"Why not? Put it in our hands. I want the ball. Any player would relish that opportunity."
That was how Ben Roethlisberger, oozing with certainty back in June, explained that he wanted the Pittsburgh Steelers' offense to go for two-point conversions after every single touchdown during the 2016 season. But is that kind of confidence warranted -- or just foolish?
Yes, the Steelersdid have the most two-point successes (eight) and the best two-point conversion rate in the league among teams to attempt the conversion more than once (an otherworldly 72.7 percent!) during the 2015 season. But it seems crazy to think they could replicate this success while increasing the frequency of their two-point attempts in 2016. Conventional football wisdom would certainly scorn such an aggressive strategy.
That said, Roethlisberger and Co. might be on to something. I examined the data and found the Steelers' two-point success to be statistically significant, after assessing varying factors such as the strength of opposing defenses, two-point attempt rate and offensive personnel. Thus, I believe going for two-point conversions after every touchdown might not be as outlandish an idea as others might think -- in fact, it might be smarter than just kicking for one extra point after every touchdown.
Don't get me wrong, this is a bold strategy
Make no mistake -- going for two-point conversions after every touchdown is a risky idea. After all, even if the Steelers converted the majority of their two-point attempts, missing just a few in one game could potentially lead to a loss. So why continue to explore the approach?
The frequency and success with which the Steelers went for two last season both marked radical departures from the NFL norm -- consider that their 11 attempts tied the 2002 New Orleans Saints for the most in NFL history, while the eight successes set a record. Moreover, the Steelers completed 72.7 percent of their two-point attempts last year, while the league completion rate (excluding the Steelers) was 44.5 percent. Given the sample size, that difference is significant enough to suggest the Steelers' success was not due to mere luck.
To determine whether the Steelers arrived at their success rate by mere chance, I looked to two-point data going all the way back to 2010 for answers. The league has attempted 362 two-point conversions since 2010 and completed 168 of them, excluding the Steelers' attempts, for a success rate of 46.4 percent. The Steelers, on the other hand, attempted 21 two-point conversions in that time, making 16 of them (a success rate of 76.2 percent). The results of a statistical test indicated that the Steelers did indeed perform better than the league average -- and they weren't just the lucky beneficiaries of a short-lived "hot streak"!
So if I'm Steelers coach Mike Tomlin, I'm starting to think about risk versus reward with the two-point conversion. Heck, I might even do the math myself: At what point does going for the two-point conversion stop being worth it?
Well, the Steelers and the league had around a 94 percent completion rate for extra-point kicks during the 2015 season. So, in aggregate, the Steelers can expect .94 points for every extra-point kick they attempt. As long as the expected points from two-point conversions is equal to or greater than .94, should the Steelers attempt two-point conversions rather than extra-points after every touchdown? That is to say, if they make at least 47 percent of their two-point conversions, are they golden? Well, maybe -- but it's not that simple. The expected gain from the two-point attempt may be higher than the extra-point attempt, but so is the risk. Missing just a few two-point conversions could set the Steelers behind their opponent.
Further, what if the Steelers' two-point success rate regresses back to the mean when they attempt more two-point conversions? What if the success rate is dependent upon certain players -- like Antonio Brown, Le'Veon Bell or even Big Ben himself -- being on the field and healthy? What if our data is skewed by the Steelers beating up on weak defenses? Tough questions, but let's take a look.
All hands on deck?
"Alright," say the skeptics. "Maybe there isn't a relationship between two-point attempt rates and two-point conversion rates ... but you have to have Le'Veon Bell, Antonio Brown and Big Ben all on the field to be successful!"
Well ... not quite. I looked at game film to find all the offensive groupings with which the Steelers attempted two-point conversions in 2015, and the results were pretty remarkable. All eight successes came with Big Ben and Brown on the field (one of the 11 attempts was with Landry Jones at quarterback, with Bell and Brown on the field; the rest were with Roethlisberger). But is the presence of Bell crucial to maintaining success with two-point tries? Surprisingly enough, it's not. When the entire trio was out there, no two-point conversions were completed -- and only one was attempted.
Of course, this doesn't mean Bell hurts the Steelers' two-point success rate. Instead, it's more a reflection of how last season went for Bell, who was suspended for the first two games and missed the last eight with injury, and the potency of the Steelers' offense despite missing one of its marquee players. Bell simply didn't rack up enough two-point opportunities to demonstrate an appreciable impact on his team's two-point conversion success rate, positive or negative.
Conventional wisdom, however, dictates that Bell -- who picked up nearly 5 yards per carry in limited action last season, and who racked up 2,215 yards from scrimmage in 2014 -- will have a wholly positive effect on his team's two-point success, since his play-making ability is all but palpable. All in all, we find that we don't need to have every single triplet on the field in order to have a dangerous two-point conversion team. And it's a good thing for Pittsburgh, which will be without Bell in the early going again this season while the back serves out a three-game suspension.
