SITE NEWS: We are moving all of our site and company news into a single blog for Sports-Reference.com. We'll tag all PFR content, so you can quickly and easily find the content you want.
Also, our existing PFR blog rss feed will be redirected to the new site's feed.
Pro-Football-Reference.com ยป Sports Reference
For more from Chase and Jason, check out their work at Football Perspective and The Big Lead.
ShareThisCheckdowns: How Much Does Time of Possession Matter?
After the Dolphins racked up a huge time of possession figure but lost to the Colts on Monday night, Brian Burke of Advanced NFL Stats started thinking about how much time of possession matters. Now, obviously, time of possession can come as a byproduct of some good things, like moving the ball and converting 3rd downs, but here he's talking about holding everything else (yards, efficiency, etc.) equal and just asking how much T.O.P. contributes to winning in and of itself.
This entry was posted on Wednesday, September 23rd, 2009 at 11:24 am and is filed under Checkdowns. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.
It's not time of possession, it's what a team does WITH their possessions. For instance, repeatedly settling for FGs after long drives and then giving up multiple TDs to the opponent in under a minute is *not* a winning formula. Right, Dolphins?
Seriously, that loss just CRUSHED me
I believe the conventional wisdom on time of possession is rooted in the widespread belief that defenses get fatigued more easily than offenses. The thinking is that if you can keep your defense off the field, they'll be fresh when the game is on the line late.
I don't believe that's true, nor have I seen any evidence that supports it, but it seems to be a view commonly held by announcers, players, coaches, etc.
I have seen games where the team that dominates TOP really is able to exert their will on the opponent in the second half, especially your smash mouth coaches, like Cowher and Schottneheimer. But, when you have an X-factor type player or an elite quarterback on the other side, it can still result in a loss. Manning would be an example. Another example that sticks out during Cowher's tenure with the Steelers is Eric Metcalf, who returned two punts for TDs for the Browns to beat the Steelers.
I'm not sure TOP really "matters" as much as it just being the result of playing really well and doing good things in a game.
Some time ago I was trying to get a list of games where a team had a 40-20 advantage or better, and I found 73 games from the 90's to this decade. The team that held the ball for over 40 minutes had a record of 66-7. The average margin of victory was 16.7.
No surprise but the teams to gain 350 yards with having the ball less than 20 minutes were the 1990 Bills in Super Bowl 25 against the Giants (371 yds), and Manning's Colts in 2000 against the Jets (375). The Colts got a little over 350 the other night in not even a full 15:00.
Why not run a linear regression and find out?
Because a linear regression won't tell you anything about the nature of the causation.
Miami's defense was fresh. There is no question about that. Their woes were caused by missed assignments and missed tackles, hence the 3 passing plays over 48 yards. I have never seen a team dominate a game so badly and still lose. I literally felt nauseous after that game.
Although the Colts were dog-tired (there is no way around it, they were on the field for 84 plays, and on their feet for 45:00), it didn't hurt them. They held strong in the red zone. You can't win by settling for FGs. The bottom line is the Colts made big plays when they had their chances, and Ted Ginn Jr. was a waste of a #9 draft pick (I'm still bitter). No but really, Indy made plays, and Miami didn't. Fasano missed a (difficult) TD pass in the end-zone. Ginn had alligator arms on a perfectly thrown ball in the back of the endzone (he thought he was deeper than he was), and Carpenter missed a FG. Make any of those plays and Miami wins.
Back to the main topic, obviously the T.O.P. possession numbers are highly correlated with winning. But, # of offensive plays seems more important when it comes to tiring out the other team. Throwing three incompletions then punting will take 30 seconds off the clock. Running 3 times then punting will likely take 2 minutes off the clock. They both took (roughly) the same effort to defend right? Is there a way to compare "Win/Loss" to "offensive play percentage?" (# of offensive plays/#of total plays)
Did that make any sense?
Miami ran 84 plays and Indy ran 34.
Miami ran 84/118 offensive plays (71%)
Ok, I'm leaving.
I'm with Roby. # of plays and # of possessions (over the course of a season) are important NFL statistics; TOP? Not so much.
One thing I used to be interested in (but never had the data to play around with) was stats like # of minutes with the lead, # of minutes tied, and # of minutes trailing. I always wondered if that had some useful predictive element to it.
Since T.O.P always seems to get all the attention, I never really heard any analysis of # of minutes leading/trailing. By its nature, the more time you spend in the lead, the more likely the clock is going to expire when you have that lead. (Think 16 games of hot potato. Grandma always seems to get stuck with said potato.) I think like scoring differential, it should be a good predictor of success.
I wonder what the ratio of run:pass plays are called when leading. Is there any kind of data showing that?
"I believe the conventional wisdom on time of possession is rooted in the widespread The belief that defenses get fatigued more easily than offenses ... I don't believe that's true, nor have I seen any evidence that supports it".
It's believable to me, and I suspect Kevin Gilbride came to appreciate the idea too after Buddy Ryan punched him out over it.
Of course it's not time of possession that wins games but points per possession -- as Manning's Colts have notably demontrated in recent years, especially during their Super Bowl year when they had such a horrid run D.
But there is the statistical argument (also advanced by Brian Burke a while back) that running the clock to reduce the number of possessions reduces the winning chances of the best offensive team. With a larger number of possessions, the best offensive team has a larger expected scoring gap over the inferior team. With fewer possessions the gap is smaller, and is more likely to be overcome by "lucky breaks" that may occur in the weaker team's favor.