Todd Haley: Advanced NFL Stats Coach of the Week

On the road against the heavily favored Colts, Chiefs head coach Todd Haley began the game with some unconventional calls and received some heavy criticism. He started the game with a surprise onside kick and went for it on 4th down on the Chiefs first offensive drive.

The surprise onside kick is, statistically, almost never a bad idea. At the start of the game, we can just use expected points to analyze the decision. Surprise onside kicks are recovered 60% of the time. Last fall, I wrote (forgive the self quote):

"The EP for a failed onside attempt is -2.1 pts, and the EP for a success is +1.2 pts. At first glance it appears onside kicks are always losing propositions. But don’t forget that you’ve always got to kickoff somehow, and a normal kickoff averages -0.7 pts for the kicking team...Solving for the break-even success rate, where the combined expected points of an onside kick equal that of a normal kick, we get...42%."

This time it didn't work out, but it was still a good call. My only criticism is that this was the second time in two weeks Haley called for the surprise onside kick. Plus, if there is one team in the galaxy that's on guard for surprise onside kicks, it has to be the 2010 Indianapolis Colts.

Later in the first quarter, after an 8-minute-long grinding drive, Haley gave the go-ahead for a 4th and 2 conversion attempt on the Indianapolis 8. We already know it's a slam-dunk good idea in normal situations, but let's do a quick check using WP.

Two-yard conversions are successful inside the 10 55% of the time. A successful conversion gives the Chiefs at least a 1st and goal at the 6, worth a 0.56 WP. A failed conversion attempt gives the Colts a 1st down at the 8, and is worth 0.41 WP. On net, the conversion attempt is worth 0.49 WP.

FGs are good from the 8 95% of the time. A successful FG ties the game and gives the ball back to the Colts after a kickoff, worth 0.47 WP to the Chiefs. A miss gives the Colts a first down at their own 15, worth 0.40 WP for the Chiefs. On net, the FG attempt is worth 0.46 WP. 

Going for it was probably the better call, especially if you think the Chiefs were underdogs. Underdogs need to take chances to win, and Haley seems to understand that. Congratulations, Todd. You've been named the Advanced NFL Stats Coach of the Week. Send me your shirt size, and you'll get a free one of these:

  • 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

11 Responses to “Todd Haley: Advanced NFL Stats Coach of the Week”

  1. Todd H says:

    I'm an XL.

  2. Anonymous says:

    Meanwhile, Jim Caldwell on the previous drive again went ultra-conservative and kicked the FG on 4th and goal from the 2. My guess is that if you checked, the Colts have made overall the worst 4th down decisions on the season. Might be interesting to look into.

  3. Borat says:


    Did the Spagnulo coach of the San Luis Osprey teems get any votes?

  4. Zach says:

    I like how both Missouri teams tried surprise on-side kicks.

  5. Anonymous says:


    Something is going on with the onside kicks.

    I am aware of 3 suprise on-side kicks this season.

    Dal/Chi, KC/Indy, and STL/Det.

    None were successful.

    You pegged the probability of success for each attempt at 62%. If 62% is the correct probability, there's only a 1 in 20 chance of all 3 failing.

    What do you think is going on? Are there other surprise onside kicks I'm unaware off that have been successful?

  6. Jeff says:

    Very interesting. There are a few things going on that would tend to confound the analysis of surprise onside kicks. There have been relatively few attempts, so if they are becoming more frequent would the receiving team be more prepared and recover more of them? Another point to consider is that the threat of an onside kick might cause the receiving team to get less yardage on the return. The traditional response to the threat of an onside kick is to put players who can handle the ball on the line. This is a terrible formation for actually returning a kick. If a team has a reputation for the surprise onside kick, this might also help their coverage on kicks where they don't attempt the onside kick.

    It would be interesting to see a team that mixed it up enough to cause these secondary effects, particularly if the shifted formations prior to the kick or attempted surprise onside kicks from different formations.

  7. James says:

    It may be that the Saints made every team more aware of onside kicks after the Superbowl. Or maybe one or more of those attempts were poorly executed.

  8. Brian Burke says:

    It's possible the cat is out of the bag. It would still be good to do some proportion of the time to keep receiving teams from dropping into their blocking schemes right away.

    The 60% rate is from the 2000 through 2008 seasons.

  9. Anonymous says:

    Kansas City had a successful one earlier this year that was called back because of an offside penalty.

    I don't buy that 62% figure anymore though. The 100% surprise is gone thanks to the Saints. You won't find too many players sleeping or trying to get an early jump running back to block.

  10. zlionsfan says:

    It can also be a sample size issue: three is, of course, very small, and the actions of a single player can weight those results heavily.

    For example, in the StL-Det game, it looked like the Lion who recovered the onside kick simply reacted quickly (I couldn't tell for sure because the weak production missed it - they didn't focus on the kick at first and actually had the Rams up 10-0 to start the game, ha ha - and then didn't show a replay). I think the Rams had noticed that the Lions' front line tended to turn too quickly when the ball was kicked; in this case, one player (or two) didn't.

    Having said that, you'd guess the surprise factor will be significantly lower the next couple of weeks or so ... the only difference would be if overconfidence outweighs preparation. "The two kicks last week didn't work, they wouldn't possibly try something this week, would they?"

  11. Patrick Haskell says:

    With respect to the 2-yard conversion rate, does it differ inside the opponent's 10-yard line, when the short field (i.e., no risk of a downfield play) allows safeties to play closer to the line and defend both the run and short pass better? My expectation is that the conversion rate is lower at that part of the field, and thus a 0.03 WP differential based on full-field data may be washed out or more.

Leave a Reply

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