Time of Possession: A comparison of 2011 to 2010

Submitted by Moleskyn on

We all know that Hoke has tried to emphasize time of possession (TOP) this season, as evidenced by his comments along the lines of "the defense plays best when they're on the sideline." So, I thought I'd take a look at the TOP numbers from this year and compare them to last year, to see if Hoke has achieved what he wanted to.

Before I get into it though, a few clarifications:

  • This isn't meant to spark a debate about whether or not you think TOP is an important statistic. While that is an interesting debate, the level of importance of TOP as a stat is irrelevant right now, since I'm just trying to see how well Hoke has achieved his goal.
  • This isn't meant to be a back-handed jab against RR, either. I am only comparing this year's TOP numbers to last year's to see the progression from one year to the next. Again, Hoke has made it a point of emphasis to try to control the ball on offense, and I'm just trying to see if he's achieved that goal.
  • I am certainly not a math wiz, so I may be off by a few seconds in some of my calculations. I converted the "minutes:seconds" format listed in the box scores into seconds, so that I could more easily figure out totals and averages, and then converted the seconds back into minutes:seconds format to make the numbers more readable; things may have gotten jumbled in the process. If you would like to see a table with all of the numbers, I can provide that.
  • My sources were ESPN's boxscores for this year, and the boxscores on MGoBlue.com for last year.

Without further ado, here's the overview:

2010

In 13 games (I included the Gator Bowl and UMass), there were a total of 780 minutes played. Of those 780 minutes, Michigan's offense possessed the ball for 353 minutes and 14 seconds, or roughly 45% of the time. The most our offense possessed the ball in a single game was 36:52, in our opener against UConn. The least we possessed the ball in a single game was 18:13, in our win over Indiana.

As an aside, our second-lowest single game TOP came against UMass at 22:22, which is why I didn't throw out that game. Games against FCS teams typically get thrown out because they would schew things in our favor; this actually does the opposite.

Our average TOP per game was 27 minutes and 20 seconds.

2011

In 12 games played so far, there have been a total of 703 minutes and 33 seconds played (keep in mind the WMU game didn't go the full 60 minutes. They only played for 43 minutes and 33 seconds). Of that time, Michigan's offense has possessed the ball for 384 minutes and 47 seconds, or roughly 55% of the time. The most our offense possessed the ball in a single game was 41:13, against Nebraska. The least we possessed the ball in a single game (not counting WMU) was 22:59, against Notre Dame. Our average TOP per game, not including WMU, is 33:35. If you include the WMU game, our average per game comes out to 31:10.

What does this mean?

Well, basically that Brady Hoke came out and accomplished one of the things he wanted to accomplish. Our offense increased their total TOP by 10%, which can't be a bad thing. Other than that, I don't think it really means much at all. Just another reason for Hoke and his staff to slap each other on the back and say "Good job."

What say the rest of ye?

Mr. Rager

December 28th, 2011 at 2:34 PM ^

Wow, everything pre-"What does this mean?" could have been written in two sentences.  You must be bored.  Thanks for sharing.

sheepdog

December 28th, 2011 at 2:56 PM ^

Its actually a pretty interesting stat and pretty important in our defensive turn around from a year ago.

In my opinion, the defensive turnaround was owed mostly to these three reasons: good coaching, good tackling, and TOP.

MGoShoe

December 28th, 2011 at 2:36 PM ^

...devotees get the TOP issue wrong is not because of what the number tells you about an offense, it's because of what it tells you about a defense.

Better defense means fewer, shorter opponent possessions -->> more own team possessions and more opportunity to score.

m1jjb00

December 28th, 2011 at 4:34 PM ^

How can it be only about just the offense or just the defense? In fact it also is sabot special teams. Imagine what a team's top would be if it gave up a td every time it kicked off or punted. That said it's more about the style of offense a team plays than anything else because that's under control of the team. Shave 5 seconds off each play from a 30/45 second clock and you have a substantial reduction in top

PurpleStuff

December 28th, 2011 at 5:36 PM ^

Good/better defense (like you said, quicker stops and more turnovers) can result in more TOP for your offense, but more TOP doesn't equal good defense.  Teams possess the ball all day against Oregon but they aren't generally doing much to stop the Ducks' offense in the process. 

