Chat about Total Quarterback Rating
Welcome to SportsNation! On Friday, we continue the Year of the Quarterback when we have Albert Larcada from ESPN's Production Analytics team stopping by to explain the new Total Quarterback Rating developed by the group.
As we're two weeks into the 2011 NFL season, the Total Quarterback Rating (or Total QBR) attempts to become a more encompassing way of evaluating a quarterback's play and replace the old passer rating which was implemented in 1973. Not surprisingly, Tom Brady leads all QBs through two weeks.
Send your questions now and join the chat Friday at 2 p.m. ET!
Albert Larcada (2:00 PM)
Hope everyone is having a good Friday! Let's get into some QBR questions...
What's the biggest factor in the QBR equation?
Albert Larcada (2:03 PM)
Passing is, and should be, the biggest factor in any QB rating system. This is no different in ours. We just take passing a little deeper and get into the situation and context where the QB's are making these passes.
For how long were you guys toying with the formula?
Albert Larcada (2:05 PM)
Our Production Analytics wasn't fully formed until early this year, but a new QB rating has been on our radar for years. Bringing in Dean Oliver from the Denver Nuggets really helped us get QBR off the ground.
Chris Fiegler (Latham,NY)
Who do you think is the best Rookie Quarterback in the NFL this season?
Albert Larcada (2:10 PM)
We have Newton as a slightly above average QB this season and Dalton as below average. The NFL passer rating has Dalton higher.
In your opinion, are there any QBs the the QBR system undervalues or overvalues?
Albert Larcada (2:12 PM)
We designed QBR so that it's really hard to "game" the system. If you are helping your team score points and win games your QBR will go up. If you don't, it will go down. The QB's that have high QBR's are the ones who have helped their teams score/win the most.
Seeing Jason Campbell in the top 10 of the ratings right now was my biggest surprise. I would never have thought him a top 10 QB, would you?
Albert Larcada (2:14 PM)
Honestly, no. Entering the season Campbell had a 49.9 QBR in the last three years, making him about as league avearge as a QB can be. He has definitely exceeded that this so far this season with arguably less talent around him.
Matt Stafford (Detroit, MI)
A fantasy owner traded Aaron Rodgers because i was just sitting on the bench. Do you think i can keep up the pace im at?
Albert Larcada (2:17 PM)
Well, Matt, I would say that is a risky move. Can you stay healthy for a full season?
I like the new ratings, but I'm not fond of the way they reward QB's who suddenly wake up in the 4th quarter, then penalize QB's who are consistently good from beginning to end. I think many teams have proven than a 17-21 point lead can be overcome so why take points away from a QB for wanting to keep adding scores even with that lead? Footbal after all is a game of momentum and the Giant's proved last year how quicly that momentum can shift.
Albert Larcada (2:20 PM)
This is a common misconception about QBR. QB's that go out in the 1st half and blow people out then play below aveage in the 2nd half have just as good of a chance as the other way around. It's all about winning. If a QB plays so well in the 1st half that he literally doesn't need to play well in the 2nd half to win, then he will have a very good QBR. See Michael Vick against the Redskins on MNF last year (the highest QBR in a game since 2008).
Luke (San Francisco)
How are sacks evaluated? Is there any way to differentiate between a sack caused by a QB holding the ball too long vs. one caused by the offensive line?
Albert Larcada (2:23 PM)
This is also a common question for us. Sacks are tricky because usually only the team itself knows exactly who messed up on the play. However, our own research and analysis done by Pro Football Reference and Football Outsiders have shown that sacks are a lot more on the QB than you might think. In general, we model sacks by the number of pass rushers. The fewer rushers the more blame on the OL because they should be able to handle it. The more rushers, the more blame on the QB. He should know he either a) needs to bring on more blockers pre-snap or b) get rid of the ball quickly because he doesn't have enough blockers.
Everyone always talks about Big Ben being an elite QB, but right now he's 21st in the ratings, below Alex Smith! This ratings system really makes you think, because I wouldn't have expected that.
Albert Larcada (2:27 PM)
Big Ben is one QB that we rate lower than most other systems. This has a lot to do with his sacks and fumbles. Entering the season Big Ben had fumbled 28 times since 2008! For reference Peyton Manning has only fumbled six times. Imagine if those had been INT numbers instead of fumbles, then I believe most would agree with Big Ben's ranking. Often fumbles put your team in a worse situation than INT's.
Kevin (Limpy) Croneigh (La La Land)
I love Jay Cutler. Best QB in the NFL.
