Saturday, January 1, 2011

Kickoff or Receive?

by Ed Anthony

A little known fact is that the team losing the initial cointoss has the option of electing to receive the ball at the beginning of the second half to electing which side to defend. THis means that if the home team loses the cointoss and begin the game with a kickoff they can then elect to kickoff at the start of the second half.

Conventional wisdom is that there is an advantage to have first possession in a half. For this reason teams always elect to receive the ball at least once during the game. But the rules are clear that the game need not develop this way.

Over the years there has been much discussion whether a team should has an advantage receiving in the first half. An argument can be made that receiving to open the second half gives a team the advantage of knowing whether they are in the lead and can play accordingly. arguments have also been made that first possession puts pressure on the kicking team to "catch up."

For year's I've felt that the greater advantage is to receive the ball in the second half. To provide my point I recently ran data on five season of football and I'd expected to find that there was at least a strong trend that teams receiving the ball in the second half had an advantage.

Below is the data I'd found for the 2005 - 2009 seasons. The results were not what I'd expected. As a matter of fact I am certain the no football fan is aware of the trend I've identified. The table shows data representing the outcome of each game by half. The team receiving the ball has consistently lost the half (except in 2008.) This result is surprising in itself. The receiving team will have as many possessions if not one more than the kicking team. Yet the data clearly show that the kicking team has won more game halves than the receiving team. (Ties were counted at the end of regulation.)














SeasonWonLostTied
2005 215 274 43
2006 223 267 42
2007 238 249 45
2008 230 193 109
2009 227 260 45



I next ran the data to see how teams receiving the ball at the start of the second half succeeded. The datas show the game results of the team receiving the ball in the second half. Again the team kicking to open the second half won more games than the receiving team.
















SeasonWonLostTied
2005 136 119 11
2006 100 148 18
2007 119 139 8
2008 106 152 8
2009 98 155 13


As none of this applies to overtime, the only conclusion one can arrive at is that given the choice, every team should elect to kick rather than receive. And if given the opportunity to kickoff at the start of each half a team should take that option. This finding does not agree with convention wisdom.

13 comments:

Tarr said...

That is a surprising result indeed!

It would be interesting to see total possessions per team per half, excluding kneel-downs.

James said...

That doesn't make any sense. Weird.

Jim Glass said...

Very interesting. Tough to figure.

Ed Anthony said...

James
I agree that it makes no sense. My first thought when I got these results was that I had a problem with the code I'd written to parse the data. I checked it and my code appears to count correctly.

I've been giving this some thought since I posted this and is a possibility that the teams receiving the ball in the second half are behind. And if both teams played to form then the receiving team would go on to lose the game. Another way of saying this is that the team receiving the ball in the first half built enough of a lead to win the game.

There are several other ways of examining the dat and perhaps one day I will take the time to do it.

Buzz said...

Ed,

Very interesting. I have long thought the same as you and was going to do a similar study during the offseason. Out of curiousity what is the simple winning % of the entire game for teams receiving the opening kickoff to the game by year?

Anonymous said...

Can you post the raw data (or where you got it)?

parinella said...

My theory to explain this is that it's all due to the opening drive. Either a) there's some sort of psych or vibe ramping up the defense or chilling out the offense or b) offensive play-calling becomes sub-optimally conservative, so the initial possession of the half is less efficient than other possessions. Maybe it's also possible that c) home teams or better teams are more likely to choose to start with the ball at the beginning of the game because they aren't afraid to turn it over.

Buzz said...

Did a quick calculation from 2002-2008. (I don't have my pbp files updated for 2009 and 2010 yet). Over this time frame the team kicking off to start the game won at a clip of 51.1%. I also simulated a coinflip for every game during this time frame to see how often it would end up heads at a 51.1% clip over such a large sample and 13% of the time it ends up at that rate or higher. So there is a 13% chance that this rate is due to pure chance. However, it does give more credence to the idea that you should kick off to start the game.

Tarr said...

"parinella", this is a site for people who take statistics seriously, not dilettantes like yourself.

parinella said...

My old friend Mr. Tarr (or is that DokTarr), I'll have you know I did a chi-square test just the other day.

Anonymous said...

little bit suprising, kind of seems to mesh with http://www.advancednflstats.com/2010/01/expected-points-ep-and-expected-points.html

exptected value of start a drive at the 20 or so isn't that great

guess the defenses are better in real life than they are on Madden

Ryan Bolland said...

I think the first half of your analysis is great, however I question the usefulness of the 2nd half. Instead of counting teams winning or losing the game, it would be much more valuable to see which team won the 2nd half (the kicking team or the receiving team). As far as the benefit of choosing to kick or receive is concerned, we should not care at all about the score entering the half, only the amount of points scored after the half. Clearly, there would be a number of variables that could confound 2nd half data (point spread, injuries, momentum, etc.) more than they would confound 1st half data, but in an effort to try to isolate the value of kicking or receiving, you should only look at winning each half as an isolated incident and hope these other variables wash out.

To this end, would it be possible to post this 2nd half analysis?

Or... am I way off?

denis said...

Sorry,but I'm not seeing this.
I've checked my databases for 2005,2006,2009 and 2010 and receiving teams are winning more halves in each and every one of those years.Receiving teams are winning 268 to 224 in 2005....not losing by roughly that amount.Overall receiving teams are winning halves at a 52% clip.I'm also only seeing 39 tied halves in 2008 not 109.I'm prepared to be proved wrong,but..

Post a Comment