ESPN: QBASE 2022

QB Class of 2022
QB Class of 2022
Photo: USA Today Sports Images

NFL Draft - Football Outsiders' QBASE projections for the QB Class of 2022 are not promising. There's a 68% chance at least one of these guys becomes an upper-tier starter, but do you want to take that bet when you can only grab just one of them? Click the link below for the ESPN+ version of QBASE, which will also run on FO later this week.

View Full Article

Comments

20 comments, Last at 09 May 2022, 1:53pm

1 I suppose quantum mechanics…

I suppose quantum mechanics holds that if all you have are anti-particles, at least one of them must have fired off a positive particle. Granted, 2013 observed that the observable universe was not a balanced one.

\For purposes of 1985 and 1988, can we include their punting statistics?

2 These articles..

...should always come with an assessment of how well this measure has done in the past to predict NFL performance.

6 Last year, Lawrence and…

In reply to by JacqueShellacque

Last year, Lawrence and Wilson were practically tied (Lawrence had a higher Elite percentage, by five points).  Of the first round QBs, Mac Jones had the highest bust rate.  Seems like the Mac Jones pick was a miss.

2020: Burrow was ranked first, but had higher bust rate (over 40%) than Lawrence or Zach Wilson.  Herbert had the second highest rating, followed closely by Tua.  Qbase hated Jordan Love (64% bust rate), and liked Jalen Hurts (only a 51% bust rate for a 2nd rounder).  Sorry about Love, New.

2019: Kyler Murray ranked 1st over Dwayne Haskins at 2, but neither was ranked that highly.  Haskins had a bust rate of 51%, a little worse than Murray's 47%.  Qbase didn't think much of anyone else, but slightly better of Daniel Jones than the others.

2018: Since I'm a Jets fan, I'll just post the DYAR predictions here:  Sam Darnold, 412 DYAR.  Josh Allen, -83 DYAR.  Oops.  Josh Rosen, 623 DYAR.  Baker Mayfield, 1480 DYAR.  Lamar Jackson, 656 DYAR.  Darnold had a bust rate of 51.9, Allen had a bust rate of 62.7, and Rosen had a bust rate of 46.9.  Qbase liked Jackson and Mayfield, so that's a plus, but it hated Allen and didn't hate Darnold and Rosen.

2017: Trubisky with a 435 DYAR and a bust rate of 49%, Watson with a 261 DYAR and a bust rate of 56%, and then Patrick Mahomes with 656 DYAR and a bust rate of 45.7%.  So Qbase liked Mahomes the most, and was sour on Watson.  A mixed bag here.

9 The problem with this sort…

The problem with this sort of thing is it assumes the determining factor of success is what they did before being drafted.....it's necessary assumption, but probably not true. Pretty sure 90% of time Mac Jones is a bust(or a late round pick and decent backup), he just landed in a good situation.  

17 It's also possible Mac Jones…

It's also possible Mac Jones turns into a pumpkin, a la Rick Mirer, although Jones' issues are different (lack of arm strength, and not inability to run a complex offense).  Now that I think of it, Jones really is Pennington 2.0.

12 "Sorry about Love, New."

No need to be sorry. I never understood his hype and he was QB5 at best and clearly not the right time for us specifically. Ironically he was supposedly put in the best situation and still comes out looking rough. I don't have much hope but maybe Seattle can give us something otherwise I don't see how any other team would be interested in his 2+1 years left.

14 https://www…

https://www.footballoutsiders.com/stat-analysis/2021/introducing-qbase-v20
It seems to do a better job predicting busts than success (probably due to larger sample size). (Most accurate pick: Trubisky, Tua and Jack Locker).  Partly it doesn't seem to every predict that high a success. The highest prediction is Marcus Mariota, which it thought would be a discount Russel Wilson (1.4, 1.5 being HOF level). 

The highest guy who both had a fairly high prediction and decent career sample size is Rivers and it's not that close. The best guy it got close to right(+- .1) was Andrew Luck, but that's mostly bc he got injured before he could exceed the projection by much. It mostly seems to predict the first couple years pretty well, and they either grow past that(and it under shoots), or they prove it right.  So it's basically somewhat accurate for the newer guys (Herbert, Burrow, Murray) but I suspect that gap will increase this year. 

It's an interesting system and it's not like NFL teams are all that good at this either. 

19 I do remember from the…

I do remember from the Watson article that the system dinged him pretty hard for teammate quality, so if you were willing to overlook that, it was pretty high on him.

3 Hindsight 2021

It's a bit jarring to see Ridder and Howell in particular so far behind Wilson and Lance despite sample size and level of competition. 

4 Not surprising for Howell. …

In reply to by terps

Not surprising for Howell.  Both Wilson and him were three year starters, and Howell had his great year in sophomore year unlike Wilson who had an even better year right before coming out.  I don't have ESPN +, so I don't know the actual ratings yet, so I can't figure why Ridder is ranked lower than Wilson though.

7 The other thing is the…

The other thing is the teammate draft levels.  No one on BYU went that high; I don't know about North Carolina, so maybe that hurts Howell, maybe it doesn't.  Ridder has Pierce and a running back that's going to get drafted too.

10 I think historical Qbase…

I think historical Qbase only goes back to 1997.

https://www.footballoutsiders.com/stat-analysis/2015/introducing-qbase

It liked Geno Smith in 2013, but hated everyone else. ''Overall, worst high misses are Peyton Manning and Matt Stafford (loved them, but not enough). Worst low-misses are Matt Ryan (hated Ryan) and Brian Griese. It despised Tannehill, who turned out to be actually decent. John Beck is the worst performer of the QB's QBase liked (non-asterisk division). 

13 Actually, the 2013 Lewin…

Actually, the 2013 Lewin Forecast predicted over a 1,000 DYAR for 6 guys: Matt Barkley, Ryan Nassib, EJ Manuel and Geno Smith's 2,064 prediction.   Zac Dysert had a 1,131 if he went in the first three rounds (LCF only applied to quarterbacks taken in the first 3 rounds).  Then there was the Asterisk Part II: Landry Jones with 2,276.  Nassib, Barkley, Dysert and Jones went too late to count, but EJ's and Geno's projections are a black mark on the LCF and therefore Qbase.

15 The QBase for Geno didn't…

The QBase for Geno didn't predict Jetsiness.

Brohm, Tebow, Smith, Kolb, and Clemens were also almost identical misses, and probably reflect that QBase has a really hard time with guys who are probably QB2s, talent-wise. It doesn't well-capture that bifurcation.

16 Matt Waldman was down on…

Matt Waldman was down on Geno, so I wouldn't proclaim Jetsiness in Smith's case.  I doubt they ruined him, as they may have with Darnold (Sanchez was a lost cause).  The Mahomes year the FO writer mentioned that Qbase had issues with loving Air Raid guys too much; hilarious that they picked that year to write a disclaimer.

20 Post draft rankings

Went

  1. Pickett - QB1 (1.20)
  2. Corral - QB4 (3.94)
  3. Ridder - QB2 (3.74)
  4. Willis - QB3 (3.86)
  5. Howell - QB6 (5.144)
  6. Strong - QB10+ (UDFA)

Strong really just doesn't have mobility and the league thought the same.