indicators that do the best job of separating winners from losers on the field

Winners and losers: beauty runs stat deep

Cold, Hard Football Facts for Feb 16, 2012

By Kerry J. Byrne

http://www.coldhardfootballfacts.com/content/winners-and-losers-beauty-runs-stat-deep/13555/

The Cold, Hard Football Facts treat football statistics much like we treat swimsuit models or cheerleaders. All are beautiful … but some are just more beautiful than others.

In the case of swimsuit models and cheerleaders, beauty is a function of pheromones, chemistry and personal taste. When it comes to our Quality Stats, beauty is much easier to define: we’re hopelessly devoted to the indicators that do the best job of separating winners from losers on the field. You should be, too.

Conversely, we kick to the curb those indicators that don’t matter in the win column, as if they were an aging old slutty ex-girlfriend looking to re-kindle the lack of romance we once showed them back in our pre-Troll days. Call them losers in the win column of life.

The Cold, Hard Football Facts are quite simple at the end of the day: we admire most those stats that win and lose NFL games. We dismiss those stats that don’t. It’s not rocket science. But it is enlightening.

Top of the list? Teams better in CHFF’s Real Quarterback Rating won a remarkable 86.9 percent of all NFL games in 2011. Way at the bottom? Teams better in Rushing Yards Per Attempt won just 47.6 percent of NFL games.

At their best, these indicators either refute or confirm conventional wisdom about what’s important in pro football and what’s not.

What we find is that the swirling chaos of pro football usually follows very distinct statistical patterns: Week after week, year after year, games are won by teams who excel in certain skills on the field.

Here are the stats that won and lost games in 2011, ranked in order from best to worst. We call it Correlation to Victory and we tracked it every single week at CHFF Insider.

It’s not “correlation” in the scientific way that statisticians use the term. It’s just the record of teams when they win a certain statistical battle. At the end of the day, that’s all that really matters.

We’ll follow up next week with a look at Predictive Rater of Victory, another stat we track week by week at CHFF Insider, which tells us how often each indicator predicted winners.

Real Quarterback Rating
Correlation to Victory (reg. season): 223-33 (.871)
Correlation to Victory (postseason): 9-2 (.818)
Total Correlation: 232-35 (.869)

Real Quarterback Rating is our very successful effort to quantify all aspects of quarterback play: passing, running, rushing TD, sacks, fumbles, etc. It essentially uses total QB production and the tried and true passer rating formula to spit out an overall rating of play by quarterbacks.

Some morons criticized this new indicator during the 2011 season. But what’s to criticize? It turned out to be the best stat in football.

It also confirms for us that winning in the NFL is all about superior play at QB. No stat in football was better at separating winners from losers.

Granted, there are MANY variables that go into superior play at quarterback. This is NOT an individual indicator. It is a team indicator. Teams with better pass protection, and better receivers, and better defenses that make like miserable (or as close to miserable as possible in the modern QB-friendly game) are more likely to win the battle of Real QB Rating.

At the end of the day, it tells us football is a very simple game: nearly 9 out of 10 NFL games are won by the team with the more efficient play at quarterback, regardless of what else happens on the field.

Scoreability-Bendability
Correlation to Victory (reg. season): 216-40 (.844)
Correlation to Victory (postseason): 11-0 (1.000)
Total Correlation: 227-40 (.850)

Scoreability and Bendability are our measures of offensive and defensive efficiency. Namely, it tells you who played the best in so-called “situational football.” And it speaks loud and clear: it doesn’t matter how many yards you generate in a game. It matters how smartly and efficiently you play the game.

You don’t need to win the yardage battle. But if you play better in the red zone, if you win the turnover battle, play better on special teams, don’t commit dumb penalties, etc, you’re going to win about 85 percent of all NFL games. Football fans know instinctively that this is true. But we quantified it.

Situational football apparently grows more important as we get later in the season. Teams who played better situational football went an incredible 124-16 (.886) over the second half of the season (Weeks 9-17) and a perfect 11-0 in the postseason.

Scoreability-Bendability was the only indicator that was perfect at identifying winners in the playoffs.

Turns out, situational football is the most critical measure of success in playoff football. Scoreability-Bendabiltiy was:

8-3 in the 2010 postseason
11-0 in the 2009 postseason
10-1 in the 2008 postseason
9-2 in the 2007 postseason
6-5 in the 2006 postseason
10-1 in the 2005 postseason
10-1 in the 2004 postseason

That’s 75-13 (.852) over the last eight postseasons for those of you keeping score at home, with the 2006 playoffs the only outlier in the bunch.

