Did “Analytics,” a.k.a. “Metrics,” win the NFC title game for Tampa Bay and make Matt LaFleur persona non grata with Green Bay Packer fans?
… “The NFL’s Analytics Revolution Has Arrived”
Many, if not most, Green Bay Packer fans are outraged at their football coach, Matt LaFleur, who is in just his second year and directed his team to the NFC title game last Sunday.
The fans believe that their coach made a horrible mistake when he decided to take a field goal with just over two minutes in its game with Tampa Bay. That decision narrowed the Tampa Bay lead to five points instead of eight, but with the probably league MVP in quarterback Aaron Rodgers, why didn’t he try to score the TD and then, hopefully, tie the game with a 2-point conversion?
Well, my belief is that is “Much Ado About Nothing.” In truth I think that this game was won in the first half, at the end, when Bucs Coach Bruce Arians made what was a very crazy call that worked.
The similarity, though, is that both decisions may have been made by metrics, or analytics.
What does that mean?
Analogy to baseball
I do not watch much baseball — I would rather watch the grass grow, it is that boring. However, in the little time I have spent doing so, I have noticed a few things that are interesting.
These managers are consumed by analytics. They will move their infield and outfield in crazy ways based upon the data that they have put together in this way. The data is so sophisticated that managers and coaches can provide players with a unique set of information that was not available previously.
That analytic revolution has been somewhat slower coming to football, but it has arrived, and it may have been the reason that Arians pulled quarterback Tom Brady off the sideline on a fourth down at his own 45 with 22 seconds remaining in the first half.
What is this data?
A writer a few years ago proclaimed that the NFL analytic revolution was finally taking place,
All of these trends point to one thing: Football’s analytics moment has arrived.
We’ve reached this high point for a couple of reasons. The rise of smarter, younger GMs and coaches is part of it. A bigger part of it, though, is the spread of the NFL’s player-tracking data, which is being shared leaguewide for the first time this season. Having access to that data allows teams to build models to analyze plays and players differently, and to simply know more about the game.
That’s been a boon to a movement that had already been embraced by a handful of the smartest teams. As other teams try to catch up, they’ve created an arms race to get the best numbers. Essentially, the smartest teams are getting significantly smarter, the average teams are trying to get better, and the dumbest teams are going to be very dumb if they don’t act soon.
Kevin Clark, “The NFL’s analytics revolution has arrived,” The Ringer, December 19, 2018
That may be the reason that LeFleur decided to go for the field goal: Because the data indicated that it was the best route to a Packer win.
LaFleur is being hammered for the decision
As a former offensive coordinator, I was very surprised by LaFleur’s decision with just over two minutes remaining. Giving the ball back to Tom Brady, with six Super Bowl wins to his credit, with two minutes left seemed like folly to me.
However, after looking at the possibilities, I can see why the analytics may have told him that. First, Brady had thrown three picks in the second half. Plug that number into the computer and it may override the fact that he has six Super Bowl rings.
Second, Brady had been erratic in the second half, too, so that is a consideration.
Third, the Packers needed to score a touchdown on fourth down and goal, they needed to score the 2-point conversion, and they had to rely on on a porous defense that gave up a touchdown at the end of the first half, which I believe was really the difference in the game.
Why was Green Bay in a man defense at that time instead of a zone? The analytics fail me here.
So, everyone in America was second-guessing that call.
SI argument was a common one
Conor Orr presents a very interesting argument that LaFleur was way off base, but he notes the analytics challenge,
- If you were anything like us, you may have spent the first few minutes after the Packers decided to kick a field goal on fourth-and-goal from the eight-yard line with 2:05 remaining in the NFC championship wondering whether there was something you might have missed.
The refinement of in-game decision-making analytics, which were some of the first widely utilized probability concepts of the modern analytics boon, are complex enough to throw the lay fan a curveball from time to time. Perhaps a field goal, which cut Green Bay’s deficit from eight to five, was the sensible maneuver.
Of course, that wasn’t the case. ESPN’s analytics department said the disparity was not large, but going for it on fourth down and subsequently trying for the two-point conversion would have given the Packers roughly a 0.5% greater chance of winning the game than kicking the field goal and taking the chance that they would get another possession and score a touchdown.
EdjSports, another fine analytics outfit, said that the decision cost Green Bay 3% pre-snap in their Game Winning Chance model. And in fairness to LaFleur, at least one model showed the field goal as a slight edge, though it's unknown if that model took into account 1) Having Aaron Rodgers and 2) Giving the ball to Tom Brady.
Conor Orr, “Matt LaFleur's Misplaced Trust Could Cost the Packers
More Than Just a Game,” Sports Illustrated, January 25, 2021
So, if the analytics showed just a .5 percent greater chance at running a play on fourth down instead of kicking the field goal, there is not much difference between the two decisions.
Practical analysis
So, in a practical sense, what if Rodgers completed a pass for the TD and the Packers tied the game with a 2-point conversion, how would the analytics have presented this scenario?
Brady has two minutes to move into field goal territory to win the game after Rodgers tied it. Remember Brady’s comeback a few years ago against the Atlantic Falcons when they had about an 18-point lead in the game? The Patriots won that game in OT after being down by a slew of points, and the hero was Tom Brady, who had not played well early in that game, either, as he had not in the second half in this game.
Can you mathematically put that information into a computer and see what comes out? Perhaps the result will be kicking a FG, like the Packers’ coach did.
Or perhaps not.
My guess is that Bruce Arians may have used analytics to score that TD right before halftime to give the Bucs a comfortable lead.
He definitely changed his mind and put the ball in the hands of his QB who had been to a slew of Super Bowls in his career.
Conclusion
I really have no idea what the metrics said in any of these situations. Why LaFleur picked his defense instead of his future Hall of Fame QB to win the game is still puzzling, despite the analytics.
A very interesting choice, and one that will be debated in the dairy lands of Wisconsin for years to come.
Comments
Post a Comment