2018 by Sky Diver

[TheElectricCompany]

1 minute ago, oldmanfan said:

What is a multivariate analysis?

Something you looked up on Google 

[jrober38]

Just now, Sky Diver said:

 

QBASE gave Wentz a 61.9% chance of failing. How well did that work out. We need some analytics on the analytics.

 

 

Sure. 

 

I'm just pointing out that in 21 years, there hasn't been a QB with a negative QBASE score who went onto become a franchise QB in the NFL. As I said, there are plenty of hits and misses in the 0-800 point range. 

 

Althought Wentz's score was low, and his bust probability was high, it wasn't negative like Allen's was. 

[Figster]

9 minutes ago, jrober38 said:

 

This isn't about guys with a higher completion percentage. This is looking at guys whose college completion percentage was very poor, and history very clearly shows that these players don't go on to become quality NFL players. 

 

We're all hoping Josh Allen bucks the trend and becomes "the guy", but every stat and bit of data you can find say his chances of becoming a franchise QB are slim to none. Most 1st round QBs fail, so whoever we picked likely wouldn't have been successful. That's just reality. It's really hard to find a QB. 

So you admit most 1st round QB's fail.

 

Perhaps that should be telling us the way college QB's are being evaluated is not very reliable.

 

 

[C.Biscuit97]

6 minutes ago, Shaw66 said:

But in this case, the guy is neither dumb nor annoying.   In fact, if he's annoying at all, it's because he relentlessly is looking to get better.  So he's annoying like Kirk Cousins.   

I was thinking more about the girls.  With the tall, big armed guys, teams ignore serious flaws that can’t get them by in the nfl.

 

i did this with EJ.  I was in full homer mood and was excited to get a qb.  He had all the physical tools, was a great dude (played his senior year with his mom battling cancer and didn’t say a word), and a lot of people trashed the pick.  Looking back, people were concerned about his accuracy.  And he completed 67% of his passes at almost 9 ypa in the ACC.

 

 

[TheElectricCompany]

4 minutes ago, Sky Diver said:

 

QBASE gave Wentz a 61.9% chance of being a bust? How well did that work out? We need some analytics on the analytics.

 

QBASE values 2+ years starting experience. 

I'm sure the odds of becoming an elite QB are significantly lower for <2 year starters. 

[C.Biscuit97]

2 minutes ago, Figster said:

So you admit most 1st round QB's fail.

 

Perhaps that should be telling us the way college QB's are being evaluated is not very reliable.

 

 

Is it 50-50 that 1st round guys become franchise qbs?  Well then it’s like 95-5 1st rounders with under 60% (in the last 10 years) become franchise guys.  

 

 

[CommonCents]

Just now, C.Biscuit97 said:

I was thinking more about the girls.  With the tall, big armed guys, teams ignore serious flaws that can’t get them by in the nfl.

 

i did this with EJ.  I was in full homer mood and was excited to get a qb.  He had all the physical tools, was a great dude (played his senior year with his mom battling cancer and didn’t say a word), and a lot of people trashed the pick.  Looking back, people were concerned about his accuracy.  And he completed 67% of his passes at almost 9 ypa in the ACC.

 

 

How long did it take you to realize that EJ didn’t have it? Did you watch EJ at FSU? 

[Sky Diver]

6 minutes ago, jrober38 said:

 

Sure. 

 

I'm just pointing out that in 21 years, there hasn't been a QB with a negative QBASE score who went onto become a franchise QB in the NFL. As I said, there are plenty of hits and misses in the 0-800 point range. 

 

Althought Wentz's score was low, and his bust probability was high, it wasn't negative like Allen's was. 

 

They had Jackson rated the #2 QB and he went at 32.

[BillnutinHouston]

13 minutes ago, Sky Diver said:

 

They have had some spectacular misses too, like Carson Wentz.

 

Im no analytics expert, but I think where analytics worshippers go too far is their belief that data from the past is always determinative going forward. It simply isn't, and the embodiment of this is the many late round picks (all of whom were measured analytically) who go on to outperform their higher-drafted peers at the NFL level. 

 

Its human nature to develop predictive models, and while these will inevitably get better with time, they will always be confounded by the unpredictability of the human element.  

[jrober38]

2 minutes ago, Sky Diver said:

 

They had Jackson rated the #2 QB and he went at 32.

 

NFL teams obviously don't use QBASE. But that doesn't mean they shouldn't. 

