Sentences to ponder

From the Pro-Football Reference Blog:

[Imagine w]e have two teams that both average a whopping 14 yards per attempt. One team completes 100% of its passes; the other 50% (for 28 yards per completion). If I were to model those two, it seems pretty clear that the team that completes 100% would score more. They would score on virtually every possession, only failing to score in limited cases where their 3 consecutive completions net 9 or fewer yards. The 50% team would also score a lot, but string together a few more droughts. I suspect my 100% completion team with 14 yards per attempt would average about 60 points a game, while the 50% completion team would average closer to 50.

At the other end of the spectrum, we have two teams that average 3 yards per [pass] attempt. If one of those teams completed 100% of their passes, they would struggle to maintain drives or even get them started [or would routinely end on 4th and 1…], while a 25% completion team would occasionally string together first downs and get into scoring range. Neither would score much at all, but if I were forced to watch both teams for 24 hours straight as punishment for all my transgressions, I’d take the team with the yards per completion to win in a non-shootout.


And later in the piece:

[I]t is pretty clear that the QB’s in a Don Coryell-based offense (Fouts, all of the Redskins QBs of the 1980’s and early 1990’s, and the Rams and Chiefs recently) are undersold by passer rating relative to adjusted net yards per attempt in terms of the value they provided, and the West Coast passers are oversold, and its because of the different philosophies as they affect completion percentage.

  • Interesting–combining those sentences makes me think of a market efficiency metric, where expected scoring for different comp % and Y/A intersects with $ paid to offense (or components of the offense).

  • Tom Rutledge

    Each example has a finance analog.

    For the first point, instead of two offenses, think of two investment strategies. Both average 0%/year (arithmetic average). Over the two years, one strategy delivered 0% in each year–the average of two zeroes is zero. The other delivered +100% in one year and -100% in the second year. That also averages out to 0%–but the second dude lost everything. That’s what -100% means. So, yes, volatility is bad, and not just because it’s scary.

    For the second point, think of a guy who borrowed all his money for a year at 10%. If he invests his borrowed money at 9%, he will always default on his loan because he’ll only have 109% at year end and he’ll owe 110%. But if there’s a chance he can make 115%, he may survive, even if on average he would tend to only make (like the first guy) 109%. (Either way, it’s a bad deal for the lender.)

    On the second point, however: if you know you’re going to make 3 yards on every play, wouldn’t you always go for it on 4th and 1?

    This is taking me back to 4th and 2 at the 2009 Pats-Colts game…I feel sick to my stomach and have to stop.

  • OldSouth

    yay kurtosis

  • Buckeye

    Well said, Tom. I was thinking the same thing re: 4th and 1. Why wouldn’t the team with 100% pass completion percentage go for it with a pass play?

  • Brendan

    All the statistics here are kinda dumb because they don’t include any concept of standard deviation, confidence interval, or attempts per game. Without those measures, I can picture a team that averages 14 yards per attempt with a 50% completion rate but scores 0 points per game (very high standard deviation, attempts per game could be high or low), or a team that averages 4 yards per completion yet scores 40 points per game (very low standard deviation, attempts per game could be high or low). You get the idea.

    My point is that yards per attempt and completion percentage only tell a third of the story; you need to quantify the frequency of “droughts” to tell another third, and the number of attempts per game tell the final third.

  • Jim

    Buckeye- You can change it to 2.4 YPC than, it does not matter because its a thought experiment and no one in reality can predicatively be 100 percent
    Brendan- Since this was a though experiment and he has not yet published any findings yet its premature to worry about those things.

  • Anon

    Even the thoughts so far though offer a conclusion that might be lost on some:

    If you’re going to throw it short, you better be good at it.

  • Mr.Murder

    Playoffs change the percentages, for many reasons. Intervals probably produce the same scoring opportunities as before. All things being equal, consistency numbers would arguably decrease. One system is more of a self correcting one in that its peaks still produce scores. The other doesn’t calculate a loss in planning and when that occurs it is to the detriment of the entire system it utilizes.

    There are going to be situations you must play from the lost yard or lost down aspect. Only one of those two models still works.

  • OldSouth

    Brendan, you’re asking for more context than is necessary for the analysis. Also, you can easily infer the standard deviations from the data given.

    50% completion on 14 yards per attempt = 14 yard SD
    100% completion on 14 yards per attempt = 0 yard SD

    As for a confidence interval or attempts per game, you really don’t need that information for this analysis.

    Humpty dumpty sat on a wall
    Humpty dumpty had a great fall

    You’re arguing that we can’t understand that story because we don’t know why humpty was sitting on the wall.

  • It’s a seductive way of thinking and certainly gives added value to the Gilman / Coryell / Zampese system, but the problem is the same one that gave rise to the Bill Walsh Offense. It’s the probability of completing passes thrown deep down field versus those thrown short or medium range, then getting YAC.

    Remember the objective of the Walsh Offense was to create the possibility of the obviously open receiver, then creating space for the receiver to run. That objective has escaped many of the modern Walsh systems, and indeed the one system where it’s obvious was not created by Coach Walsh. It’s the Colts Offense. Perhaps the most efficient passing system today next to the Saints’ Offense.

  • Joseph White

    If you love R&B music then you gotta love Chris Brown also. He’s been my music idol for the past few years. ”