Albert Larcada (2:27 PM)
When did you finally say, "I think we got this....this is the version of the formula we want to go with"?
Albert Larcada (2:32 PM)
I'm not sure we ever say that. As long as the data continues to become more and more granular we (and every other analyst) should continue to push football analytics forward. We have a lot of great video-tracked data we used to create QBR, but we can also track and add more.
Does your team winning the game factor into the QBR at all?
Albert Larcada (2:34 PM)
In a vacuum, no. There is no "extra credit" for a win. However, since 2008 the team that has a better QBR wins roughly 86% of games, so QB's that have high QBR's tend to win.
spencer (Beverly, Ma)
Could you list the top 5 qb in the new QBR system. Your picks
Albert Larcada (2:36 PM)
Sure. 1) Tom Brady. 2) Ryan Fitzpatrick. 3) Matt Hasselbeck. 4) Tony Romo. 5) Matt Stafford.
Is it true that you can never achieve a perfect 100 in QBR unlike with the old passer rating? Could you explain how that works?
Albert Larcada (2:39 PM)
That is true. You can never reach a 100.0 QBR. The closest so far was Vick vs. the Redskins last year on MNF with a 99.8. The reason being that QB's can never be perfect. If a QB goes 20-20 for 400 yards and 6 TD's how can we give that a 100.0 when he could have thrown for 401 yards and 7 TD's. Wouldn't that have been a better game? Go back and look at every 158.3 in the NFL passer rating, I promise you will find at least one (likely several) flaws in that game.
Percentage wise ...how much did the old system carry over to the new system
Albert Larcada (2:41 PM)
I believe Pro Football Reference did a study on this, but I don't remember their results. The correlation was something like 75%? We didn't design the system to correlate (or not correlate) with the NFL passer rating.
When are you going to add defensive adjustments, and why haven't you done it yet?
Albert Larcada (2:46 PM)
Defensive adjustments is something we left out on purpose. Given all the splits we do with QBR it wouldn't be practical to add a defensive adjustment to every situation. Also, def adjustments are not as practical early in the year because we really don't know what defenses are good yet. It is something we plan to do in the future.
Tim (New York)
you say its all about winning in the new QBR but only 3 of the top 5 are 2-0 when there are four other teams that are undefeated so far? explain?
Albert Larcada (2:49 PM)
Five of the top 7 are undefeated QB's (Brady, Fitzpatrick, Stafford, Rodgers, and Schaub). Grossman is 10th. Sanchez frankly has not played as well as other QB's this season.
Brady being first in the rankings right now is no surprise to anyone.....but is there any QB in the ratings right now that makes you stop and think?
Albert Larcada (2:53 PM)
Seeing Matt Hasselbeck 3rd made me do some research. He has NOT been a good QB in QBR over the last three years, one of the worst actually. It turns out he has really dominated on 3rd down. he has the third most 1st down passes, the 2nd best comp pct, and has been sacked, intercepted or fumbled on 3rd downs this year. Keeping drives alive is a huge part of QBR.
If you have to sum up what the QBR is trying to measure, what would you say?
Albert Larcada (2:53 PM)
We are trying to measure all of a QB's contributions toward scoring points and winning games.
patrick throckmorton (winchester,va)
how well do you think rex grossman will do this season and does he have what it takes to take the skins to the playoffs ?
Albert Larcada (2:56 PM)
Currently Rex is 10th in QBR this season. Given that his QBR was 25.9 since 2008 entering the season (given, a small sample size) I cannot be confident he will keep this up.
Bryan (Springdale, AR)
What do the ratings have to say about who performs better under stressful conditions, such as trying to come back in the 4th quarter, 2-minute drills, or playing without starting receivers or playing with injuries?
Albert Larcada (2:58 PM)
You sound like Trent Dilfer! This is exactly something he would ask. Trent's three biggest things he told us we should take into account before we started the QBR process was 3rd down, 4th quarter, and red zone performance. These are all things that we weight pretty heavily given how often they decide games.
Matt Stafford (Detroit, MI)
if i can stay healthy all year. and If megatron is healthy will i be an MVP contender?
Albert Larcada (3:00 PM)
Sure, Matt Stafford. Anything you say. Seriously though, in a short period of time Stafford has performed well. Check out Chris Sprow's article on ESPN.com, he does a good job of breaking him down.
Albert Larcada (3:01 PM)
Thank you guys for all the questions. Look forward to chatting again soon!