This is a follow up to Nestor's article about going for it on 4th down.

 

• First, already talked about going for the FG with a 4th and 2 at Cal’s 6 yard line. When it was pretty clear that Bruins needed TDs instead of FGs, CRN decided to play it safe.
  
  
• Second, we saw the same situation play out again later in the third quarter when with a 4th and 3 on Cal’s 28, Neuheisel again went with the Donahue mindset of playing it safe and going for three.
  
  
• Third, this is where probably many will not agree with me but I will share it anyway. Jahvid Best’s 93 yard run I thought was the game changing moment that pretty much decided the outcome of the game. However, the pivotal decision in my mind came right before UCLA punted in that second quarter. Bruins had 4th and less than a yard to go on their own 46 yard line. Up until that time Bruins were doing all right on offense (actually we had a decent day on offense against a slightly better than average defense yesterday). Anyway with a less than a yard to go and down by only 7 (14-21), Neuheisel took the safe route (the Donahue one) and decided to play for field position by depending on Locke’s leg. Well Locke did boom one (for 46 yards) but it didn’t do much good as the D got obliterated by Best in the subsequent big play. What I am saying here that’s the kind of moment when Neuheisel instead of playing by the conventional Donahue rules, could send his signal to rest of his team that he had enough trust in his young talent and offense to go for less than a yard.

 

Strongly recommend this whole 4 part article to anyone who has time and is interested in football strategy....anyway, the summary of it is that they calculated the expected points from each yard line, ran some probabilities of conversion percentages on 4th down and field goals from every yard line, as well as the expected points from punts. NFL data but the probabilities for NCAA ball probably isnt too far off.

I was North Campus but even I could follow and crunch some numbers (ie, read a graph), so all of you can understand this. I went ahead and figured out how these decisions played out according to the probabilities. Here are the three charts that are used.

4th and 2 at the 6-yard line: The expected point value of a field goal at the 6 yard line is about 2.2, maybe 2.3 points (3 points - almost guaranteed field goal from that range, minus 0.7 points, which is the expected points you will give up when you kick off following a FG). On the 2nd chart, a 4th-and-2 attempt at the 6 yard line will usually net you about 2.8, 2.9 points, factoring in the probabilities of converting that as well as the expected points for having a first down at the 4 yard line and subtracting the expected points that you would lose if the other team took possession on the 6-yard line. You can see here that having a 1st down backed up deep in your own territory actually gives you negative expected points, so if you go for it and fail, it's not all that bad to cough up the ball deep into Cal territory -  Cal is more likely to give up points down the road if they take possession backed up that deep (obviously, there are always exceptions, as Jahivd Best proved, and these are just odds). In this case, according to the numbers, going for it is a wiser decision.

4th and 3 at the 28-yard line: The expected points difference here is a little closer. A FG from that range will probably net you about 1.2 points (probability of converting a FG from that range, subtracted the EP for the other team receiving possession after the score/them receiving possession 7 yards behind the line of scrimmage on a miss), while going for it looks to be about 1.5 or 1.6 expected points. A little closer this time but going for it still looks to be the wiser decision.

4th and 1 at the 46-yard line: Here you go to the expected value of punts - factoring in the average distance of a punt from each yard line, and the expected points the other team will have from possessing the ball at that yard line. From the 46-yard line it's a little less than 0, maybe -0.1 expected points. From that distance the average is about 30 yards, giving Cal possession at their own 10, where their expected points for that drive is 0.1. If you look at the 2nd chart, the highest blue line is 4th-and-1, where the expected points are better than any other distance. At the 46-yard line, it's about 1.1 expected points to go for it, clearly better than the value for a punt. Again, this is a better decision according to the odds.