From Week 9 through the Super Bowl, here in the 2011 season, teams that played better situational football went 135-16 (.894).

In other words, smart teams win games. And they win more often when the games get bigger and the competition gets tougher.

Passer Rating Differential
Correlation to Victory (reg. season): 201-55 (.785)
Correlation to Victory (postseason): 9-2 (.818)
Total Correlation: 210-57 (.787)

Traditional passer rating has plenty of critics. We dealt with those clowns throughout the 2011 season and long before that. Hell, even ex-coaches like to tell us that passer rating doesn’t matter.

How these people reached the highest levels of coaching without understanding that the NFL is all about passing efficiency says a lot about the stranglehold our arch-enemy conventional wisdom has on the game today.

Hell, Brian Billick is a Super Bowl champion coach! A Super Bowl-champion coach smart enough to follow us on Twitter (@football facts, by the way). And even he doesn’t understand the power of passer rating.

That power is evident here: teams that win the Passer Rating battle won nearly 80 percent of all NFL games in 2011. That’s pretty powerful. And it’s also pretty consistent with pro football history.

By the way, only two teams upset the statistical apple cart in the 2011 postseason. The Giants and Patriots each won their conference title games despite the fact that the opposing QBs that day were more efficient.

But as we all saw, both conference champions had a lot of breaks go their way: the Giants in the form of two special teams flubs by the 49ers and the Patriots in the form of a dropped TD pass and missed goal by the Ravens in the final seconds.

If not for those big breaks, the team with the higher passer rating would have been a perfext 11-0 in the 2011 playoffs.

Interceptions
Correlation to Victory (reg. season): 151-42 (.782)
Correlation to Victory (postseason): 6-2 (.750)
Total Correlation: 157-44 (.781)

We’ve long said that interceptions are the most important single play in sports. The way they correlate to victory pretty much proves it. We call it the CHFF Interception Ladder. We need to update it. But short version: each INT decreases your chances of winning by about 20 percentage points.

This is the only indicator we track that looks only at the impact of a single play (or several of them in a game) and its Correlation to Victory is phenomenal.

Put another way: you throw INTs, you lose football games. Simple as that. You throw more interceptions than the other team, you lose nearly 80 percent of all NFL games.

We did not look at fumbles, because it seems that fumbles are random acts with a low Correlation to Victory. Interceptions are function of the quality of QB play, defense, scheme, coaching, etc. However, we should track fumbles in the future just to prove they’re random (or maybe refute our own expectations).

Real Passing Yards Per Attempt
Correlation to Victory (reg. season): 186-70 (.727)
Correlation to Victory (postseason): 7-4 (.636)
Total Correlation: 193-74 (.723)

We often tell people you can tell who won a football game simply by looking at a team’s average per pass attempt. Well, it’s not that cut and dry. But the fact of the matter is that the team with the higher average per attempt wins about 3 of 4 NFL games.

That’s pretty compelling for a stat that’s so simple. In fact, we’ve long said that it’s a disservice for analysts, writers and even the NFL broadcast networks to show a quarterback’s total yards, but not his average per pass attempt.

The total passing yards are meaningless. The average per attempt goes a long way toward telling us who’s going to win the game.

By the way, 2011 was a down year for Real Passing YPA. In most seasons since 2004, teams that won the YPA battle won more than 75 percent of games. In 2009, it was nearly 80 percent: 204-52 (.797).

Passing Yards Per attempt
Correlation to Victory (reg. season): 182-74 (.711)
Correlation to Victory (postseason): 8-3 (.727)
Total Correlation: 190-77 (.712)

Real Passing Yards Per Attempt includes the impact of sacks. In other words, it looks at net passing yards and counts sacks as attempts. It’s more of a team indicator, which is why we use Real Passing YPA as one of our official Quality Stats.

Plain old Passing YPA here is merely a function of passing yards divided by passing attempts – more an individual stat for quarterbacks.

Even then, it’s a fairly telling indicator of success – but slightly less telling than Real Passing YPA. In other words, the impact of these stats follow the pattern that you might expect. Passing YPA is important, but it’s even more important when we start to measure the impact of sacks.

