Success Count

Most readers are familiar with the Success Rate (SR) stat. It's pretty straightforward. SR is the percentage of plays in which the result of a play results in an increase in net scoring potential, defined by an increase in Expected Points (EP). SR makes sense as a metric for running backs, quarterbacks, and receivers, and it certainly makes sense for team offenses and team defenses. But it doesn't make sense for individual defensive players. In its place, a slightly different variant of the success metric, called Success Count (SC), has been added.

On all the individual defender stat pages here, I had included SR simply because I already had it coded for teams and offensive players. I thought it might be interesting and possibly reveal something, but no insight materialized. SR for defenders doesn't make sense for the same reason that total WPA or EPA, as derived from play-by-play descriptions, doesn't make sense.

For example, consider two plays involving a free safety. In one play, the safety blitzes, strip-sacks the QB, and returns the fumble for a touchdown. That would be a success under any definition. In the other play, a running back slices through a wide hole on the strong side of the formation, breaks three tackles, and streaks down the sideline until the safety saves the day with a shoestring tackle. Our hero safety would have a SR of 50%, which tells us very little about what really happened.

Instead, let's just add up all the sacks, interceptions, passes defended, QB hits, forced fumbles, fumble recoveries, tackles, and assists that result in team-level successes. Success Count is that simple. The regular season leaders for linebackers in 2010 are Lawrence Timmons, Ray Lewis, London Fletcher, LaMarr Woodley, James Harrison, Cameron Wake, and James Laurinaitis. Not a bad group. SC can tell us a lot more about quality performance without penalizing a player for making plays in some situations.

Put simply, SC is the same concept as SR, but it's a simple counting stat rather than a rate stat. Another way to think of it is "positive plays made."

We could do some other clever stuff with SC. For example, we could calculate 'Success Factor'--something analogous to Tackle Factor (TF). TF measures the proportion of a team's tackles a player makes, which accounts for the fact that worse defenses allow longer drives and make more tackles. TF also compares each player's proportion of team tackles to how many tackles his position should typically expect, which accounts for the fact that some positions will naturally make more tackles than others. Success Factor would account for the same considerations.

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

13 Responses to “Success Count”

  1. Dave B says:

    I'm not sure success rate doesn't come with a lot of noise from the situational sage of the player. Specifically for RB'S, WR's and TE's. Success rate over average might tell you something more. Take the theoretical RB that only comes in on short yardage 2yards or less to go. On an average team his success rate would be 60%.

    I like the success rate stats for teams and QB's but find it marginally useful for skill players. RB's are hard to compare as well since their receiving and rushing success rates are lumped together

  2. James says:

    I'm thinking along the same lines as Dave. I was wondering if there was a way to compared a runningback's results versus league average results on a play-by-play basis.

    Piggybacking Dave's example, if the short yardage RB was successful on those plays 70% of the time, then he'd be +10%. The opposite situation would be a draw play on third and long, the RB's success rate is going to be way lower but it doesn't really tell you how he does compared to average.

    Sample size would be a big problem, especially at the extremes, and it would take quite a bit of coding and parsing, but I think it would be worthwhile to see the results. Obviously 1st and 10 would be the easiest to compare due to sheer numbers.

  3. James says:

    On a completely different topic, would it be possible to get a play-by-play tab on the game pages that include the WPA and EPA numbers? I know you can scroll over the chart and get some of the numbers but it's hard to get all of the plays, especially towards the ends of games.

    It'd also be really cool if you could sort the plays by the player involved and see if one play killed them or they stunk it up the whole game.

  4. Brian Burke says:

    Good suggestions. If I had the time, I'd do it all! I'll get there eventually.

  5. Ian B says:

    I second the idea to get a play-by-play grid that shows all of the plays and the changes in WP. I would love to sort the grid by both the absolute value of the change and the net value to see which plays broke open games, in total and for my team.

  6. Joseph says:

    @Dave B. & James:
    I know that Brian doesn't agree with everything that Football Outsiders does, but what your asking for sounds like seasonal DVOA for individual "skill players." They also have a minimum number of plays for a player to be ranked (iirc, 200 passes for QB's, 100 carries for RB's, and 50 targets for receivers), thus culling out the "situational" guys. Although I don't believe I've ever seen it, I believe they also rank those players players who get between 50% to 99% of those baselines in a different group, enabling you to compare situational players/backups to each other--with the "small sample size" caveat.
    It's not the same as WPA or EPA, but it's something.

  7. John says:

    What about when a player blows a play though, like say a interception attempt that was thrown right at the defender who simply muffs the play and misses the the catch. He should be penalized for that.

  8. Brian Burke says:

    Couldn't agree more. But that stuff just isn't in the play descriptions we get. On the other hand, he kinda-sorta is penalized by SC for blowing what would otherwise be a success, because he doesn't get a 'success' for that play.

    (Actually a muffed int would probably go down as a 'pass defended' and would be a success, as it should be depending on your point of view.)

  9. John says:

    Good point, but to go even further then maybe SC should be weighted, i.e. Interception worth more than a pass defended. Just a thought.

  10. Brian Burke says:

    That's what +EPA does.

  11. John says:

    Oh yeah. Oops. New to the site, still learning.

  12. Sampo says:

    I want to know how are the most succesful running backs!

    SC, +EPA, +WPA for RBs!

  13. DY says:

    Looking for the Titans to pull a big upset against Broncos.

Leave a Reply

Note: Only a member of this blog may post a comment.