Gary Pinkel has always mentioned that playing defense is all about leverage. If two guys are pursuing a runner who is running toward the sideline, the job of the first guy isn't necessarily to tackle him, but to make him cut back toward the middle of the field to be tackled. (I'm horrifically paraphrasing here, I realize.) Whereas if you miss a tackle on the outside, the guy might run for 70 yards, if you turn him inside and there's a missed tackle, there are about 4 others guys in pursuit to make the tackle. It's not necessarily about making the big play yourself--it's about making it harder for the runner to make the big play. Or something like that.
Why am I mentioning this? Because my latest BTBS idea--Win Correlations (WinCorr)--further suggests that it's not necessarily how many big defensive plays you make that determines how well you do...it's more about leveraging the offense into uncomfortable situations (a.k.a. Passing Downs). I've covered this a bit in the past, but these numbers demonstrate that principle even further.
As I mentioned the other day in my Colorado preview, WinCorr is, in short, the correlation between a given statistical category and wins/losses. As you'll see, they can serve a couple different purposes: 1) we can use them to identify the most important of all the BTBS categories we've looked at to date, and 2) we can look at team-specific WinCorr's to develop a unique footprint for each team. We'll look at the former today and the latter tomorrow.
To illustrate the various uses of WinCorr, let's jump right in.
National WinCorr
There are two ways to look at WinCorr on a national level--determining which statistical categories are most tied to winning a specific game and determining which categories are most tied to winning seasons...i.e. being a good team. We'll look at both.
As I said above, we compare each statistical category with overall wins and losses, but...how do we come up with a number for wins and losses when we're talking about a single game? We have two options: either we 1) give wins a 1 and losses a 0 and run correlations off of that (the black and white way), or we 2) compare the stats from each game with the % of points a team scored in that game (the gray area way). (So if a team wins 20-10, instead of giving the winning team a 1, we'd give them a 0.667, as they scored 66.7% of the game's points.)
The former is cleaner (and leads to lower correlations, obviously), but the latter is probably a bit more telling. It determines a difference between winning 24-23 and winning 41-3. Plus, the correlations are simply stronger using % of pts, so that's what I'm going with here.
Two other things to note: 1) I ran Spearman correlations for these numbers--if you're a nerd, you probably know what Spearmans are, and if you don't, you probably don't care; and 2) below is a list of the strongest correlations...meaning there is the possibility of a negative correlation on the list with positive correlations. I did this because we're looking at what most directly impacts a game, not what impacts it in a positive or negative way...if that makes any sense.
Also...for each of these, I'll get rid of the obvious ones--you don't need lots of stats to figure out that things like '% of pts' and 'total points' are going to be highly correlated to wins.
Okay, one more thing: I've tried to highlight the most important information in boldface, so if numbers make your eyes glaze over, skip right to the bolded parts.
WinCorr (Correlations using % of pts)
- PPP, close game (0.682)
- S&P, close game (0.678)
- PPP, overall (0.642)
- S&P, overall (0.634)
- Total EqPts (0.617)
- Total EqPts, Non-Passing Downs (0.597)
- Total Rushing EqPts (0.587)
- Passing S&P, close game (0.583)
- Total Rushing EqPts, Non-Passing Downs (0.582)
- Total Rushes, Q4 (0.579)
- Success Rate, close game (0.578)
- PPP, Non-Passing Downs (0.575)
- Passing S&P, overall (0.573)
- Passing PPP, close game (0.565)
- S&P, Non-Passing Downs (0.565)
- Passing PPP, overall (0.563)
- Success Rate, overall (0.540)
- Total Rushes on 1st Down (0.534) ???
- Total Rushing EqPts on 1st Down (0.529)
- Rushing PPP, close game (0.529)
- Rushing S&P, close game (0.523)
- Total EqPts on 1st Down (0.521)
- Total Line Yards, Non-Passing Downs (0.517)
- Total Passes, Q4 (-0.516)
- Total Rushes (0.510)
I listed 25 because the correlations for all were over 0.500. Pretty strong correlations abound. And I realize your eyes probably glazed over looking at that list, but this tells a few really interesting stories.
- Success on Non-Passing Downs is crucial. Here's where the 'leverage' idea comes into play. Stopping a 1st-and-10 rush for 3 yards instead of 5 creates a much less comfortable situation for the offense. Little things like that could be seen as just as important as big hits and singular big plays over time.
- What strikes me as most interesting here is that PPP is worth a smidge more than S&P. The idea of S&P (Success Rates + Points Per Play) is to combine the efficiency of Success Rates and the explosiveness of PPP. However, PPP's correlation is 0.642 (in close games, 0.682), while the Success Rate correlation is only 0.540 (in close games, 0.578). Granted, that's only a 0.1 difference, but that's still a difference. And therefore it drags down the overall applicability of S&P. May have to think about retooling the idea of S&P.
- It's also interesting that PPP (the ability to make big plays) and S&P are more important to the passing game, while pure EqPts (the Pts you've racked up over the course of the game, not the average) is more important to the rushing game. Not totally sure what that means yet, but it's interesting. It's like the threat of a good passing game is as important as actually performing well in the passing game.
- One of my initial suspicions rings true--in the end, the 'close game' numbers are a bit more important than the 'overall' numbers. This makes sense--and it's why I initially created a 'close game' measure in the first place--but it's nice to get some affirmation on it.
- Obviously the presence of "Total Q4 Rushing Attempts" (as a decent positive correlation) and "Total Q4 Passing Attempts" (as a negative one) is a bit of 'correlation vs causation' here. It's not that you win more because you're rushing in Q4--it's that you rush more in Q4 because you're winning. This does, however, verify that bit of conventional wisdom.
