​Steelers and Seahawks Teach Us That Good Offenses Should Go For Two

One of the most hot-button issues in football analytics is the two-point conversion. Whether it's criticizing a coach in catch-up mode for not going for them soon enough, or just a straightforward analysis of why teams don't go for it more often, this is a subject that just won't die. Coaches have tended to approach two-point conversions conservatively, since what might be considered a "risky" attempt by media and outside observers can be the next ticket out of town for them. Whenever you can isolate the effect of the coach as a net negative, it's a huge disincentive forward. And that's what has mainly stopped teams from going for two.

You'll see a lot of percentages dragged around based on score-recent history, such as what happened when Jack Del Rio went for two at the end of the Oakland-New Orleans game to secure a win. But these lack two real context inputs: how skilled the kicker is, and how much better or worse the offense is than the defense. I can tell you that NFL kickers made 94.5 percent of their extra-point tries in 2015, and that the number is hovering around the same rate so far this year. I can tell you that over the five year period between 2010-2014, two-point conversion percentages hovered around 47.5 percent.

But this doesn't account for the scenario. In the case of that Raiders-Saints game, the Raiders have one of the best offenses in the league, and the Saints have one of the worst defenses. The Saints also have a terrific offense, while the Raiders defense started out bad and has settled at around middling. There was no contest—going for two put the ball with the unit that Oakland should have had more faith in.

What you saw this weekend was the full extension of that philosophy, with Pittsburgh going for it on four separate occasions. Now, they failed all four times. But ultimately, what did it cost? The Steelers were robbed of a chance to be down only two with 1:55 left in the fourth quarter, but they scored the touchdown anyway. This offense is always going to have a chance to score and make a game of things.

Seattle, against New England, took an even bolder step and went for two to try to put the game away. Up seven, with a chance to kick an extra point to make it an eight-point game with four minutes left, Seattle went for two as well. They also failed, but it didn't change the math much. New England went from needing a score and a two-point conversion to just needing to score. Had Seattle made the two-point attempt, the game would've been in the hands of onside kicks rather than Tom Brady.

Football analytics has breached the question of going for two from a percentages standpoint, but what will sway coaches more than ever is matchup data. While these teams combined to go 0-for-5 on Sunday, it wouldn't be surprising to see more good offenses take the risk, further tilting the field in their favor by asking mediocre offenses to succeed at the two just to keep up.