A yard is a yard is a yard is a yard?

This "Inside the Numbers" examines whether rushing and passing yards (per attempt) matter equally. Conventional wisdom holds that pounding out that extra six inches on the ground matters more than your receiver running his route six inches deeper, what what does the data say, and what does it all mean for the Card? Woody Hayes, prepare to roll over in your grave.

Recently, I discovered that by taking the highly complicated and not at all obvious step of setting my computer's clock back a year, I could convince a certain statistical analysis software program that, no, its license hadn't expired after all. (Don't worry, my school buys us a new license each September anyways, but it's nice to be able to tool around for the month or two in the meanwhile. Though I don't know how much faith I should put in SAS Inc. if they fall for the ‘ole "tell the computer it's 1948" trick.) Anyways, ever since, I've been manipulating the stats in trying to understand more about what really helps teams (and Stanford) win in football, with a special interest in anywhere the hard numbers differ from conventional wisdom. Last episode, we showed that interceptions had double or triple the effect of fumbles on a team's chances of winning, and us fans would be better served by counting a pick as two fumbles, instead of using the accepted one-to-one ratio when calculating turnover margin. Probably good news for the run-heavy Card.

Today, I want to look at the relative importance of excelling at passing versus excelling at rushing, both offensively and defensively. Do note that it's kind of like Emeril here, as I'm crunching the numbers in real-time as I write the article. So POW! for one, and second of all, hopefully we find something interesting like last time, otherwise I don't know what I'm going to do for 1,000 words as our soufflé slowly disintegrates before our eyes. Okay, here goes…

The total yards statistic, second only to points in conventional analysis of a football game, of course implies that passing yards and rushing yards are of the same value. In a most obvious sense, of course, they are – whether you convert third-and-four on the ground or through the air, the chains still move. However, just like a fumble and an interception both give the other team the ball, yet the latter appears to cost your team twice as dearly, maybe there's a difference between the two. Ask any coach, most famous of all Jim Harbaugh's old college coach, and he'll tell you rushing the ball and stopping the run are the two most important tasks in football, bar none. Let's see if that Bo knew football after all:

I broke down every team's performance in games played over the last three college football regular seasons into one of four quartiles for both their yards per carry and their yards per pass. (I used yards per carry and yards per pass to control for the fact that team have different run-pass balances. Just looking at raw rushing and passing yardages would always put a team like Navy toward the top, even if they had a pretty poor running day and it took them 55 carries to get 180 yards.) For yards per carry, 2.71 yards per carry was a 25th percentile performance, 3.76 was the average (well, median if we're getting technical) and 4.89 was a 75th percentile performance, so those were the cut points that created our four quartiles. For yards per pass, 5.33, 6.71 and 8.30 were the 25th, 50th and 75th percentiles.

Disclaimers: Note that the quartiles aren't exactly 25 percent apiece but 24.5 or 25.2 percent, because I'm only crazy enough to take the cut points to the hundredths place. Note also that the wins and losses sum not to a .500 record but about a .520 mark because, indeed, FBS (Division I) teams have collectively won about 52 percent of their games the last three years. How is this possible? No ties and, no, we haven't found the ever-elusive moral victory, but FBS teams play FCS (Division I-AA and down) teams on occasion, and unless they're Michigan, they win. But enough of the foreplay.