Also, every team is trying to force three and outs and turnovers.  No defense intentionally gives up long drives.  On the flip side, teams can/do differ in how they want to control the tempo on offense.  They can be more oriented toward the run or the pass and can attempt to pick up more big plays or move methodically down the field.  At the end of the day what matters is the end result of those drives, not how long they take.  That is why TOP in and of itself means essentially nothing and just reflects a combination of offensive style of play with a slice of defensive success.

 

 

DonAZ

December 28th, 2011 at 2:41 PM ^

I like TOP in general, but I don't think it's very predictive of victory.  Too many other things factor in -- turnovers, red zone success, scoring efficiency of the opponent when they have the ball.

Somewhere the must be a statistic that factors in what a team does with those minutes of possession ... points/minute of posession?  Net points/minute -- offensive possession efficiency vs. defensive points given up per minute? 

DonAZ

December 28th, 2011 at 4:03 PM ^

Do you have evidence to support?  I admit that on the surface that would seem to be true ... but for teams like Oregon (dead last in TOP for 2011) it doesn't matter.  I've seen articles suggesting TOP was only weakly correlated to victory (with turnovers being an example of a major statistic more strongly correlated).

Tuebor

December 28th, 2011 at 2:49 PM ^

I think in this league you have to be a physical running based offense.  This TOP improvement shows how much we have improved.  Thanks for running this analysis.

Spontaneous Co…

December 28th, 2011 at 2:56 PM ^

I believe your attempts to minimize the importance of your analysis reflect your efforts to avoid RR bashing which magically appears in most of these types of "comparisons."  Regardless of the cause, it is a very good thing to possess the ball more.  I agree with the commentor that says other factors are critical to determining victory besides TOP, but TOP is a valid way to judge how much your offense is controlling the game, how much your defense is controlling the game or some combination of those two things.  Whichever aspect of "control" it demonstrates, we should all be thrilled to see that stat improve. 

The only way I would discount TOP is if a team dominated that stat but somehow managed to only score a dozen points a game or gave up 40 a game.  But even then that doesn't mean TOP is not important, it means your team is failing miserably in other aspects, i.e. turnovers, special teams, red zone efficiency, giving up huge plays, etc.

go16blue

December 28th, 2011 at 3:23 PM ^

Outside of a few outliers (see: Oregon), TOP is a pretty good measure of overall team effectiveness. There are better metrics, but TOP shows when your offense is doing well (you therefore have the ball more) and when your defense is doing well (your opponent has the ball less).

It is not is a stat to aim for, or to build your team around. There's no sense in trying to grind it out just so you can win TOP unless your team is outmatched; in that sense TOP is a useless statistic. That having been said, its certain evidence of our improvement this year.

DonAZ

December 28th, 2011 at 4:12 PM ^

The benefit of TOP can be mitigated by (a) turnovers, and (b) lack of success in the redzone.

Burning up 7 minutes of clock per possession and scoring 0 or 3 points is a good way to lose ballgames against teams with a more aggressive attack.

TOP is imporant.  It just isn't the most telling statistic.

I'd argue 3rd down conversion rate is more predictive than TOP.

Blue in Seattle

December 28th, 2011 at 3:24 PM ^

excerpt from Brian's post after the Nebraska game, "This is Michigan; This isn't Michigan"

and read all the way through the article to this blockquote, and read the article linked to it.

 

Media, conventional. My man Nick Baumgardner on the lopsided time of possession:

One of the residual effects of Michigan's stellar defensive day was a lopsided time of possession battle. The Wolverines held the ball for 41:13 while Nebraska had possession for just 18:47.

"Residual effects." My man.

Read the AnnArbor.com article at this link, "Michigan Special Teams play up to their name"

where you can read the full TOP section written by Nick Baumgardner,

 

Michigan dominates time of possession One of the residual effects of Michigan's stellar defensive day was a lopsided time of possession battle. The Wolverines held the ball for 41:13 while Nebraska had possession for just 18:47. "You play really good defense when you get to watch your offense," Hoke said. "Our offense, not in every game, but (it does) have a little advantage in time of possession. "One of our best defensive calls is (our offense) on the field. And on defense, we (limited them to) 3-for-13 in third down conversions and helped ourselves get off the field."

Emphasis is mine, but basically TOP is just the result that you get when your defense creates 3 and Outs, AND your offense completes the possession with 7 points.  For further education please use the MGoBlog search function to find many of Brian's articles gushing about his love of Hoke's Game Theory approach of "Controlled Aggression".  Oh Hell here's the "money quote" showing Brian's acceptance of the Hoke era.

 

Controlled aggression.

How would you characterize the first year of the Hoke era if given only two words? I don't think you could do better than sniping a couple Hoke used to describe Denard's game:

"Denard went out there as a quarterback of Michigan and went out there to help his teammates and be accountable to his teammates. He couldn't do it by himself and no one ever does, but I thought he played an aggressive, controlled football game."

Controlled aggression. From Mattison's okie blitzes that get an unblocked guy while dropping seven to Borges going for points in the fourth quarter Saturday to Hoke's decisions to go for it on fourth down to Hoke's ability to not strangle Hagerup (better man than all of us), "controlled aggression" is the story of Michigan's 2011… and its future.

I could not have been more wrong about Hoke. He's not the milquetoast win-by-not-losing sort. He's not even average. He has a gut feel that is on par with every RPG minimaxing engineer out there. Forged by the fires of MAC defenses, Hoke has learned to push when he should and pull back when he should. I would not want to play poker against him.

I know Hoke talks about toughness and physicalness even if the latter isn't really a word, and that's fine and important. It's half of the equation. The other half is putting your guys in position to take advantage of that. Hoke does that. MANBALL: pretty much not pejorative anymore.

Please take this as an educational response to your post, which is weak.  Afterall I could have just replied, "TOP, man you just don't 'Get It' "

Hoke = Lloyd only for those who only take the time to look at the surface of this Michigan team.  If Lloyd had coached that Ohio game he would have lost.  Hoke did not, because Hoke's MANBALL = Controlled Aggression, the result will likely be TOP, but more importantly it means more POINTS for MICHIGAN and less for the opponent.

 

Nardudeski

December 28th, 2011 at 4:05 PM ^

Does T.O.P. even matter? After all of the time we've put into caring about it on here, have we ever looked back and checked to see if you can even correlate (or better, argue causation of wins due to T.O.P.) T.O.P. to wins? 

Top 10 Longest T.O.P's from 2011: Temple, Standford, Army, Kansas State, Georgia, Virginia Tech, Central Florida, Texas, Nevada, Western Kentucky.

Top 10 Shortest T.O.P's from 2011: Auburn, Western Michigan, Indiana, Tulsa, Oklahoma State, Miss. State, Houston, Troy, Middle Tennesee, Oregon, Maryland. 

Conclusion: MOAR T.O.P.? 

Blue boy johnson

December 28th, 2011 at 5:13 PM ^

I will add a couple more stats that may be relevant to what Hoke is trying to accomplish

This years D faced virtually the same amount of plays per game as the  2006 team. Don't know what it all means but it is interesting.

TOTAL PLAYS BY OPPOSITION

2011:  727/12  60.5/game

2010:  963/13   74/game

2009   838/12    69.8/game

2008   836/12    69.7/game

2007   898/13    69/game

2006   789/13    60.7/game

micheal honcho

December 28th, 2011 at 5:15 PM ^

I think first downs is a more telling and important statistic than TOP, although the 2 are most likely correlated to some degree.