Here's another long list--it's WinCorr over the course of a season. See if you can pick out trends.
WinCorr - season numbers (Correlation to '% of pts')
- EqPts Per Game (0.752)
- PPP (0.749)
- PPP, Non-Passing Games (0.748)
- S&P, Non-Passing Downs (0.745)
- S&P, close games (0.733)
- PPP, close games (0.725)
- Rushing PPP, Non-Passing Downs (0.722)
- S&P, 1st downs (0.711)
- Success Rate (0.708)
- PPP, 1st downs (0.706)
- Rushing PPP (0.702)
- Rushing PPP, close games (0.693)
- Passing S&P (0.692)
- Success Rate, close games (0.690)
- Rushing S&P (0.687)
- Rushing PPP, 1st downs (0.687)
- Rushing S&P, Non-Passing Downs (0.686)
- Rushing S&P, close games (0.682)
- S&P, 3rd downs (0.681)
- S&P, Q1 (0.675)
- Success Rate, Non-Passing Downs (0.675)
- Success Rate, 3rd Downs (0.674)
- Passing PPP (0.663)
- PPP, Q1 (0.654)
- Rushing S&P, 1st Downs (0.653)
- Of the 25 stats on the list, 8 were either related to Non-Passing Downs or 1st Downs. Leverage, leverage, leverage.
- EqPts and PPP continue to exceed S&P and Success Rates in importance. Ten of the 25 most important stats were variations of PPP...11 if you include EqPts Per Game.
- Interesting, though, is the fact that over the course of a season, overall PPP becomes more important than Close-Game PPP. Figure that one out.
- Nine rushing statistics on the list...2 passing statistics. Interesting. And these categories have nothing to do with pure quantity of rushes (which means the 'correlation vs causation' argument really doesn't come into play).
'+' Number Correlations
Since I spent all that time developing the '+' Number concept, you knew I was going to look at that too. What's funny, though, is that for this one, the correlations with Win % were significantly stronger than the correlations with % of Pts. I mean seriously, correlations in the 0.9 range? That's significant.
So that's what we're going to use.
Win Corr - + numbers (Correlation to Win %)
| Offense | Defense |
|
|
- I guess if you're a "Defense wins championships" kind of person, then here you go: Defensive EqPts+ is more tied to Win % than Offensive EqPts+. Granted, we're talking 0.919 vs 0.899, but...again, it's a difference!
- While EqPts+ are more tied to winning on the defensive side, S&P+ numbers are more tied to winning on the offensive side. So...pure numbers are more important defensively, while averages are more important offensively? Something like that? I dunno. Can't make much of that.
- Four more 'leverage stats' show up on the defensive side (S&P+, Non-Passing Downs; S&P+, 1st Downs; Rushing S&P+, Non-Passing Downs; Rushing S&P+, 1st Downs), however they're conspicuously absent from the offensive list. Granted, they're still important (garnering correlations of 0.663, 0.573, 0.557, and 0.482 respectively), but they're rather significantly less important. Can anyone explain that one? Creating leverage against the offense is better than...creating leverage against the defense? Eh?
So we can reach some pretty interesting conclusions from this data, and most of it comes back to the idea of leverage. I have data broken out for all quarters, all downs, the redzone, etc., and by far the most significant category is how teams perform in Non-Passing Downs.
Passing Downs = Turnaround?
This brings me to an interesting question: if Passing Downs = death, on average, then would the teams with the best numbers on Passing Downs be privy to a possible turnaround in luck the next year? In other words, are Passing Downs a lot like Turnovers? Is success in the category somewhat arbitrary...and does it even out over time? Being that we only have one year of play-by-play data so far, all we can do is take a look at the best (and worst) teams in the category, speculate, and see what happens at the end of the year. When I have multi-year data, it's going to be fun to tie all these season stats to success the next season, so we can see which stats are the best predictors of future success, but alas that is not an option just yet.
S&P+, Passing Downs (Offense)
1. Nebraska
2. Florida
3. Oregon
4. Texas Tech
5. Tulsa
6. Washington State
7. Kentucky
8. Hawaii
9. Louisville
10. West Virginia
Now...Florida, Texas Tech, Tulsa, Kentucky, Hawaii, and Louisville were six of the best passing teams in the country, so their presence on the list should surprise no one. Plus, Oregon and West Virginia had great all-around offenses as well. But...Nebraska? Washington State? Last year it seemed like NU was about 110% likely to run a screen to Marlon Lucky on third-and-long, and while it got yards (decent PPP), it certainly didn't move the chains very often (low success rate). Maybe they ate up enough yards each time to get them to #1, but...that just seems weird. I'm officially putting NU and Wazzu on my "watch" list. If they underachieve (not that either could do much worse than last year), this may be one reason why.
What about the bottom of the list?
120. Utah State
119. Northern Illinois
118. Idaho
117. Iowa
116. Temple
115. UCLA
114. Navy
113. Tulane
112. Akron
111. Notre Dame
Okay, none of these teams were known for their offense (or for having a good team at all), but there are a few major-league teams on the list that, if this measure means anything, could be in for a nice turnaround season--Iowa, UCLA, Notre Dame.
Summary
One thing I'm going to accomplish in the future is the entry of all the regular (i.e. non-BTBS) statistics as well. I would obviously like to see what the WinCorr's are in regard to things like 'yards per carry' (as opposed to my version, PPP). I have some tools at my disposal now that should make that pretty easy to compile, but we'll see when it actually happens. In all, though, I'm extremely satisfied with the WinCorr idea. Like I said, tomorrow we'll take a look at what a specific team's WinCorr says about them. In the meantime, let me know what you think!