Preying on the weak?
We have to be careful to account for the defenses the Steelers played against when attempting two-point conversions. After all, it might be easier to convert two-point attempts against defenses that may be weak in the red zone or weak overall. As such, it's important to see if the Steelers loaded up on two-point attempts against weaker teams.
As a baseline test, I found the correlation coefficients between the two-point attempt rate of the Steelers and defensive rankings of the Steelers' opponents in their games, both in total defense and red-zone defense. Total defense and two-point attempt rate had a correlation coefficient of .229, while the correlation coefficient relating red-zone defense and two-point attempt rate was .477. These numbers seem to corroborate the argument that the Steelers relied on weak defenses to inflate their two-point conversion data, particularly against weak red-zone teams. Moreover, the average defensive rank of the teams the Steelers attempted two-point conversions against (18.89 overall, 13.89 in the red zone) was much lower than the average rank of teams the Steelers did not attempt two-point conversions against (9.43 overall, 11.42 in the red zone) -- meaning the Steelers saved their extra-point heroics for weaker defenses.
We should note that Roethlisberger missed three of the seven games in which the Steelers did not attempt a two-point conversion with injury. It makes sense that Tomlin would hesitate to attempt two-point conversions without his best quarterback. But what about the other four games -- against the Bengals in Weeks 8 and 14, against the Broncos in Week 15 and against the Ravens in Week 16 -- in which the Steelers did not attempt a two-point conversion despite Roethlisberger's presence? Once again, the average rank for the defenses against whom the Steelers did not attempt a two-point conversion with Roethlisberger under center (7.75 overall, 10.25 in the red zone) was much better than the average rank for the defenses the Steelers did attempt two-point conversions against.
Since we found that the Steelers converted their two-point attempts against weaker defensive teams, one could argue that it might not be so shrewd to attempt them against opponents of a higher caliber. However, I would argue that increasing the frequency of two-point attempts might still be a viable approach. While it would bring more risk against better defensive teams, there is still the potential for more reward. Remember that the Steelers posted the fifth-best scoring-drive efficiency rate in the league in 2015 (40.3 percent). In other words, the Steelers are likely to have plenty of scoring chances with which to buffer that risk.
What about the Steelers' kicker troubles?
There's another potential explanation for the uptick in two-point tries to consider. Pittsburgh lost kicker Shaun Suisham for the year with a torn ACL in August 2015. With the league moving extra-point attempts back (from a distance of 20 yards to a distance of 33 yards), would the team be pressed to go for two more often, given that it would be relying on a cobbled-together solution at the kicker position?
Well, the Steelers did attempt five two-point conversions the rest of the preseason (making three) after losing Suisham, and they continued the trend into the regular season, going three for four on two-point tries through Week 4. Replacement kicker Josh Scobee, meanwhile, was six for seven on extra-point tries in that same span. He also missed two high-profile field-goal attempts in an overtime loss to the Ravens in Week 4, prompting his release and the signing of Chris Boswell. But while the Steelers might have made headlines with their kicker woes, the fact is, over the course of the season, Scobee and Boswell missed just one extra-point kick each, giving the team a 32 for 34 success rate (94.1 percent) on extra-point kicks, which was about the same as the rest of the NFL in 2015 (94.2 percent). While we can't know whether a lack of confidence in their kickers led the Steelers to try for two more frequently, the numbers suggests that their kickers were as reliable with extra-point kicks as the rest of the league's kickers.
Conclusion
So is going for two-point conversions after every touchdown crazy? Maybe -- and obviously, in their first chance to put the strategy to the test against the Redskins in Week 1, the team refrained, opting to kick for the extra point after every touchdown. But I still think the Steelers' success in the recent past implies that it might become a viable and consistent option for post-touchdown scoring, and should at least be considered by coaches and general managers around the league. For high-scoring offenses like the Steelers, who were tied for fourth in the NFL in points scored during the 2015 season, aggressive two-point conversion strategies could have made a huge difference -- especially in the four games that the Steelers lost by one score or less. Consider that the team went for two in just one of those losses, a successful attempt in Week 1.
Understandably, there may be some concerns, as a sudden deviation in offensive strategy this large is unprecedented; however, I believe Big Ben and Tomlin can find themselves ahead of the NFL if they capitalize on their otherworldly two-point success rate now -- before the rest of the league has a chance to catch up.
Sri Nimmagadda is a junior at Princeton University studying public policy, finance and statistics/machine learning. Sri has extensive experience working with data analytics and sports data as a former research intern with the NFL.
