Advanced stats and our QBs, 2015 edition

[Tsailand]

Previous thread was here: http://es.redskins.com/topic/385827-advanced-stats-and-our-qbs-2014-edition/

 

Here are the preseason stats from this year.  Of course they should be taken with a grain of salt, due to the nature of preseason games.  RG3's stats also suffer from a small sample size -- only 19 dropbacks total.

 

 

Passer Rating (not actually an advanced stat):

Cousins: 103.9

McCoy: 115.8

RG3: 54.6

 

 

Yards per Attempt:

Cousins: 8.2

McCoy: 7.3

RG3: 3.4

 

 

ANY/A (YPA, with sabermetric adjustments for sacks, ints, and TDs):

Cousins: 7.81

McCoy:  7.87

RG3: 1.69

 

 

QBR: Couldn't find preseasons stats for this.

 

 

DVOA: No preseason rankings.

 

 

 

 

 

[Califan007 The Constipated]

Those stats aren't very advanced lol...

[StonewallJackson]

Small sample size obviously, but pitiful from RGIII.

[bowhunter]

#sacks/attempt:

 McCoy     3/57= 5.2%  (Good)

 Cousins   1/53= 1.89% (Great)

 Griffin       3/13= 23.1%  (really really bad)

 

Interception/Att

  McCoy    0/57= 0% (Great)

  Cousins 1/53= 1.89% (Good)

  Griffin     0/13= 0% (Great)

 

Touchdowns/Att

  McCoy  4/57= 7.0% (Great)

  Cousins 2/53= 3.77% (Good)

  Griffin   0/13= 0% (not good) *Partial credit for Garcons drop would be 1/13= 7.7%

Thats all I got, I never really "advanced" past simple division. Mary Beth sat beside me in Algebra class though, I was the hardest part of that class.

[Tsailand]

Week one stats!

 

Adjusted Net Yards Per attempt from http://www.sportingcharts.com/nfl/stats/player-adjusted-net-yards-per-attempt-statistics/2015/

Cousins' ANY/A was 3.69, which is bad.  He ranks 26th out of 32, just below Andrew Luck and above Teddy Bridgewater.

The second INT hurt him, perhaps unfairly.  However, in the long run this will be balanced out by cases where he throws easy INTs that the other team drops.

 

 

QBR from http://espn.go.com/nfl/qbr

QBR is on a scale from 0 to 100. By design, 50 is average.  Week one, the highest QBR was Mariota with 95.7, the lowest was Winston with 6.7, and Ryan Tannehill had exactly 50.0

Cousins was 38.2, 23rd out of 32, just behind Andrew Luck who had 38.7

 

 

DVOA from http://www.footballoutsiders.com/stats/qb

Cousins was -6.0%, which means (I think) that he was a little bit worse than a hypothetical average QB.  They rank him as 19th out of 33 for week one.

 

 

The takeaway:  Cousins had a below-average game.  He ranked in roughly the bottom 1/3rd of QBs last week. Bad but not horrible.  Also, as in past years, DVOA likes Cousins a bit more than the other two stats do.

[ncr2h]

Keep in mind dyar and dvoa aren't adjusted for defensive rankings until after week 4. So if the Dolphins end up having a below average defense, his 11 dyar will get adjusted downward (similar to when Griffin put up 100+ dyar week 1 2012 against the Saints, who ended up being the worst defense in the league). If the Dolphins end up being a top defense, his performance will get adjusted upward.

It will be interesting what happens over the coming weeks. The consensus is that the defenses we face early on are extremely difficult. That won't stop cousins' detractors from declaring him a bust, and likely won't stop me or others from seeing a solid performance (barring a full game meltdown). If he can maintain a solid performance against stl and nyg I will be very impressed.

[Always A Commander Never A Captain]

Tannehill had a 50 QBR? I thought experts were supposed to have input and influence the stat. He was terrible. Started rewatching the game on NFL Game Pass, he had one wiiiidddeee open touchdown opportunity to Kenny Stills, no defender within 15 yards. He flat out missed, and Stills went for it but even if he caught it he wouldn't have had even one foot inbounds.