Top quartile: YPC: 75%, YPP: 81%
Third quartile: YPC: 59%, YPP: 59%
Second quartile: YPC: 47% YPP: 40%
Bottom quartile: YPC: 24% YPP: 25%
(Percentages are win percentages for teams in the given performance quartile for the given stat. Each group has about 1,100 observations. YPC = yards per carry, YPP = yards per pass. Note that these can be interpreted identically as offensive winning percentages or defensive losing percentages, because for every time a team had a 3.5 YPC game and lost, some other team allowed that same 3.5 YPC game and won.) If you squint hard enough at the numbers, you can deduce some differences between your team's chance of winning depending on how well it ran versus how well it passed, but then again, people can see the Virgin Mary in their French toast. The bottom line is that gaining and allowing yards per carry and yards per pass appear all equally important – but relative to other teams' performances, not relative to absolute yardage. That is, a top-25 percent day against the run or pass will result in approximately the same 70 to 80 percent winning percentage. Of course, however, there's no top-25 percent yard line on the football field. Because the yards per pass quartile cutoffs (2.71, 3.76 and 4.89) are slightly further apart than the yards per run's (5.33, 6.71 and 8.30), we'd expect each additional yard per run to move a team further up the rankings, and thus have more of an effect on a team's win percentage than another yard per pass. Let's see: 4-5 YPP: 23%
5-6 YPP: 34%
6-7 YPP: 46%
7-8 YPP: 57%
8-9 YPP: 72%

1-2 YPC: 21%
2-3 YPC: 33%
3-4 YPC: 50%
4-5 YPC: 61%
5-6 YPC: 72%

Well, guess our soufflé did just crumble apart after all, as the win percentages are about the same. Then again, this is a really interesting finding in its own right. Bo doesn't know football after all. Let me explain.
Conventional wisdom would have dictated that being able to pound the ball on the ground would be more important than throwing for another few yards per pass. For one thing, teams run more than they pass at the college level. For another, there's less variability in rushing yardage so it's a more effective offense (though with lower per-play returns, as you can see), which is the whole rationale behind three yards and a cloud of dust. Plus, teams turn to the run in goal-line and short-distance situations when that extra yard is most important. We think of running backs churning for those precious last nine inches and receivers stepping out of bounds a yard or two before a would-be hit. However, this data suggests our receivers need to take a lesson from our tailbacks and our DBs need to stick their tackles like our linebackers – that extra yard per pass is every bit as influential on Stanford's chances of winning.

Don't get me wrong, there's plenty of evidence out there that suggests rushing prowess is more important than passing ability in football. Heck, I ran another test (called a logistic test) to double-check these results, and found that each additional yard per carry is associated with a 1.7-fold increase in the probability of winning, compared to a 1.5-fold increase for each additional yard per pass. That doesn't sound like a major difference, but it is statistically significant with our 4,000-observation sample. Intuitively, the last dozen national champions all had top-ten rush defenses nationally, and good but not always great pass Ds.

Then again, I dare anyone who's watched all the Stanford fourth-quarter collapses of the last five years and tell me a pass defense isn't every bit as important as a rush D. (Plus, of course, the difference between 1.5 and 1.7 isn't all that great.) Most convincingly, remember Florida, winner of the last two national titles and running? Who's their running back? They don't have one, really. When not busy winning Heismans, spreading God's word in prisons or performing circumcisions in the Philippines, their quarterback doubles as the running back. Imagine winning straight two national titles with a team that only passes when it absolutely has to, and then only when its tailback drops back to throw. There is no way that could happen today.

In conclusion then, I think we disproved conventional wisdom again today. Unlike turnovers, where interceptions smart twice as badly as fumbles, a yard is a yard is a yard, whether earned or defended on ground or in air. Paul Revere ("one if by land, two if by slant route") would be disappointed.

What does it all mean for Stanford? I think this is bad news in the short term, because Stanford was better on the ground both offensively and defensively last year, and I don't expect either to change this season. Expanding the horizon to the foreseeable future, the Card probably will be better defending the pass than run, given questions at the LB position and the stockpile of talent at DB (and, of course, rush end). Plus, with an emerging star at QB, stud receiver recruits (and RB recruits, to be fair), and the most questions on the offense's future coming on the OL, I think Stanford should be return offensive to its days of yore, where it lit up scoreboards through the air. Good news for us CardinalManiacs: the purists can cry, but Luck to Patterson and Jones should be every bit as effective as three yards and a cloud of dust.


Are you fully subscribed to The Bootleg? If not, then you are missing out on all the top Cardinal coverage we provide daily on our award-winning website. Sign up today for the biggest and best in Stanford sports coverage with TheBootleg.com (sign-up)!


The Bootleg Top Stories