Negative Pass Play Percentage
Correlation to Victory (reg. season): 177-74 (.705)
Correlation to Victory (postseason): 8-2 (.800)
Total Correlation: 185-76 (.709)

Negative Pass Play Percentage is one of the indicators we track as part of our Offensive and Defensive Hog Indices. It measures the percentage of snaps that end in a sack or INT, both on offense and on defense.

We broke it out separately in our Correlation to Victory chart this year and it’s pretty interesting. Teams that suffer a lower percentage of Negative Pass Plays won more than 70 percent of games in 2011.

Interestingly, interceptions alone (see above) correlate much more highly to wins and losses. The difference, of course, is that we look at total interceptions only above. Here in Negative Pass Plays, we’re looking at INTs and sacks as a percentage of plays.

It seems that in terms of Negative Pass Plays, total volume is more important than percentages and efficiency.

By the way, Super Bowl XLVI was the one 2011 postseason game in which the two teams tied in Negative Pass Play Percentage. The Giants and Patriots each suffered three Negative Pass Plays in 43 dropbacks (40 attempts, 3 sacks for Manning; 41 attempts, 1 INT, 2 sacks for Brady).

Third-down Success
Correlation to Victory (reg. season): 178-71 (.715)
Correlation to Victory (postseason): 4-6 (.400)
Total Correlation: 182-77 (.703)

One of the great clichés in football is that teams need to convert their third-down opportunities on offense and get off the field on third downs on defense.

Hey, for once conventional wisdom is right!

Teams that were better on third down won more than 70 percent of all games in 2011. Interestingly, it did not correlate to victory too often in the playoffs.

But that 2011 postseason number was a historic outlier. Teams that won the third-down battle went 9-2 last year in the playoffs; 7-4 in 2009; 10-1 in 2008, etc., etc.

Rush Yards
Correlation to Victory (reg. season): 174-81 (.682)
Correlation to Victory (postseason): 6-5 (.545)
Total Correlation: 180-86 (.674)

Coach Bill Parcells once said that it’s not how well you run the ball, but how often. He’s dead-on-balls accurate. Running often is important for two reasons.

One, it keeps defenses honest in a sport in which games are won and lost by the passing game. And, two, it’s a sign that you are winning when you are able to run more often.

As a result, teams that run for more yards tend to wind close to 70 percent of NFL games. But as you’ll see below, unlike passing effectiveness, rushing effectiveness is not a big deal. It’s just the commitment to run that makes teams successful.

But when are teams running to win and when are they running because they’re winning? That’s more difficult to pin down.

Somewhere out there is the mother stat that combines rushing attempts with passing effectiveness – somewhere in that nexus is where you’ll find the secret to winning games: running a lot and passing very effectively.

The 1960s Packers might be the best classic example of a team that ran often, with hugely varying degrees of success, but that consistently won the battles of passing efficiency.

Passing Yards
Correlation to Victory (reg. season): 129-127 (.504)
Correlation to Victory (postseason): 7-4 (.636)
Total Correlation: 136-131 (.509)

We’ve been harping on this issue for years. Winning in the NFL is all about passing efficiency, and little about passing volume. And the success, or lack thereof, of teams that pass for more yards provides all the confirmation you need.

Fans of imaginary fake football may get all excited by the big 400-yard passing day. But when all is said and done, a big passing day is largely irrelevant. In fact, in many cases, it’s a sign that you’re probably losing and need to pass for more yards to try to catch up.

But as the Cold, Hard Football Facts prove, that passing effort does not always lead to victory. Teams with more passing yards barely won more than 50 percent of games in 2011. It’s a coin flip.

Rushing Yards Per Attempt
Correlation to Victory (reg. season): 122-134 (.477)
Correlation to Victory (postseason): 5-6 (.455)
Total Correlation: 127-140 (.476)

Teams can run to daylight all day long. They can dominate in the trenches defensively and make life hell on opposing running backs. But at the end of the day, that effort is so distantly related to victory that they can marry in 47 states.

You might call it Minnesota Vikings syndrome.

The Vikings habitually run the ball well and stop the run. Been that way for years. The 2011 season was no exception: 5.2 YPA on offense (second); 3.9 YPA on defense (sixth). How’d that work out for them? Not well. The Vikings embarrassed themselves this year witha 3-13 record despite a great running back, a dominant defensive lineman and high expectations.

Sorry, “establish the run” crowd. But the impact of an effective rush offense and rush defense are massively overrated.