I don't know how anyone can rewatch that game with All 22 and come away thinking Tannehill was anything other than poor.

[RedskinsFTW Next Year]

Tannehill had a 50 QBR? I thought experts were supposed to have input and influence the stat. He was terrible. Started rewatching the game on NFL Game Pass, he had one wiiiidddeee open touchdown opportunity to Kenny Stills, no defender within 15 yards. He flat out missed, and Stills went for it but even if he caught it he wouldn't have had even one foot inbounds.

I don't know how anyone can rewatch that game with All 22 and come away thinking Tannehill was anything other than poor.

 

Worst game I've see Tannehill play. But interceptions will kill your qbr. Even though tannehill should have been picked off couple times, he still had 0 and put up over 220 yards. But yea, he looked bad.

[Tsailand]

Week 2 stats.

 

ANY/A: 7.28, pretty good.

Cousins' average for the season is 5.39, which puts him 24th out of 34 QBs ranked.

 

QBR: 72.4, 12th out of 32.

On the season, his QBR is 51.6, 22nd out of 32.

 

DVOA: no game-by-game rankings.

For the season they now have him 12th out of 34, right between Rivers and Tannehill.

[ncr2h]

Week 2 stats.

ANY/A: 7.28, pretty good.

Cousins' average for the season is 5.39, which puts him 24th out of 34 QBs ranked.

QBR: 72.4, 12th out of 32.

On the season, his QBR is 51.6, 22nd out of 32.

DVOA: no game-by-game rankings.

For the season they now have him 12th out of 34, right between Rivers and Tannehill.

FYI - while there's no dvoa weekly rankings, there are weekly dyar rankings. I belive dvoa is just dyar divided by number of plays, or something similar. The weekly dyar reports can be obtained from the "quick reads" reports which are published each Tuesday.

Cousins received 11 dyar in week 1 and 77 in week 2. If you take prior year dyar stats and divide them on a per game basis, pro bowl QBs typically average somewhere along the lines of 1000+ dyar per season, which translates to 62.5 dyar per game. This means that Cousins played at a pro bowl (or near pro bowl) level in week 2, while his week 1 performance was that of a solid but unspectacular starting QB.

It should be noted that week 2 was the 4th time that cousins has achieved back to back games with positive dyar. Rg3 has only accomplished that 2 times since his injury against Baltimore in 2012. For all the talk about how inconsistent he is, by dyar he has actually been relatively consistent - it's just that when he has a bad game, he has a BAD game. And those bad games have mainly come against the Giants.

Tomorrow is the biggest game of cousins' life so far. National TV game against a team where he has historically struggled. A solid performance and/or a win tomorrow will go a long way toward changing the narrative on Cousins. If he wants to move from the Ryan Fitzpatrick, Josh McCown tier in the mind of casual nfl viewers and wants to start being looked at as a legitimate long term starter, a good performance tomorrow is the first step.

[ncr2h]

Cousins' week 4 performance against the Eagles was good for 141 dyar, which is the highest performance of his career. It's also the highest by a redskins QB since at least 2012. Here are the highest dyar games by redskins QBs since 2012 (per quick reads, published each Tuesday on footballoutsiders.com):

1. 141 - Cousins vs Philly 2015

2. 138 - Cousins vs Philly 2014

3. 130 - Cousins vs Jacksonville 2014

4. 129 - Griffin vs Dallas 2012

5. 115 - Griffin vs New Orleans 2012

In the 13 games where Cousins has played more than a half, he has 9 games with positive dyar, including 6 games of 50 dyar or better and 5 games of 75 or better. In the 24 games played more than one half by non-Cousinses since the start of 2013, there have only been 9 games with positive dyar. Of those, 0 have been for 100 or more dyar (cousins has 3), 1 has been for 75 or more (cousins has 5), and 5 have been for 50 or more (cousins has 6). Non-Cousinses have achieved positive dyar just 9 times in their 24 games (cousins has 9 in his 13 games).