However, football isn't math, the odds are only odds. We punted, backed Cal up to the 7 yard line, where their expected points are actually about -0.5, or a positive 0.5 for us. In reality they scored 7. The odds don't always play out but they are the averages of what is expected to happen. Obviously there are going to be different depending on your team and players - Kai Forbath isn't getting 1.5 points on a field goal from the 28, he is getting almost guaranteed 3 points every time (minus 0.7 on the ensuring kickoff, the EP of Cal for getting the ball at the 27-yard line) because he does not know how to miss.

However I do think that we should have gone for at least the first and the last situation, short-yardage deals where the odds of converting are slightly higher (see the gap between 4th-and-1 and 4th-and-everything else on that middle graph). For what it's worth here is their graph of what to do/not do on 4th down, based on these findings. Here is a graph that has all three of the above ones merged, you can clearly see that there were higher expected points at the 6, 28, and 46 yard lines for going for it (colored lines) than kicking (black line). Interesting to note that 4th-and-2 is almost always a better decision than kicking from anywhere on the field, and 4th-and-1 is always a better decision, according to this data.

There's a ton of reasons why you don't go for it on 4th down - at that site they write about how people fear potential loss rather than prospective gain, even though a punt is considered a "loss" it doesn't really feel like it, while a failed 4th down does and is rarely considered a "good call". If you miss a FG that's a loss, but it's the kicker's fault - it was still a "good call". Job security is probably another, a guy like Pete Carroll is not getting fired anytime soon so he can afford to be more aggressive on 4th down. Going for it and not converting in a national championship Rose Bowl game that you end up losing would have fans calling for your job with almost any other coach. I feel like if Neuheisel had gone 0-3 on those 4th downs we'd be saying the other thing.

Addendum

Here is some more reading from the site (great insights into the game if you have time) on the topic of 4th downs in “no-man’s land” – area where you could either attempt a long field goal or a short punt – again, NFL statistics, but I don’t think that this matters as much, more of a general coaching thing.

On 4th down from identical field positions, coaches tend to attempt FGs more often with shorter distances to go and punt more often with longer distances to go. For example, when kicking on 4th and 1 from the 32, coaches went for the FG 100% of the time. But when kicking on 4th and long (7+ yds) from the 32, coaches went for the FG less often—80% of the time.

Interesting, that for something that does not matter at all if you punt or kick a FG (distance to go), there is still a mental effect on the decision. I’ll just copy since I can’t really say it better than they did. In 4th and 1 they kicked the FG about 85% of the time, compared to only about 65% of the time, from the same yardline.

It’s as if teams with 4th and short feel like they came so close that they really want to come away from the drive with points. And teams with 4th and long feel more resigned to giving up the hope for points, preferring to pin the opponent deep in his own territory. Any other theories?

The closer you are, ie, a 4th-and 2 near the goalline, coaches may feel that they need to at least come away with something. I went back and checked the drive charts for those two FG decisions – both were 50+ yard drive that started in our own territory. Maybe if we had started these drives as a result of a turnover, in Cal territory, we would have gone for a 4th down or two, as you would feel as if you hadn’t invested as much in the drive and would be more likely to gamble at the end. Studying field position vs. 4th down decisions would be an interesting study.

I think, NFL or no, that these statistics and numbers show that there is definitely a mental component to making these decisions, that it’s the result of either by the fear of coming away with no points, the perception of greater risk, pressure from fans and media, conventional norms, poor play by the offense, etc rather than simple probabilities and expectations of success. However I don’t think that going against these numbers automatically makes the decision “bad” – these other factors do exist and they do need to be taken into consideration.

These may be statistics from NFL games but I do believe that there are many of these “mental roadblocks” among most coaches at any level that prevent them from rethinking the norms – teams have been punting on 4th down for a century (even though it was a different game back then, when offense were not rolling up and down the field), so coaches tend to punt on 4th down when it would appear that this may not always be the wisest decision to score points according to the long-term odds.

I think that punting for field position is valid in cases, as in games where your offense isn’t moving well (a throwback to those old days when punting became the wisest decision).