Advertisement

Dan Campbell's 4th-down decisions lead to more misguided and unnecessary debate on analytics

Brian Callahan is the new head coach of the Tennessee Titans and he was asked a question that every new coach and general manager will be asked for the next few years, at least.

It was about analytics.

It's a somewhat misunderstood, catch-all term, a faceless thing that's easy to blame when a fourth-down decision doesn't work out. But Callahan seemed to understand the role of analytics very well. His answer was a reasoned, measured explanation in the middle of an unreasonable debate on the matter.

"Analytics, it's a term that's a broad-stroke term but there's a lot of details that go into it," Callahan said in his introductory news conference. "At the end of the day you're using hard, concrete data to help inform decision-making and that's great. The more smart people, the more information that you're given, the better decisions you can make."

Easy. Numbers and statistics can make you smarter and guide decisions, though shouldn't necessarily be treated as an absolute. It shouldn't be more controversial than that, but it is.

The Detroit Lions lost the NFC championship game, and Dan Campbell's decisions on fourth down were a big part of the conversation afterward. (Photo by Kevin Sabitus/Getty Images)
The Detroit Lions lost the NFC championship game, and Dan Campbell's decisions on fourth down were a big part of the conversation afterward. (Photo by Kevin Sabitus/Getty Images)

Detroit Lions head coach Dan Campbell's fourth-down decisions in the NFC championship game are going to lead to a long offseason of debate on the matter. Heaven help us if Super Bowl LVIII comes down to a controversial decision that's blamed on "analytics."

San Francisco 49ers coach Kyle Shanahan has already shared his opinion on analytics, and he sees how it can be a detriment.

"I’ve gone to these meetings a lot in other buildings, and even here," Shanahan said in an interview with KNBR in October, via Niners Nation. "You meet with an analytics department and they bring out this whole book and there’s two million numbers to memorize.

"I’ve watched coaches try to do that, and you just melt during games. It’s too much info. And you realize, most of this stuff is pretty natural. And some of it is 50/50. Some of it is your preference.

"I try to keep my mind on it and go with what’s natural. And whenever there are those obvious ones, like when to call timeout here with two minutes, or 2:30, you know, all of those things where analytics are 100% right with and the math is totally there. I don’t even want to waste the time to learn that stuff because, ‘Hey guys, when it’s that automatic, just tell me, and that’s when I’ll do it.’

"But when it’s not that automatic. If it’s a 50/50 thing, just always let me go with my gut and my experience."

Dan Campbell wasn't following just analytics

Fourth-down decisions have become synonymous with analytics, in part because the fourth-down revolution started with a head coach and team that embraced statistics. In 2017, Philadelphia Eagles head coach Doug Pederson started going for it on fourth down regularly and at that point, it was a novel concept. A lot of it was based on the study of numbers, which indicated coaches weren't going for it enough.

Fast forward a few years, and most coaches have followed Pederson's lead. It's not strange to see coaches go for it in mostly any situation. It's also not uncommon to see a graphic on your TV screen before a fourth down saying whether the numbers agree or disagree with going for it.

A lot of the debates still revolve around 50/50 coaching decisions that are based on more than blindly following numbers, like Campbell's two controversial fourth-down decisions. Neither worked out because the passes were incomplete, and the two field-goal opportunities the Lions passed on were big talking points after Detroit lost 34-31. On the second one, the Lions passed on a 48-yard field-goal attempt that would have tied the game to go for it, and Jared Goff threw an incomplete pass. Did the numbers lead Campbell astray? Not really.

"I just felt really good about us converting, getting our momentum and not letting them play long ball," Campbell said after the game. "They were bleeding the clock out, that's what they do, and I wanted to get the upper hand back. It's easy hindsight and I get it. I get that. But I don't regret those decisions."

This embedded content is not available in your region.

There's a lot in that answer and nothing about numbers or statistics (also, a lot of analytically inclined people would cringe at the mention of "momentum" because they don't believe in it).

Based on his comments, Campbell seemed to base his numbers on the feel of the game and momentum, not numbers. You wouldn't know it if you search "Dan Campbell" and "analytics" this week.

What is the role of analytics in the NFL?

As with most things in life, the most reasonable place to be on the NFL's analytics debate is not on the far edges but somewhere in the middle. You can quickly find people who believe it's still 1988 and you "always take the points," and others who believe there's no such thing as momentum and if the numbers indicate there is any edge in a decision, that's unquestionably the right call and those who disagree are fossils.

Here's the thing about the endless debate over Campbell's fourth-down calls: They both were practically coin flips. The first fourth-down call, with the Lions ahead by 14, gave Detroit a 90.5% chance to win by going for it and a 90.3% chance to win by kicking a field goal, according to ESPN Analytics. The second decision said the win probability was 38.8% with a field goal and 39.1% by going for it. We're debating 0.2% edge for the first decision and 0.3% edge for the second. Those are practically rounding errors, though those on both sides of the debate would lead you to believe there was a black and white answer to be followed. The truth is there's something to be said about pushing a 14-point lead to 17 when you're in the position the Lions were in Sunday, and there's an argument to be made that nothing that happened afterward in the 49ers' comeback changes, no matter what Campbell decided. There was no right or wrong. It was just a 50/50 decision that didn't work out. And certainly not a referendum on analytics.

Let's go back to Callahan's comments to guide us instead. Analytics, to most reasonable people, are statistics that guide decisions but don't make them. That isn't just the decision on whether to go for it on fourth, but drafting and free-agency decisions, whether to try a 2-point conversion after a touchdown and any other difficult decision a coach or GM can make. As Callahan said, it's not bad to be smarter about those decisions before they're made.

We're still a long way from that type of nuance in the debate. After all, it took almost 100 years for NFL coaches to realize that going for it on fourth down isn't really a gamble. Just wait until the Super Bowl result changes on a decision that can be blamed on analytics.