So far in his career, Cousins appears to be a major upgrade for this team. Really happy for his performance last week... That was huge.

[Larry]

But, does all that mean he's a really great QB?  Or does it mean he's bipolar, and his good games are really good (and let's not talk about the other ones)? 

 

---

 

I'll also say, I checked out their site, yesterday, and I've gotta say that I'm looking kinda dubious at their stats. 

 

They say our special teams are #18 in the NFL

[SkinsFTW]

I don't really hold it against Cousins that Tom Coughlin has his number, he always has Tom Bradys number too. As long as he keeps improving week to week he might be getting the contract that was expected for RG3 and we might not need to blow anymore 1sts on qbs.

[Tsailand]

I'll also say, I checked out their site, yesterday, and I've gotta say that I'm looking kinda dubious at their stats. 

 

They say our special teams are #18 in the NFL

 

That's up from 29th last week.  These rankings are based on just the current season.

 

In general, our special teams are dreadful every year, but so far it hasn't been as bad as it could have been.  I expect that in a few weeks we'll give up two return TDs in the same game, or something awful like that, and fall back down to 32nd where we belong.

[Tsailand]

Week 3:

ANY/A: 4.7, bad

QBR: 39.7, 20th out of 32 ranked QBs.

DYAR (thanks, ncr2h): negative 19 yards, 22nd out of 32. This is a per-game stat, not per-play. I think it's basically saying that a replacement-level QB like Colt McCoy would have gotten us 19 more yards against the Giants. Replacement level is not a good thing -- it means you are a backup.

(Last year, RG3's DYAR was an absolutely dreadful negative 374 yards, even though he played less than half a season.)

Week 4:

ANY/A: 6.6, average

QBR: 76.9, 8th out of 31.

DYAR:+141 yards, 2nd out of 31.

Through four weeks:

ANY/A: 5.53, 23rd of of 34.

QBR: 58.8, 13th out of 32.

DVOA: 13th out of 32.

In summary: He's playing like a real NFL starting QB, but not at a pro bowl level.

Those of you who expect HOF QB play out of a QB who has less than one season's worth of NFL games under his belt need a reality check. Or you can <STAFF EDIT, RULE 13, AVOID OVERLY EXPLICIT CONTENT>to 2012 highlight films of the read option while you wait for Dan Snyder to throw away a bunch more picks and money after whomever ESPN hypes next. Drew Brees and Peyton Manning are free agents in 2017!

[TSO]

Tsailand, let's avoid the type of baiting involved in your last few sentences. Just a verbal warning, but let's keep it classy from here on out or we'll have to take action. I have no doubt you'll understand, thanks.

[PokerPacker]

Tsailand, let's avoid the type of baiting involved in your last few sentences.

based on the context, looks like he also managed to involve the other kind. Unfortunately this is a written medium, so homophonic puns don't work as well as the homographic variety.

[Tsailand]

Week 5:
ANY/A:  4.51, bad.

QBR: 52.6, 13th out of 28.

DYAR:+33 yards, 13th out of 30.



Through five weeks:

ANY/A: 5.36, 26th of of 35.

QBR: 57.6, 16th out of 33.

DVOA: 12th out of 35.

[Tsailand]

I'll omit the week-by-week stats for weeks 6-8 and skip right to the rankings at the halfway point of the season:

 

ANY/A: 5.35, 26th of of 32.

QBR: 56.2, 20th out of 32.

DVOA: 12th out of 35.

[Tsailand]

The other day, someone mentioned that Kirk would be the top free agent QB after this season.  How true is that?

