4th Down Briefs

About a year ago I made a powerpoint about when teams should go for it on 4th down. The goal was to have something other than a 4-part blog post or a scholarly research paper that could be used to convince the doubters out there. I never published it because I was never really thrilled with it. I wanted to keep it as short as possible, knowing most people generally don't have lots of time to devote to abstract notions of "Expected Points" in the NFL. The result was an unfortunate eye chart of bullets, sub-bullets, and graphs.

I think it's salvageable with some work, at least as a presentation for a willing audience. But I decided to start over with something much simpler, designed for the uninitiated on the topic. I got the powerpoint religion and abandoned my bullet-ridden format for something I hope is more compelling and more entertaining. My hope is that it's more accessible, and it gets emailed and twittered around a little.

Here is the revised version:


And this was the original which goes into another level of detail:

  • 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

17 Responses to “4th Down Briefs”

  1. Nathan Jahnke says:

    Very nice, I like it.

  2. Nathan Jahnke says:

    To let you know when I tweeted this, it is titled with forIt being one word, and when I click on the link, the title of that also has forIt as one word.

  3. James says:

    I like it, about as straightforward as you can get, and a big improvement over the original. I would have reworded EP somewhat to something like, "Expected points is an average of how many points you can "expect" to score from that field position." Payton's also speaking in some weird 3rd person on slide 28.

    But again: simple, explains the concept, gets your point out there, shows how this would affect the game inside the six and on 4th and 2, and allows someone to find out more. Well done.

  4. Kulko says:

    Its a fine slideshow, but I think it can still use a bit of work.

    I would choose an example which is simpler , like only comparing FG % and going for it with a 4th and 2 from the 12 or such.

    Also I would probably leave out the last three slides and only tell them that when you go into the details with somebody who wants to know. They just sound smartass and you are presenting to people who believe stats guys are to full of themselves anyway.

  5. mermel says:

    I'm not even sure you need an example. The first couple slides asking whether the decisions are made on gut instinct or reasoned analysis should be the focal point. You need to convince people that reasoned analysis should be used before you tell them the best reasoned analysis.

  6. Ryan says:

    @Anon:
    Not sure copyright issues really come into play, unless he's making money off it... certainly looks like fair use to me.

    @Brian:
    Well done, as always. I do wish you had that "bottom line" slide from the Details, though... I think even for number-averse people who need visuals to understand something, graphs can help more than photos. I think a mix between the two would be ideal. (In fact... why hasn't anyone invented a "choose your own adventure" PowerPoint? Let the user choose between "Explain this to me more..." and "I get it, I get it... let's move on." Is that possible? It should be, if it ain't.)

  7. Ian Simcox says:

    Very good Brian. Doesn't tell the site regulars anything new (we're all EP and WP geeks now) but very useful for introducing the concept to people who doubt you when you say "4th and 3 - go for it". The funny thing is that there's such a stigma around failed 4th downs currently, but if coaches went for it more often then a stopped 4th down would cease to be a rare 'momentum turning' event and would become just the norm.

    On a technical question, is it right to consider the average punt landing at the 14, or should you take the average EP of the entire distribution of punts from the 37 (i.e. x% are returned for TD, y% are touchbacks, z% land at w-yard lard etc)? These may well be the same number, I just wondered whether they were.

  8. Unknown says:

    As a spectator I hate when they punt. Boring. I want them to take more risk. This slideshow is a great bit of marketing to help educate all football fans. I hope that one day you will make "the chart" that the coaches need and that one day after that "the chart" will make it into the hands of coaches on game day.

  9. Anonymous says:

    Tres bien! The NFL should hire you to take this presentation along and give talks to all 32 teams.

    Hopefully this would save me tearing my hair off the moment a coach decides to kick on 4th and short in the end-zone.

  10. Q says:

    I think the revised version is actually much better ; you can have too much text on one slide, just one-two or maybe three sentences are enough.

    But i think you shouldn't do the math (1) and talk about the minus ep effect of the kick after a score (2) because : 1- If you can get people interested by your analysis then you can explain the math and show how you can build a chart. Most coaches will be turn off by the maths stuff I think, and (2)- this is a detail and a debatable point i think ( In theory you can say the kickoff is worth -0.3 point, but in a real game 7 point is 7 point; the name of the game is tackle the opponent because teams do have alternate possession so when you try to score you are next suppose to stop your opponent -or any score would be be worthless-)

  11. Q says:

    Oh and !i think the big deal is also that even if you're stop on a 4rh and goal, your opponent is stuck and with real game data you can show that the "goin' for it' team has more chances to score next

  12. Q2 says:

    Oh Ithink one very important thing would be bringin' up the fact that if the "go for it" team fails on 4th and goal, the opponent is stuck and demonstrate with actual real game data that the "go for it" team will eventually scores most of the time.

  13. Anonymous says:

    Looks great!

    BTW: Miles Austin's data doesn't add up on the receiver stats page. His yards and TD's are lower than the official stats. Every other receiver I checked worked out correctly.

    There may be a mistake there

    Keep up the great site!

  14. J.R. says:

    The second presentation's 16th slide is missing a key piece of information: how large is the EP advantage of each choice? Given that the line is the break-even (EP advantage = 0), it must get larger and larger as you move away from the line, but how much larger? Could you re-plot this as a 3-D column graph where column height is the EP advantage, or as an array of dots where dot diameter correlates to EP advantage?

    Pie charts stink, but perhaps a pie chart showing how much of the EP "pie" each option gives you at each coordinate might be the way to go. Right on the line it's 50/50 or 45/45/10. That also gives you a nice way to illustrate how you build the contour chart: show the pie chart for 4th-and-3 at the 37, then "zoom out" to show an array of pie charts. Leave only the largest slice at each dot, and you're left with a three-color contour chart.

    I love the EP analysis - keep up the great work!

  15. Tarr says:

    1st presentation, slide 10: the negative sign is cut off on the far left.

    Good stuff.

  16. Anonymous says:

    Outstanding work. Absolutely outstanding. Kudos to you.

    Only criticism I would have is the frame "On the goal line, offenses should normally go for it anywhere inside the 6." On it's face, that statement is self-contradictory.

  17. Anonymous says:

    My quibble with "adjusting for game situation" is this: If you think your chances of making the conversion is lower than the average, how is punting going to help, if you are still handing expected points to the opposition? Maybe fewer than if your conversion probability were higher, but maybe not. If your team is objectively inferior, that means the other team's expected points are higher. If your team is objectively inferior, you need to take your best shot even if it is not so good.

Leave a Reply

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