[eball]

4 minutes ago, BillnutinHouston said:

 

Im no analytics expert, but I think where analytics worshippers go too far is their belief that data from the past is always determinative going forward. It simply isn't, and the embodiment of this is the many late round picks (all of whom were measured analytically) who go on to outperform their higher-drafted peers at the NFL level. 

 

Its human nature to develop predictive models, and while these will inevitably get better with time, they will always be confounded by the unpredictability of the human element.  

 

That right thar sounds purdy smahrt....

[C.Biscuit97]

3 minutes ago, BillnutinHouston said:

 

Im no analytics expert, but I think where analytics worshippers go too far is their belief that data from the past is always determinative going forward. It simply isn't, and the embodiment of this is the many late round picks (all of whom were measured analytically) who go on to outperform their higher-drafted peers at the NFL level. 

 

Its human nature to develop predictive models, and while these will inevitably get better with time, they will always be confounded by the unpredictability of the human element.  

Well taking anything as gospel is dumb.  But there are statistics that should factor in decision making.  Let hitting on 17 in blackjack.  You could pull a 4 but all evidence is you’re better staying.  

 

I feel the Allen pick is more of a gut feeling type pick.  

[Figster]

11 minutes ago, C.Biscuit97 said:

Is it 50-50 that 1st round guys become franchise qbs?  Well then it’s like 95-5 1st rounders with under 60% (in the last 10 years) become franchise guys.  

 

 

What are the odds of a 6th round QB becoming the best signal caller the league has ever seen. Or a guy like Kurt Warner is bagging groceries one minute yet goes on to win a championship.

 

I could care less about the odds...

 

with all due respect Biscuit

Edited July 25, 2018 by Figster

[jrober38]

6 minutes ago, Commonsense said:

How long did it take you to realize that EJ didn’t have it? Did you watch EJ at FSU? 

 

I gave up on EJ after the San Diego game in his second season. He showed no improvement from year 1 to year 2, was missing wide open receivers and seemed to have regressed. I defended him pretty thoroughly as a rookie, but after 4 games as a sophomore I felt his benching was completely justified. He looked completely lost. 

 

I bought into the idea he was a raw player with lots of upside and I overlooked all of his deficiencies as a passer. I won't do that again this year. 

[Sky Diver]

4 minutes ago, jrober38 said:

 

NFL teams obviously don't use QBASE. But that doesn't mean they shouldn't. 

 

If Philadelphia had used QBASE they would have missed out on Carson Wentz. QBASE gave him a 61.9% chance of being a bust, about the same percentage they have assigned to Allen.

Edited July 25, 2018 by Sky Diver

[jrober38]

3 minutes ago, Sky Diver said:

 

If Philadelphia had used QBASE they would have missed out on Carson Wentz.

 

Not necessarily.

 

As I've said a few times, the only really meaningful aspect of QBASE is that guys with a negative score never work out. There are tons of misses for guys with scores in the 0-800 range. 

 

All the teams who picked QBs with a negative score would likely be very happy had they used QBASE. Just imagine if we'd not drafted JP Losman (-192 QBASE) and had a 1st round pick the next year that we could have used on Aaron Rodgers (+1198 QBASE). 

 

Picking players with a negative QBASE score has never proven to be a wise decision. If you use QBASE as a binary type of test to eliminate some players, if you took all the guys with a negative score off your draft board your odds of picking a successful QB would go up because you'd only potentially pick a player who actually had a chance of being successful. 

Edited July 25, 2018 by jrober38

[Sky Diver]

1 minute ago, jrober38 said:

 

Not necessarily.

 

As I've said a few times, the only really meaningful aspect of QBASE is that guys with a negative score never work out. There are tons of misses for guys with scores in the 0-800 range. 

 

All the teams who picked QBs with a negative score would likely be very happy had they used QBASE. Just imagine if we'd not drafted JP Losman (-192 QBASE) and had a 1st round pick the next year that we could have used on Aaron Rodgers (+1198 QBASE). 

 

Picking players with a negative QBASE score has never proven to be a wise decision. If you use QBASE as a binary type of test to eliminate some players, if you took all the guys with a negative score off your draft board your odds of picking a successful QB would go up because you'd only potentially pick a player who actually had a chance of being successful. 

 

From an analytics perspective, do you think that using the 2nd overall pick in the draft on a player that has a 61.9% chance of being a bust make sense to you?