 

Here are the 29 NFL QBs who will be free agents in 2016, according to http://www.spotrac.com/nfl/free-agents/quarterback/

 

Brandon Weeden
Brock Osweiler
Bruce Gradkowski
Case Keenum
Chad Henne
Charlie Whitehurst
Chase Daniel
Dan Orlovsky
Drew Stanton
Jimmy Clausen
Joe Webb
Josh Johnson
Kellen Clemens
Kellen Moore
Kirk Cousins
Luke McCown
Matt Cassel
Matt Flynn
Matt Hasselbeck
Matt McGloin
Matt Moore
Matt Schaub
Michael Vick
Ryan Fitzpatrick
Sam Bradford
Scott Tolzien
T.J. Yates
Tarvaris Jackson
Thaddeus Lewis

 

(RG3 is not on the list because he is currently under contract for 2016.  Colt should be on the list, but isn't -- not sure why.  Let me know if they've missed anyone important.)

 

Most of these guys haven't seen much play time this year, because they are backups. The only one who has seen enough play time this year to be "qualified" on any of the advanced stats lists, and is ranked above Cousins on any of them, is Ryan Fitzpatrick.  As it happens, he ranks above Cousins on all three lists.

 

So there you have it: Cousins is on track to be only the second most sought-after QB next season.... except that he just turned 27, while Fitzpatrick turns 33 in a couple weeks.

[TD_washingtonredskins]

So there you have it: Cousins is on track to be only the second most sought-after QB next season.... except that he just turned 27, while Fitzpatrick turns 33 in a couple weeks.

 

Which is probably what tips the scales. It's possible that people would consider Fitzpatrick a backup at this point simply due to his age. He's an interesting case though...he hasn't "played" as much as some other QBs and is having a very nice year. I think for a 2-3 year deal he would make a quality starter. 

 

I'll bet Cousins is considered the best available QB just because it's feasible that he could lead a team for 6-8 years if everything worked out. 

[Tsailand]

Bringing this thread back up.  Not bothering with game-by-game stats since Cousins is clearly no longer on a game-to-game leash.

 

 

Current rankings after fourteen games:

ANY/A: 6.72, 12th of of 35.  Carr is 11th and Smith is 13th.

QBR: 66.0, 12th out of 32. Eli is 11th and Brady is 13th.

DVOA (through 13 games): 11th out of 33. Brees is 10th and Rivers is 12th.

 

 

 

For salary purposes, the main comparison is Fitzpatrick, who becomes a free agent in March.  His rankings:

ANY/A: 14th.
QBR: 9th.
DVOA: 13th.

[skinsfan35yrs.]

Bringing this thread back up.  Not bothering with game-by-game stats since Cousins is clearly no longer on a game-to-game leash.

 

 

Current rankings after fourteen games:

ANY/A: 6.72, 12th of of 35.  Carr is 11th and Smith is 13th.

QBR: 66.0, 12th out of 32. Eli is 11th and Brady is 13th.

DVOA (through 13 games): 11th out of 33. Brees is 10th and Rivers is 12th.

 

 

 

For salary purposes, the main comparison is Fitzpatrick, who becomes a free agent in March.  His rankings:

ANY/A: 14th.

QBR: 9th.

DVOA: 13th.

For us uninformed, what do ANY/A and DVOA stand for?

[Always A Commander Never A Captain]

I feel like comparisons to a veteran QB like Fitzpatrick aren't the best.  You know who he is and what he does on the field.  Kirk's last 8 games are far better than his first 6 games.

 

We don't know what Kirk can do.

 

Maybe for these stats and taking them in multi-game increments, or maybe turning them into a graph?

[Tsailand]

For us uninformed, what do ANY/A and DVOA stand for?

 

From the previous thread:

 

There's an advanced version of YPA called ANY/A - adjusted net yards per passing attempt.  Basically it takes sacks and lost sack yardage into account, penalizes you 45 yards per INT, and rewards you 20 yards per passing TD.  These two numbers aren't arbitrary like the basic passer rating calculation, but based on advanced research.

 

Football Outsiders has a state called DVOA (Defense-adjusted Value Over Average).  The calculations are also secret, but they take everything into account, including the strengths of opposing defenses.  They rank every QB with at least 100 passes this year: