Attention All Offensive Coordinators

Fourth down and short in the middle of the field does not mean you need to run a goal line play with the jumbo package. The average play in the NFL gets 5 yards. What's the completion percentage for short passes in this league, like 75%?

Line up and just run a standard play. Use play action. Throw an intermediate or deep pattern once or twice a season to keep defenses honest. How about a bubble screen if the corners don't press. And if they do, let them get burned.

Do not telegraph your intent to plunge for the line of gain. That only makes the defense's job easy. Spread out and force the defense to defend the entire field and your success rates will improve. Punish the defense for crowding the line of scrimmage. Think of fourth and short as an opportunity to know exactly what the defense is going to do and take advantage of it.

On the goal line, it's different. There is very little of the field to defend, and yards gained beyond the goal line are irrelevant. Jumbo packages and plunges make a lot more sense.

  • 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

26 Responses to “Attention All Offensive Coordinators”

  1. Ian Simcox says:

    This is one thing that's always amazed me. Offensive coordinators are some of the most creative thinkers around, and yet they come out with such vanilla plays on 4th down. It's like they are being paid to make 4th down look like some kind of special situation that's so much harder than any other down.

    Why do Madden players go for it all the time? Because they pick normal plays. If you were forced to goal-line plunge every time, you'd never go for it.

  2. Eddie says:

    Madden references aside... Football statistically speaking is not baseball. There isn't enough sample size for a lot of data comparison. Not every offensive line is the same (I know this all too well because I am a Bears fan). There are situational comparisons that statistics as they are currently given do not account for. Football is not a rate sport like baseball and basketball. There are too many things that can't be accounted for.

  3. Anonymous says:

    Eddie, when you have league wide data, that might change the conclusion for SOME teams, but not all. The goal line data for teams is terrible. The strategy of running against 8 in the box is proven to be terrible. The conclusion might not be that ALL teams shouldn't use goal line, but that MoRE teams should. That much is basically proven and intuitively correct as well.

  4. OnSolThree says:

    They pull this junk, then they go "See, going for it on 4th down is too risky! Stupid statistics saying otherwise!"

  5. Ian Simcox says:

    Just to dive a bit more in this (using the amazing Game Play Finder over at P-F-R), in 2011 teams gained a yard or more on 86% of all 1st-10 plays. On 3rd-1, that drops to 77%. The average gain on those plays gaining at least one yard on first down is 8.72 yards, on the 3rd downs it's 6.21 yards.

    I don't believe defenses let offenses have those yards easier on first down and then 'step-up' on 3rd and 1. Professional sportsmen tend to play at their hardest on every down. With a lower success rate and a lower gain on successes, that tells me that offenses aren't using all the field in short yardage situations (for more evidence, 6% of 1st-10 plays go for 20+ yards, compared to only 3% of 3rd-1).

    As for the argument that not all offensive lines are the same, well in 3rd and 1 situations last year the worst team for converting them was St Louis and they managed to convert 64% of them. Moving to 4th and 1 plays, league wide 54% were converted, with those successes averaging 4.88 yards and barely 1% of them going over 20 yards.

    The pattern is beginning to emerge then, that the more you call short yardage plays that result in lower gains even when successful, the more your success rate drops. The answer - call normal plays. If they're good enough to get you 10 yards they are good enough to get one.

  6. Anonymous says:

    The success of the jumbo package basically rests on the ability of the offensive line to out push the defensive line. It would seem that if you have a weak offensive line (Bears), it would be advantageous to run normal plays. It's unlikely that your line will be able to out push the stronger defensive line.
    I think the selection of the jumbo package is just another example of risk aversion. If you run a jumbo package you're looking at a gain of 1 or 2 or a loss of 0 to 1 yards. If you fail, the critics will blame the offensive players for not pushing harder not the play call. If you run a normal play and fail (maybe even lose a few yards), the critics will question the play call. So form a coach's standpoint, it better to call the conventional play and have the players get blamed for failure instead of calling a normal play and have the play call (the coach) blamed for the failure.
    Les

  7. Anonymous says:

    At least some teams have the brains to go for it in the first place, even if the play calling trend is bad. The Broncos had a 4th and 2 around the Bucs 45 today and punted. I think that decision making bothers me more.

  8. Sean S says:

    Are you talkin' 'bout Da Bears or something?

  9. Anonymous says:

    Great point again Brian...Seattle did exactly this on 4th & 3 late in game 6 yard pass over the middle..& kept the game alive!

    The big problem is that going for it on fourth down is such a novelty for almost all coaches they don't have a long metal history of the feel of it...its too new & they are spooked by it..give them 10 years of seeing it and doing it then they will evolve in the meantime the few coaches ahead of curve can feast..

    By the way..I ran a regression on the 5 key efficiency stats to wins to get a basic expected wins for the the last 12 years.

    Almost all teams are within 1 game per season
    (expected to actual wins) Only one team outperforms
    their expected win total by 2 wins a season...Any guesses?
    Bill B. of course. (FYI Detroit had the worse record -1.5)

    So is he a) lucky
    b) cheating
    c) doing something that the efficiency stats aren't picking up..game management/play calling etc)

    Andy Reid was right on total 0 diff.

    Dan

  10. Brian Burke says:

    Ian does the legwork. Excellent!

  11. Anonymous says:

    Ian, I disagree with you on doing normal plays. On 4th and short, the 1st yard is infinitely more valuable than yards 2-10. So a 75% shot at 1 yard is better than a 50/50 shot at 10 yards or better. That one yard is so valuable that teams should risk lowering their chances at 2-10 yards in exchange for a higher 1+ yard chance

  12. Anonymous says:

    I see the Jumbo plays to be a lot like bunting in baseball. Most SABR types hate it and it's used way more often than it should. While it increases your chances of scoring one run, it hurts your chances of scoring more than that. Coaches use the Jumbo because they think it is more certain way to pick up one yard. There's more to 4th and 1 than just a yard, because if you don't get it, the penalty is bigger than 1st-3rd downs. The risk and payoff are different on 4th down than the other downs. That's something you need to consider on 4th down. Since you only need 1 yard, stability or reliability is more important. 20, 30 years ago the Jumbo plays were probably more reliable, but anymore, they are not.

    I know Brian has talked that coaches are motivated by SR more than Yards/play numbers. What are the SR numbers by down? EPA? WPA? That may give us a better picture than just yards per play.

  13. Anonymous says:

    just one point. Defenses play it differently. On 4th and 1, they will very definitely attempt to stop the 1 yard.

    On 1st and 10, they basically want to keep the offense to 4 yards or less.

    It is not appropriate to apply 1st and 10 situations and outcomes, to a 4th down situation. There in no reason to think the same outcomes will hold.

  14. Ian Simcox says:

    @Anonymous: That's the problem though. Defenses know this too. By refusing to go for longer plays, offenses also reduce their chances of gaining 1+ yard. There is no trade-off. It's not a case of "giving up a chance of 10 yards for a greater chance of 1 yard". It's actually "giving up a chance of 10 yards for a lower chance of 1"

  15. Ian Simcox says:

    @Anon re: just one point

    But if defenses lined up like it was 4th and 1 in a 1st and 10 situation, how would you play it if you were the offense?

  16. Robert Bilotta says:

    These posts are why I like this blog. The play calling on 4th and short has always bothered me.

    Even if disregard statistics, shouldn't common sense or intuition tell you if you are going to be predictable, you will lower your success rate. Instead of always trotting out the jumbo package, try doing something different. And my biggest irritation is when there is a slow-developing run play, against a stacked 8-in-the-box defense. You think that is a quality choice? I am guessing the offensive coordinator just call these plays automatically.

  17. Anonymous says:

    I would be fascinated to see the defense react to the offense sending out the 4 wide receiver package on 4th and inches. They'd probably be forced to burn a timeout or take a 12 men penalty trying to make the last minute subs

  18. ff says:

    Anonymous, they would play it like how teams play the Patriots on 4th and short and Brady lines up in 5 wide. Amost guaranteed QB sneak so they play man to man and bring everyone really close to the QB to stop the sneak.