Anyone who has played the game of football, even at a HS level either on offense, defense or both understands that a 1st down is essentially a small victory within the game. It demoralizes the defense and empowers the offense. When you string them together in a long drive you are reinforcing your dominance over a team, this projects to subsequent possesions.

If an offense comes out and scores against you on 3 plays, it has an effect on the defensive personel, but its not the same as a long drive full of 1st downs. A quick score is similar to a kick return or punt return for touchdown. A momentary set back that can be quickly shrugged off as we re-focus on stopping them.

I know this is just emo bullshit to the stat heads who think the game is won and lost on paper, however, anyone who has actually spent time playing organized football at the varisty level or better knows of which I speak.

Does this automatically ensure victory? No, nothing does. But it does build a teams confidence in themselves, their scheme and their personel that usually leads to consistency and more victories than not.

DonAZ

December 28th, 2011 at 6:01 PM ^

I agree with that.

Further, I'd flip that and say one of the more demoralizing things for an offense is to march down the field in a time-burning possession and fail in the red zone ... either due to turnover (worst case scenario) or a missed third down conversion leading to a field goal (concession points).

Or, put another way ... a great boost for a defense is to cause a ball possession offense to fail in their drive.

ppToilet

December 28th, 2011 at 7:36 PM ^

of scoring more points than the opponent.

You don't win the game by holding the ball longer as that is just a means to an end.  If you can hold the ball longer and still score points, then that is better than just scoring points.  Keeping your defense off the field is a good thing, unless your defense is so dominant that they will cause turnovers and/or score points themselves.

Victory is defined by scoring more points.  Any other statistic is only useful inasmuch as it correlates to (or influences) scoring more points than the opponent.  There is no single statistic or variable that will do this for a variety of reasons including that the variables are not independent of each other.  The best one can do is rate the relative predictive value of one variable relative to another (I'm not entirely familiar with what Mathlete tracks in the database, but the model created seems to do this on a play-by-play basis).

Getting back to the original post, I think the question was framed correctly as 'did Hoke meet his objective?'. The alternative question is 'does T.O.P. score more points?' and that is pointless taken on its own.

treetown

December 28th, 2011 at 9:30 PM ^

There is no single parameter which predicts victory outside of the game score. As others noted, the time of position reflects not just defense but the interaction between defense and offense. Each minute is worth roughly two plays, so a four minute difference is a plus/minus 8 plays - or at least one good drive. What is that worth?

A similar effect is seen in basketball. A number of years ago during the "Bad Boys" era, the Detroit Pistons was one of the top defenses in the NBA but many insightful observers noted a key factor was their deliberate play on offense. Simply using most of the time during each possession limited the opposition's chances. At the risk of stirring up the whole RR good/bad right/wrong issue, it may be that the TOP shows how well the offense and defense are in sync with each philosophically. A defensive minded coach which doesn't mind winning 14-3 or 21-17 wants a defense that is hard to score on. So, ideally the offense should hold on the ball for long drives and rarely turn the ball over. See the New York Giants under Parcells during the Super Bowl run. Their offense was not spectacular but their defense was stifling and in the Super Bowl against the Buffalo Bills, the Giants held the ball for 40 minutes; keeping the high powered Bills attack under Jim Kelly off the field for much of the game.

Hoke's offense under Borges isn't as spectacular but is a good fit with his defense - They are off the field in general quickly and don't give up that many big plays or scores (not perfect but better).

On the other hand if one wants to play a hyperactive offense like the Oregon Ducks, coaches should look not just at their offense but their defense. They take chances and try to create the equivalent of an offensive big play - sending players on run blitzes or slants in hope of "blowing up" a play and creating a really bad down and distance. A riskier defense which might create turnovers, blow up the other teams plays and keep up that hyperactive intense atmosphere is a good fit with that offense; it might give up a big play but if the offense is scoring it is good gamble.