ESPN tries to build a better quarterback rating

Getty Images

Quarterback rating is a statistic every football fan knows but few really get. Why is it scaled so that 158.3 is a perfect score? Why does it only include passing, and not other aspects of quarterback play like avoiding sacks and running? Why does it penalize interceptions but not fumbles? Why does it treat a six-yard completion on second-and-5 the same as a six-yard completion on third-and-11?

The NFL adopted passer rating as an official statistic in 1973, and the quest to improve passer rating began almost immediately. Now ESPN is poised to take the biggest step yet toward an improved rating that will be both a better assessment of how good a quarterback is and an easier stat for average NFL fans to understand.

The new stat, which ESPN is calling the Total Quarterback Rating (Total QBR), will be unveiled during a TV special on Friday night. Developing Total QBR was a joint venture between some of the statheads at ESPN and NFL analysts like Trent Dilfer, Jon Gruden and Ron Jaworski. The bottom line, as one of ESPN’s statisticians put it in a press release, is, “If you want one stat that measures the totality of a quarterback’s performance, it’s QBR.”

Unlike passer rating, Total QBR considers everything a quarterback does except handoffs. Quarterback runs, sacks, fumbles and penalties, all of which are ignored by passer rating, are included in Total QBR.

Total QBR also includes the situation of the play, such as down and distance, field position and the time left in the game. So a quarterback who runs for two yards on third-and-1 will be rewarded in the system, while a quarterback who runs for two yards on third-and-10 will be penalized. And a quarterback who racks up a bunch of passing yards while his team is way behind in the fourth quarter, playing against a prevent defense, won’t be treated the same as a quarterback who racks up a bunch of passing yards in the process of helping his team build a lead in the first half.

One of the aspects of Total QBR that could be both a strength and a drawback is that it considers data that the average fan doesn’t have access to, like how far a pass travels in the air, and whether the quarterback was under pressure when he threw it. That could be a great benefit of Total QBR because it incorporates detailed information that only comes from film study, not from the box score. The drawback, however, is that it means fans can’t see for themselves exactly where Total QBR comes from — fans just have to trust that the distance the ball traveled was correctly measured, and how much pressure the quarterback felt on the play was correctly assessed.

Despite all those extra elements that go into Total QBR, the stat is designed to be easier to understand than traditional passer rating. Toward that end, Total QBR is based on a scale where 100 is perfect and 50 is average. No more perfect scores of 158.3.

If you’re curious how Total QBR works in practice, Mike Sando of ESPN.com writes that the top quarterbacks of 2010, according to Total QBR, were Tom Brady, Peyton Manning, Matt Ryan, Michael Vick, Aaron Rodgers and Drew Brees. The worst quarterbacks, according to Total QBR, were Derek Anderson, Brett Favre and Jimmy Clausen.

That seems roughly correct, but then again you didn’t need an advanced stat to tell you that. And right now we don’t know all the details of exactly how this stat is calculated.

If ESPN is committed to this stat and is able to clearly and concisely explain it on the Worldwide Leader’s NFL broadcasts, then fans will quickly become familiar with it and it will soon become a staple of how we talk about quarterbacks. On the other hand, if the stat comes across as too convoluted — or if it doesn’t really seem like much of an improvement on the current passer rating — then this will all feel like a rather pointless exercise. We’ll be interested to see how it’s presented on Friday night. And we’ll be interested to hear what PFT Planet thinks.

70 responses to “ESPN tries to build a better quarterback rating

  1. They try to build a better rating to justify all there @$$ kissing they do for the top Qbs they listed! What a joke! Secret formula!

  2. “”The drawback, however, is that it means fans can’t see for themselves exactly where Total QBR comes from — fans just have to trust that ESPN correctly measured how far the ball traveled, and correctly assessed the amount of pressure the quarterback felt on the play.””

    Fans won’t trust ESPN, and the traditional passer rating will still be used….so this will just make things more complicated to people who have no idea that ESPN decided to invent some new stat that seems very similar to the old stat. It will have 0 credibility unless they show both stats right next to eachother….so what exactly do they think they’ve accomplished?

    Eventually, the numbers will be so predictablle due to the nature of averages that people will think its like an exchange rate or converting pounds to kilograms.

  3. I was excited to read this until i was about half way thru the post and realized its going to be more complicated to equate the total QBR then figure out what those awful sores are on Al Davis forhead

  4. So

    colt McCoy > sam Bradford
    Eli manning > ben roethlisberger
    Matt Ryan > Philip rivers

    Nice try ESPN.

  5. JaMarcus Russell QBR … -0.24 Which is prob his Blood Purple drank content as well.

  6. Amen. It’s about darn time. The existing QB rating is meaningless garbage. Means NOTHING to NFL coaches.

    One key passing stat they do consider when measuring the overall productivity of a teams passing game, is YPA (yards per pass attempt. A 7+ yard YPA is outstanding.

  7. Well, strictly speaking it only includes “passing” because the statistic is “Passer rating” not “Quarterback rating.”

    It may seem like an outdated statistic, but honestly “Passer Rating Differential” (the difference between your QB’s passer rating and the passer rating of opposing quarterbacks) is actually a stat that strongly correlates with success in the NFL, so the old statistic absolutely should not be done away with, unless we can find a most strongly correlated statistic than PRD.

    If ESPN wants to come up with a metric to measure overall quality of quarterbacks so they can talk about it on television… sure, whatever.

  8. I am all for it. 158.3 always seemed to be the absolute most random figure. ESPN may be on their high horse here, but I don’t think anybody accepts the current QB rating as a true tell of a quarterbacks performance.

  9. ehh… not to excited about this.. can’t believe ESPN is making a big deal about a stat. Now every sports network is going to come up with their new stat too..

    if it works, great. I’m all for it. but, its just a stat.

  10. This is very good news. The old rating was invented in a lab, this is more real world, it appears, giving credit were it is due. More mobile QBs will get their due as well. The numbering system is easier to follow too. Say good-bye to 158.3.

  11. Great idea and definitely should be a success.

    Can someone please do this now for the world golf rankings…they are such a disaster!!

  12. If Kerry Byrne was not involved in its development, then I don’t want to hear about it. Cold Hard Football Facts is stats-geek heaven.

  13. With Favre leaving the NFL and Newton joining it, the three worst QBR scores in 2011 will conveniently be located in the same city.

  14. Wow, ESPN is actually doing something worthwhile? This could be a pretty good stat if it’s handled correctly.

    Knowing ESPN, number of Twitter Followers for each QB and the Thursday Night Football rating each starting QB pulls in will be part of the equation.

  15. That the NFL began tabulating convoluted QB ratings almost a full ten years before they tracked sacks tells you everything you need to know about the league’s statistics department.

    I’m skeptical of ESPN’s plan, but I hope to heck they succeed.

  16. If they’re going to that extent, might as well breakdown INT’s into 2 catagories (like a wild pitch or passed ball in baseball)…. why penalize the QB for balls that hit the receiver right between the numbers or get tipped up and intercepted.

  17. Why do I get the feeling it’s going to magically rewrite Dilfer as a non-terrible QB?

  18. the top quarterbacks of 2010, according to Total QBR, were Tom Brady, Peyton Manning, Matt Ryan, Michael Vick, Aaron Rodgers and Drew Brees. The worst quarterbacks, according to Total QBR, were Derek Anderson, Brett Favre and Jimmy Clausen.
    _________

    That’s exactly what the current passer rating says are the top QBs and the worst QBs.

  19. What about a kneel down, that loses a yard doesn’t it? How about a spike to kill the clock? I really think this is over nit picking a system that works just fine.

  20. Reading the comments above really makes me wonder what it would have been like if there was the internet when things like the sack became stats, or the three point line was put into basketball.

  21. I like the concept and it sounds pretty reasonable. Looks like a likely improvement over QB rating.

    My problem is that QB rating, and even QBR, are team statistics. But we name them just after the QB. I mean, does this really add that much that passing yards, 1st downs, and offensive points don’t already convey?

    I have a question. What are the top 10 QB’s in passing yards, TD’s, and offensive points? If that list looks just like the QBR list then I have to ask what the point is. You take a bunch of stats that all add towards the goal of scoring points. Why not just look at offensive points at that point?

  22. These stats are pretty close to meaningless. This is a team sport, not individual. Every other player will effect the rating. QBs surrounded by talent will look better. O-line issues will make any QB unproductive. Hands of stone receivers factor in also. Stats can be manipulated to say anything you want them to. Why account for one factor but ignore another? (Because it skews the data away from an intended representation) There are too many minute details that are subjective and therefor unreliable. Are fantasy stats that big of a deal?

  23. udub says:

    Reading the comments above really makes me wonder what it would have been like if there was the internet when things like the sack became stats, or the three point line was put into basketball.
    …………………………..

    comparing simple stats to this espn joke, QBR! its not a good comparison. Sacks simple! 3 point shot explains itself!

  24. i am all for a new rating system but if espn has a hand in it “no thanks”. why don`t you stick to baseball espn since even though football season has unofficially started you keep hammering baseball crap down all our throats. its perfect for you. a network and a game thats past its time and very boring.

  25. house068 says:Aug 1, 2011 5:31 PM

    Yes… so Josh Freeman is one of the top QBS in the league… Right…

    —————————————————–

    Obviously, you haven’t seen him play. The guy is a beast. Believe me, he is the reason for 10-6 Bucs team.

  26. Nothing ever goes wrong when ESPN decides to block off time in their schedule to announce some kind of “decision”.

    Also why is ESPN advocating this and not the saber guy who came up with it. If this is the brain child of Trent Dilfer, then there’s gotta be something better. I heard him on a podcast say that to be a good QB like Sanchez you had to literally want the keys to drive the car. Is that taken into account in this stat? Also are they going to trademark the stat?

