Writers of Pro Football Prospectus 2008

19 Oct 2005

Week 6: Be Aggressive

This week Rich Eisen takes aim at the dreaded "Prevent Offense." He wonders why teams who jump out to early leads subscribe to the offensive philosophy that goes something like this: "don't make any mistakes, even if we go three-and-out on every possession of the second half." He also has some thoughts on the Pats, Joe Vitt, and notices that Seahawk receivers are actually ... gasp! ... catching the ball this season.

Posted by: P. Ryan Wilson on 19 Oct 2005

17 comments, Last at 20 Oct 2005, 1:41pm by Richie

Comments

1
by jim's apple pie (not verified) :: Wed, 10/19/2005 - 9:10pm

Of course, if a team passes trying to get the first down and fails, they're ripped by the media for not running to keep the clock moving. This is just a case of, if it works, then it was the right thing to do. If you get that first down at the end of the game, then whatever you did was the right thing, be it run or pass.

Also, I say always go for the two point conversion to win. Why not settle everything in one play, when your offense only has to gain a couple of yards? Are your odds actually better by playing for overtime, hoping to get the ball first or to stop the other team, hoping to get in field goal range, and then hoping your kicker doesn't blow it? Where is that odds guy when you need him?

2
by ABW (not verified) :: Wed, 10/19/2005 - 9:16pm

On Monday, an intrepid member of the New York media broached the subject with Coughlin, who replied with an incredulous: "Think about two? When you can tie the game to go into overtime? The percentages are not in your favor."

Remind me not to ask Tom Coughlin any questions having to do with percentages or the relative strength of the Giants defense and offense....

3
by BlueStarDude (not verified) :: Wed, 10/19/2005 - 10:26pm

Before going for 2 to win, you would at least have to consider the strength of your defense and of the opposing team's offense, as well as the opponet's return abilities and how good/poor your kickoff coverage is.

Wasn't Tice's one shining coaching moment going for 2 to win a meaningless game a few years back?

4
by thad (not verified) :: Wed, 10/19/2005 - 11:14pm

Gee i wished he had put more 80's references in there.
That article was horrible.

5
by MAW (not verified) :: Wed, 10/19/2005 - 11:15pm

He's dead on about the Manning Mastercard commercial. It's just as fresh and funny as last year's, and the clincher is the epilogue, where he says "Can you sign this for my brother? He's a big fan."

It cracks me up everytime.

6
by zlionsfan (not verified) :: Thu, 10/20/2005 - 12:56am

I think Manning's face helps those commercials out. He actually looks sincere in them. Kind of a Woody Harrelson-type character.

I like the guess-the-Ian-and-Solomon game. If Detroit were hosting Cleveland, that would be it for sure. The Lions have been an analog game for at least the past two weeks, which is nice - the bar where I watch games now has only HD TVs, so if the owner doesn't come around and switch the TV to 4:3, I don't have to see the score or the time left in the game.

I realize that isn't healthy behavior, but right now, it suits me just fine.

7
by Israel (not verified) :: Thu, 10/20/2005 - 1:48am

Are your odds actually better by playing for overtime, hoping to get the ball first or to stop the other team, hoping to get in field goal range, and then hoping your kicker doesn’t blow it?

Are there stats on overtime injuries someplace? Is it of any relevance to consider whether going into overtime increases the chance of injury - particularly when guys are tired? I'm wondering if that itself might make going for two at the end look more attractive.

8
by Kibbles (not verified) :: Thu, 10/20/2005 - 2:05am

Re #1: The number that I always hear is that 2-point conversions are successful about 40% of the time. That means if you think your team is as good as or better than theirs, then it's much smarter to play for overtime, where you have at least a 50% chance of winning.

On the other hand, if you think your team is much worse, and ESPECIALLY if you're a bad team with a good offense, then it's probably smarter to go for the 2PC to win it. Your odds will be a little bit better than 40% (because you have a good offense), and your odds of winning in OT will be a little less than 50% (because you're a worse team).

If you're a bad team with a better defense than offense, then you're better off going for Overtime and hoping for a miracle (INT, sack, fumble recovery, winning the coin toss and actually mounting a drive).

9
by Becephalus (not verified) :: Thu, 10/20/2005 - 5:21am

See I could swear I heard it was 48%, still not 50, but a lot closer. I cannot remember where I heard that (I wish I did). Perhaps it was just the percentage from one years and not an all time percentage?

10
by mawbrew (not verified) :: Thu, 10/20/2005 - 10:09am

I think Eisen is exactly right about the need to stay aggressive on offense. When you are up only one score, it's foolish to be more concerned with the clock than with maintaining possession. I think that's remains the case until you can be certain that the opposition will have less than 30 seconds left (and no timeouts!) should you have to punt.

11
by B (not verified) :: Thu, 10/20/2005 - 10:10am

There's always a chance you could miss the extra point, so kicking the PAT instead of going for two gives you a ~48% chance of winning the game.

12
by Sophandros (not verified) :: Thu, 10/20/2005 - 10:22am

#3: That game was against the Saints a few years back. I was there and almost broke my hand punching the chair in front of me.

13
by Theo (not verified) :: Thu, 10/20/2005 - 10:38am

How can your odds be better or worse when you go inovertime?
"when in overtime, you probably lose"
What about the other team?

14
by Ryan Mc (not verified) :: Thu, 10/20/2005 - 10:46am

I'm still waiting for a team to line up for an extra point and then run it in for two and the win on a fake. Any team ever tried that before?

15
by MJK (not verified) :: Thu, 10/20/2005 - 11:37am

William Krasker at footballcommentary.com has talked a lot about optimal strategies in the end game.

From a win probability point of view, here's how you figure it:

Prob of winning if you go for 2 = prob of making the 2 point conversion

Prob of winning if you play for OT = (prob of kicking extra point)*(0.5*prob of winning if you get the ball first in OT + 0.5*prob of winning if others get the ball first).

Krasker generally assumes that the probability of winning in OT is 0.5 (since both teams have played to a tie thus far, it makes sense that they might be equally good), but that might not necessarily be correct, as some people have already pointed out in this thread. Since teams make the XP ~98% of the time, by his assumptions it only makes sense to go for 2 if you think your odds are better than 49%. It's a pretty close call.

16
by Daniel Warehall (not verified) :: Thu, 10/20/2005 - 11:44am

The problem with going for two is that, if the team misses and loses, the coach is responsible.

If the team loses in overtime, it's the players' fault.

I think, even if the chance was 60%, most coaches wouldn't try it...

17
by Richie (not verified) :: Thu, 10/20/2005 - 1:41pm

(I didn't read the article.)

But, it's always such a fine line between "sitting on a lead" and "running up the score." Dallas vs Wash was the perfect example. Dallas had a small lead, but it seemed safe since Washington's offense was useless. Dallas seemed to run a lot of lame plays, and didn't do much on offense, and it let Washington come back.

On the other hand, I was watching a local high school game (starring a junior QB who is expected to be the most recruited player in the country next year), the home team was completely overmatched. These teams didn't belong on the same field. The good team left their main guys in until one drive of the third quarter. It felt like they were running up the score. But, sheesh, the team needs to let it's guys get some work in.