AFC Championship Analysis

Ben Roethlisberger had a rough day passing the ball yesterday, posting just 1.5 Adjusted Yards Per Attempt (AYPA). But he made critical completions when it mattered most, and he made a big difference on his 4 scrambles, converting 3 first downs and a TD. Ultimately, BR put the game away with his final two throws, each for 14 yds and each converting first downs. Those throws were worth 0.06 WPA and 0.08 WPA alone. BR ended up with a 0.29 WPA for the game, 5.0 EPA, and a 51.6% SR. 5.0 EPA doesn’t exactly scream MVP, but then again, when the final whistle blew, PIT won by exactly 5 points.

One big reason for the win was the Steelers’ rushing attack. Rashard Mendenhall ran at a 4.5 YPC pace, accumulating 0.13 WPA, and 3.9 EPA. A better job of tackling would have really changed the complexion of the game, as Rex Ryan lamented in his postgame press conference. Removing BR’s scrambles, the Steelers only had a 42% SR on their running plays, which is right at league-average. Contrary to conventional wisdom, PIT is not a great running team, but the Jets allowed Mendenhall to gain too many yards after getting through the first level of defenders.

Should the Jets have attempted an onside kick after scoring to make it a 5-point game with 2:56 left to play and all 3 timeouts remaining? The Steelers were clearly expecting it, so we need to go with the 20% success rate rather than a 60% success rate for unexpected onside attempts. A success is worth 0.31 WP. A failure is worth 0.10 WP, which was adjusted for the fact that the Jets had 3 TOs. That makes the break-even probability of success 24%, which is more than the Jets could expect. Kicking deep was the right call, and having all 3 TOs is what tips the scales.

Mark Sanchez played well with a 6.3 AYPA, only 2 sacks, and no interceptions, but he couldn't make the plays when they counted most. He had only 0.04 WPA and 3.2 EPA.

Could the Jets have been more aggressive on their 4th downs? Not really. My numbers show only one bad decision, on which only a crazy person would agree with me. In the 2nd quarter, down by 10, they had a 4th and 7 from their own 43, which would have net the Jets only an additional 0.016 WPA.

The stop on 4th down at the goal line didn’t turn out so bad for the Jets. It actually turned into 9 points. The safety gave them 2, and on the subsequent possession, they ended up scoring 7 more. It’s a good case study of why coaches overlook the value of 4th down conversion attempts. Sure the Jets would have preferred a TD in that situation because they were down by 14 well into the 4th quarter, but generally speaking, it’s a good example of why handing the ball over deep in opponent territory isn’t always such a bad thing.

Another point about 4th downs in the game is that one of BR's interceptions came on a 4th and 1 at the NYJ 32. On 4th down, turnovers aren't such a big deal. You've got very little to lose by forcing a pass if necessary and giving a receiver a chance to make a play. It's partly why I think coaches underestimate how successful 4th and long attempts can be. It's also one reason why BR's WPA was so high despite having a very poor AYPA.

Advanced Individual stats for
PIT: offense and defense

NYJ: offense and defense

  • 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

22 Responses to “AFC Championship Analysis”

  1. Ian B says:

    The jets defense link is taking me to the steeler's defense page.

  2. Anonymous says:

    Can we discuss the ridiculousness of the Jets calling a timeout after the Steelers first play following the kickoff with less than 3 minutes left in the 4th quarter? Why haven't coaches realized that you shouldn't call the timeouts until you HAVE TO? Calling a timeout there is a waste if the Steelers convert for a first down...which they did.

  3. Brian Burke says:

    Link fixed. Thanks.

    I've heard the complaint about the Jets' first timeout a few times today. Maybe I'm missing something about the Jets' first timeout. It makes sense to me.

    If you don't take it, and you do get a stop, then you just cost yourself 40 seconds.

    If you do take it, and you don't get a stop on the first series, what has it cost you? Those 40 seconds are either going to be burned on PIT's first series or on their second, right?

  4. Doug says:

    Re: the timeouts, I think it's less an issue of "waiting" and more an issue of being smart regarding the two-minute warning. The value of a timeout can be measured in the amount of clock you prevented the opponent from bleeding off if you hadn't called the timeout, i.e., how early in the play clock you can get the timeout called. The two-minute warning is essentially a timeout, so it's useful to try to plan to have the two-minute warning occur as early in the play clock as possible to maximize its value. A timeout after a play ending at 2:38 is MUCH more useful than a timeout after a play ending at 2:43---in the latter case, your opponent is going to have to run another play before the two-minute warning anyway, and the two-minute warning should occur almost immediately after the next play, which is ideal.