  27. Will this system and points for playing for New England and the two New York teams. Because then they could justify even more stories about those teams. Considering we have seen a report from all three of those camps everyday since they opened , yet i have not seen one report from tampa, the Panthers, the Falcons, The lions, the Raiders, Not even the Saints and the only time they have showed the steelers it was to talk about a guy who ended going to a New York team …..

  28. I’ve developed my own QB rating system. It takes into account the terribleness of the team that the QB is on, and the selected QB’s natural ability. For example, under my system, Tom Brady and Peyton Manning get one point every time they throw a TD. Jimmy Clausen and Colt McCoy get 100 points every time they throw a TD. So far, Manning and Brady are still way ahead.

  29. I don’t care what changes they make as long as they normalize the score so that 100 is the highest possible score.
    You can do it w/ the currecnt rating by multiplying by 100/158.3. It’s very simple.

  30. The reason why Osi’s flipping out is because he claims the Giants promised to re-negotiate the deal by his third year.

    ________________________________

    In that affidavit, Umenyiora swore that back in 2008 Reese promised him that by now, “if I was currently playing at a high level, we’d either renegotiate my current contract so that it would be equal to that of the top five defensive ends playing or I would be traded to a team that would do that.”

    _________________________________

    As far as Osi’s current comments are concerned, he can negotiate a gauranteed contract, but would have to take less money. See Peyon and Plaxico’s contracts.

  31. I like the idea of this new stat but it’s still flawed. It doesn’t adjust for the quality of receiver being thrown to. But it’s still better than the previous stat.Of
    course the real problem with this new stat is that it is subjective – define “under pressure”

  32. udontknowjaq says:
    Aug 1, 2011 6:04 PM

    comparing simple stats to this espn joke, QBR! its not a good comparison. Sacks simple! 3 point shot explains itself!

    ————————————————–

    I’m not comparing sacks to this as a stat, or the 3 point line to this as a concept. I’m talking about the reactions to these things if the internet was around then. It would have been similar. People don’t like change. Everything was new once. The sack didn’t count, and then it did. No one would think about it not counting now since its established, but when it first came about if things like Pro Football Talk were around people would be giving the thumbs up to comments bashing it.

    Why not let things play out and see if the TQBR becomes legit or becomes a glowing hockey puck gimmick?

  33. And you needed a complex stat to determine that Favre, Anderson and Clausen were the worst?

  34. Woohoo! 2 of 3 bottom feeders (Clausen, Anderson) are Panthers (Anderson as of about five mins ago walking out to Panthers’ practice field). Save us, Cam!

  35. There is only one stat that matters: wins vs losses.

    Last year the Chargers had the #1 offense and defense and look what their season was like.

    It may be fun to talk stats when you’re drinking beer with your buddies on game day but they are meaningless otherwise.

  36. udub says:
    Aug 1, 2011 7:31 PM
    udontknowjaq says:
    Aug 1, 2011 6:04 PM

    comparing simple stats to this espn joke, QBR! its not a good comparison. Sacks simple! 3 point shot explains itself!

    ————————————————
    I’m not comparing sacks to this as a stat, or the 3 point line to this as a concept. I’m talking about the reactions to these things if the internet was around then. It would have been similar. People don’t like change. Everything was new once. The sack didn’t count, and then it did. No one would think about it not counting now since its established, but when it first came about if things like Pro Football Talk were around people would be giving the thumbs up to comments bashing it.

    Why not let things play out and see if the TQBR becomes legit or becomes a glowing hockey puck gimmick?
    …………………………………………

    I understand what your saying but this stat has stats within stats & stats we can’t even see for ourselves all these things your talkin bout you can judge 7 see for yourself

  37. don`t know where fitzpatrick ranked overall but i hear he was just slightly behind peyton manning in “size of forehead”.

  38. WOW!!!!! Man does ESPN wish they could do things to themselves that our girls(ladies, guys do 4I u) do for us. They are so obsessed with themselves, they have to make their own way to grade a QB, so then they can say oh this QB is great cause we are ESPN, and we say so. The QB is only as good as the men that surround him, ask a great QB what made him great, bet he says his teammates. Not some bull sh** rating. GUYS HELP? WHY DOES ESPN ALWAYS HAVE TO EMBARRASS A GREAT GAME, LIKE COMING OUT WITH SOMETHING SO DAMN STUPID, JUST SO THEY GET THE PUBLICITY. IT SICKENS ME PEOPLE WATCH THESE NO GOOD PROMPT READERS, THAT DON’T KNOW THEIR ASS FROM THEIR ELBOW ON ESPN. CAN SOMEONE WITH MONEY, PLEASE GIVE US A NETWORK THAT SHUTS THESE, THINK THEY KNOW MORE THAN US, IDIOTS UP. US TRUE FANS KNOW MORE THAN THESE DUMB ASSES WILL EVER KNOW.

Leave a Reply

You must be logged in to leave a comment. Not a member? Register now!