I have a little bit of an obsession with fourth down decisions. It has always puzzled me when coaches would purposefully give the ball to the other team, even more so when a first down and more points were just a few feet away.
As football has moved from an Ol’ Boys network to becoming more analytically-minded, this mindset has slowly shifted — not as fast as I would like, but it has been noticeable.
With all this in mind I will make an effort to try to take a look at all of the fourth down decisions the 49ers make this season and compare them to what would be considered optimal. To accomplish this I will use the handy Fourth Down Calculator from Advanced Football Stats. This will give us both the expected points and the change in win probability based on the decision made.
Fourth down decisions against the Packers
The 49ers had a total of 10 fourth downs. They went for it once, punted (or were in punt formation) six times, and attempted three field goals. Below is the list of decisions that the 49ers faced:
|Quarter||Score Differential||Field Position||Yards To Go||Time||Actual Choice||Optimal Choice|
|1||0||SF 35||2||12:42||Punt||Go For It|
|2||0||GB 43||5||11:57||Punt||Go For It|
|2||7||GB 30||6||1:32||Field Goal||Go For It|
|4||0||GB 9||9||14:17||Field Goal||Field Goal|
|4||3||GB 36||2||2:42||Go For It||Go For It|
|4||3||GB 15||4||0:26||Field Goal||Go For It|
Harbaugh did a pretty good job, overall. The 49ers only faced a pair of 4th-and-short situations, with Harbaugh taking the more aggressive (and more importantly, optimal) choice once. In all he went 6-of-8 on his decisions, with two of them being so borderline that I wouldn’t call them non-optimal choices.
Let’s take a closer look at the decisions that would be considered non-optimal:
In this situation the numbers suggest that going for it will yield both more expected points and a higher probability of winning the game. Going for it will give teams an expected points of -0.43, compared to -0.72 for punting. It will also increase the chances of winning the game by 2% to 49%, compared to 47%.
This wasn’t a huge deviation from optimal, but seeing as the Packers had one of the worst rush defenses in the league last year while the 49ers had one of the best rush offenses this one would point to an opportunity for Harbaugh to be aggressive. Since an offense like Green Bay’s is going to get their points, it would have been nice to see the 49ers take the aggressive approach. But it wasn’t that big of a deal to see them punt.
Second Quarter, 11:57 remaining, Ball on GB 43, 5 yards to go:
This is one of the punts that really drove me crazy. This part of the field it is really too close to the opponent’s end zone to have a lot of upside from punting. The fourth down calculator agrees — going for it will give teams an expected points of 0.53, compared to -0.04 for punting or -1.37 for attempting a field goal. It will also increase the chances of winning the game by 2% to 52%, compared to 50%.
Second Quarter, 1:32 remaining, Ball on GB 30, 6 yards to go:
This is one of the fourth down decisions where one could argue either way. The expected points are higher going for the first down while the win probability is higher attempting the field goal. In the end it probably depends on what a coach has more confidence in, his offense/defense or his special teams.
The calculator says that going for it will give teams an expected points of 1.21 compared to 1.00 for attempting a field goal, however going for it will also decrease the chances of winning the game by 1% to 76%, compared to 77%.
I can’t really blame Harbaugh too much here; with the clock winding down on the half it is important to get points and with his faith in the defense he would hope that they could keep Green Bay off the board (they didn’t, but we have the advantage of hindsight).
Fourth Quarter, 0:26 remaining, Ball on opponent 15, 4 yards to go:
The final decision is another one where you could argue both sides convincingly. If the 49ers go for it and convert, the game is over. If they fail, it still most likely forces Green Bay to go 85+ yards to win or 50 yards for a chance to kick a game-tying field goal with about 20 seconds remaining. Kicking the field goal has about an 80% success rate and gives the ball to Green Bay with 10-15 seconds to go all the way for a touchdown.
Here is what the calculator gives us for that situation: going for it will give the teams an expected points of 2.17, compared to 2.06 for attempting a field goal. Going for it will also increase the chances of winning the game by 1% to 98%, compared to 97%.
Either way, the 49ers were in a great position to win and did.
I will update this after every game to keep a cumulative tally of how Harbaugh does this season.
- 6-of-8 optimal decisions
- -1.18 expected points from those decisions
- -4% on win probability