    As it pertains to yesterday's game, by taking the strategy they did, the Jets got the Steelers into 3rd down at 2:00 with no timeouts. But by skipping their second timeout at 2:43 like I suggested above, the Jets could've instead forced the Steelers into 2nd down at 2:00, but with the Jets holding two timeouts, which is obviously a better situation.

  5. Ian Simcox says:

    On 4th down turnovers - the really beneficial thing is that most defenders are stat hogs and will take the pick to pad their stat line rather make the right play and slap the ball down on 4th down (and as Marlon McCree can tell you from the 2006 playoffs, sometimes returns end up in a lost fumble).

    We all know this play:

  6. David Gurian-Peck says:

    Should the Jets not have gone for 2 after their final touchdown?

    If they had, how, if at all would, that have affected the decision to kick it deep on the ensuing kickoff?

  7. James Sinclair says:

    Before that pass on 3rd-and-6 that put it away, I think I remember Simms (or Nantz) saying something to the effect of "they're just going to run it here and run the clock down as much as they can before punting." I'm sure I wasn't the only reader of this site who thought "yeah, or they can throw for a first down and win the game right now."

    I'd bet the numbers support Tomlin's decision to pass, but I'm just guessing. Any dissenting views?

  8. Fitz says:

    I like what Tomlin said about the decision for the final pass of the game: "We weren't going to play not to lose"

  9. Anonymous says:

    Another thing to consider about the timeouts: it seems much more valuable to have them on offense rather than use them on defense. Coaches seem to want to give their offense as much "time" as possible, i.e., use all their timeouts on defense to keep as much time on the clock, but aren't those timeouts put to better use on offense?

    Allowing your offense time to huddle and call a play has advantages over calling the play at the line in hurry-up. And of course, the biggest advantage to having timeouts on offense is that you won't have to spike the ball - that is, you'll have all 4 downs rather than just 3. Of course, there will be less time on the clock, but I still feel like it's not enough to justify calling timeouts so early on defense.

    Try and get a 3 and out without calling a timeout. Of course, there is a threshold - if a 3 and out without using a timeout leaves you with 20 seconds, then a timeout is necessary. Not sure how one would go about caluclating such a threshold.

  10. Anonymous says:

    Forgot to mention...having timeouts on offense opens up the middle of the field as a downfield option as well.

  11. Brian Burke says:

    You only need 1 timeout to keep the middle of the field open for the offense. Otherwise, 40 seconds is a lot of time to watch tick off the clock. A well-run offense can probably line up and call a play in 10 seconds from whistle to snap. That 30 seconds of difference is probably the overriding consideration.

  12. Anonymous says:

    So why do we see QBs spike the ball if they can call the play in the time it takes the refs to set the ball in position? There's something gained in huddling up and establishing a play, and hopefully that gain outweighs the loss of down.

  13. Brian Burke says:

    True. But I would suggest offenses spike the ball far too often.

    A great example is the week 17 CIN-BAL game. CIN needed a TD in the final minutes. They drove to the BAL 8, and then threw a short pass up the middle to the BAL 2. On second down, Carson Palmer spiked the ball to stop the clock. 3rd down incomplete. 4th down incomplete. Game over.

    The only thing was that there was over 20 sec left on the clock. Oops.

  14. Anonymous says:

    How about an in-depth analysis of late game situations involving timeouts/stopping the clock, etc. ?

    It might be difficult, if not impossible, to quantify everything involved, but it should be possible to set up the benefits and costs of given situations in a nice little abstract equation. Then everyone can throw in their two cents.

  15. Jonathan says:

    Re Timeouts: Good point about calling timeout at 2:43 vs. 2:38, but you are forgetting one important factor: if you let the offense run the clock inside of 2:06, you give them the option of throwing the ball with impunity. They know the clock will stop anyway, so why not pass? This is also why it's a bad idea to *call* timeout at 2:04. You save 4 seconds on the clock, but that's not worth the strategic play calling advantage that you have now granted the offense. 2:10 is right around the best time to call timeout & use the 2 minute warning to your advantage (conversely, 2:50 is a great time to let the clock run & force the opposition to snap the ball with 2:10 to go).

    Saving your timeouts for offense is ABSOLUTELY THE WRONG THING TO DO if you are trying to save time. Fact. You save 40 seconds by using your timeouts on defense. You save less time by using them on offense. Note the qualifier, though: "if you are trying to save time." There is an advantage to wasting time/holding timeouts: if you have timeouts at the end of a drive, you can use them strategically to make sure the opposition does not get the ball back with time on the clock.