  19. Ian Simcox says:

    Ff, ok so if that's how you stop the 5 wide set, why don't defenses do it on 1st, 2nd or 3rd down? Because if its not a sneak, you are in trouble.

    And that's the point. If the defense knows what's coming its easy for them. So mix it up. Make them defend like its 1st and 10.

  20. Howard says:

    What is especially maddening is some of these same coordinators will go to empty backfield on 3rd and 1- which again makes them predictable.

  21. Howard says:

    Another thing have been thinking about. Would be interesting to see what the run/pass ratio is on two point conversions compared to same distance on regular 3rd and 4rth down plays.

  22. Anonymous says:

    Monday, December 03, 2012

    @Anon re: just one point

    But if defenses lined up like it was 4th and 1 in a 1st and 10 situation, how would you play it if you were the offense?
    _____________________________________________________

    on first and 10, you can throw it 50 yards and miss. On 4th and 1, if you throw it 50 yards and miss, you take a huge hit in Win Probability.


    I think people get confused when looking at expectation values of plays, and thinking, if we reran this exact scenario 1000 times, then it would be marginally better to go for 4th and 1 all 1000 times.

    in reality, that play only happens once. And there are only 16 games in a season. Coaches need to avoid huge drops in winning percentage, more than they need to optimize expectation values.

  23. Ian Simcox says:

    Well that's actually a different argument you're making there. We are talking about how best to maximise your chances of converting, and if the defense comes up 8 in the box and press coverage, then you have far more chance of converting if you throw over them than if you run into them. Do that a couple of times, opposing defenses will learn that you are not scared to throw for a 4th down, will have to back off a bit and your rushes will have more chance of succeeding.

    As for avoiding huge drops in win probability - say you have the ball at midfield, 4th and 1 in a tie game with 7:30 left in the 4th quarter. Your WP if you punt for a touchback is 0.47. If you go for it and fail, turning the ball over 30 yards further up the field, your WP drops to 0.31. But, if you succeed in your conversion attempt, your WP is 0.69. If huge drops are to be avoided, huge gains must be sought after by a similar logic.

  24. Anonymous says:

    I think another point is that you are often leaving up to a ref's somewhat blind spot when running jumbo or running a sneak. Yeah you can challenge, but often the replay views are inconclusive on these type of plays.

  25. Anonymous says:

    As for avoiding huge drops in win probability - say you have the ball at midfield, 4th and 1 in a tie game with 7:30 left in the 4th quarter. Your WP if you punt for a touchback is 0.47. If you go for it and fail, turning the ball over 30 yards further up the field, your WP drops to 0.31. But, if you succeed in your conversion attempt, your WP is 0.69. If huge drops are to be avoided, huge gains must be sought after by a similar logic.

    --------------------------

    In my opinion, that does not follow at all. Granted, increasing WP is obviously the goal. And, in your scenario (or any) many many factors go into the decision of punt/'go for it', but when faced with the choice of maintaining .47 or having a 50/50 chance at dropping to .31, that maintaining .47 is not the wrong choice.

    The strategy of avoiding drops in WP is valid.

    Thing is, the probability of a successful conversion is unknown, so one cannot make a quantitative determination. This site assumes this probability is known, and therefore it can do arithmetic and come up with some result like punt = .47 WP and go for it = .53 WP therefore go for it is the right answer, but the assumptions are not valid.

    I'd also point out that the actual WPs do matter. A team can certainly risk a .60 or .90 WP, where the worst outcome still leaves you ahead. Or the other situation, where you risk .10 WP vs a .30 WP, where you are likely to lose anyway and you are desperate. But scenarios where you risk likely-to-win for likely-to-lose should be avoided. Even if the expectation value of 1000 trials indicates a small advantage for risking it.

  26. Daniel B says:

    "The average play in the NFL gets 5 yards"

    Average gain isn't the relevant stat here. What percentage of plays gain more than X yards is what you need. Although in both cases the stat won't account for the D going more all-or-nothing than usual.

Leave a Reply