  16. Jonathan says:

    Re 3rd & 6: I'll take a bite at a dissenting view.

    WP for a team down by 5 with 1:50 to go: 0.12
    With 1:10 to go: 0.07

    42% breakeven is the baseline. However, this doesn't not account for timeouts. The Jets had none, so immediately, both numbers go down. This ultimately increases the breakeven conversion %, since it makes low-variance plays more attractive to the Steelers.

    Just to come up with a guesstimate, I plugged 1:30 and 0:50 into the WP calc, as an attempts to compensate for the lack of timeouts. I came up with 0.09 and 0.05. That's a breakeven of 56%, and I sincerely doubt the actual breakeven was any lower than that. That's a tall order under normal circumstances.

    OTOH, here's one thing that tips the scale back in favor of calling a pass play: Big Ben could have taken a sack without costing his team field position in that area of the field, so he had the luxury of being able to wait until a receiver was wide open. Remember: the 56% number is for 3rd down conversions. It's more of a conversion-to-incomplete pass ratio. So...maybe it was the right call. But a strong case could be made for running the ball.

  17. Jonathan says:

    Re 2 pt conversion:

    I think you just wait to go for two. You're already down five, so realistically, you have to score another TD to have any chance at winning it. Since you already know this to begin with, there is no real advantage to pulling the outcome of the conversion attempt forward in time. About the only disadvantage to going for it immediately is that, in the event of a failed conversion you trail by six. You then expose yourself to the slim possibility of scoring a TD to tie it and then missing the PAT.

  18. Adam says:


    Ran thru that with WPA too re that last pass. I came up with .08 WPA if Steelers punt to 10 and Jets get ball with 1:50 left, vs. .03 if Steelers punt to 10 with 1:10 left. But whatever, similar enough. I think issue though is that its not just implied risk/reward, its game over on 1st down. So between that and Ben can just eat the ball like you mention, probably worthwhile gamble. Especially if you trust Ben to only throw if play is clearly there.

  19. Anonymous says:

    As far as using a timeout on offense versus defense goes, in EVERY case it is more beneficial to use it on defense.

    Think about it, what is the worst thing that can happen? 40 seconds can run off the clock. If you don't use the timeout on defense, the opponent will ALWAYS use the entire 40 seconds. On offense, AT WORST, you can run a normal play (call in a play, huddle up, the whole deal).

  20. James Sinclair says:

    Yeah, the 3rd-and-6 play seems like a case where the analysis would have it about even, but all the situational factors favor passing.
    – The Jets probably expected a run (at least, until they saw Pittsburgh's five-wide formation).
    – Presumably, Roethlisberger's a lot less likely to throw an interception in a situation like that, where the risk substantially outweighs the reward (not that I'm still bitter about how Matt Ryan ended the first half against the Packers or anything).
    – As you (Jonathan) said, a sack is just as good as a run for less than six yards—maybe even better, since it would give the punter more room to work with, and a coverage sack would take several seconds longer to develop.
    – The Jets had the "momentum," to the extent that there is such a thing.

    As for going for two, the most compelling argument I can think of against it is that the gameday line was Pittsburgh minus-4. Just sayin'.

    Anyway, one tangential thought on timeouts. I don't remember if it happened this weekend (seems like it probably did, though), but it drives me crazy how willing most teams are to use a timeout to prevent a five-yard penalty—usually delay of game. Obviously, there are times when those five yards are crucial, but there are other times when a 1st-and-15 or a 2nd-and-12 isn't so bad if it means having an extra timeout at the end of the game.

  21. Anonymous says:

    When exactly does it become a good idea for the offense to stop the clock on defense rather than hold on to the timeouts? Assuming a 1 possession game, should the losing team (on defense) be calling timeouts, say, as long as the offense can run at least a 3rd down play after the two minute warning? Where is the threshold? Would we rather have 4 minutes and no timeouts, or 2 minutes and three timeouts? More time gives more flexibility in offensive play calling. Then again, it also gives more time to the other team should you score TOO early. A lot to consider...

  22. Steve says:

    This is a little tangential, but while we're on the topic of timeouts, I've always wondered:

    2d half, your team has the ball. There's a screw-up getting personnel in, the play clock is at :02, no way to get the play off - so QB calls timeout.

    My sense is that, more often than not, your team is better off taking the 5-yard delay-of-game penalty versus burning a timeout. There are a lot of ways to get 5 yards, but it's impossible to get that TO back. And while I'm sure there are some situations where you want those 5 yards, I think those are few and far between.

    Any thoughts?

    (Apologies if this has been analyzed before, I'm a relative newcomer to this great site.)

Leave a Reply

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