Analytics with Alison: Overtimes, faceoffs, and the connection in between

A little over a week ago, Seattle was playing its 23rd game of the season in Detroit. Having forced overtime, the Kraken were fighting for a second standings point from the game when a stoppage in play was called. And that's when a question was raised by J.T. Brown.
"What is the percentage of teams that win in overtime that have also won the opening draw (in overtime)? May not be even a correlation, but would be cool to know."

There wasn't a ton of time to break down ideas in the moment, but Brown's general thought process was "teams that get possession are then able to hold the puck for a long time."
The Kraken have played in only two overtimes, but Brown's theory seemed to be in play. When Seattle went to overtime in Columbus, it was the Blue Jackets that won the opening draw. 2:16 later, Patrik Laine sealed the victory for his team. In Detroit, Dylan Larkin of the Red Wings won the initial faceoff in bonus time and, while it would take a shootout, they too, were victorious over the Kraken.
Does a faceoff win to start overtime make it more likely a team will win the game?
Let's dig in.
First and foremost, we went to Sportlogiq to ask them if they could tell us how many teams that win the initial faceoff in the "fourth period" end up winning the game. The answer? If we include all NHL games back to the 2017-18 season, 1085 games have gone to overtime. Of those contests, 718 have been decided in 3-on-3 play (or 4-on-3 if a penalty is called), and not gone to a shootout. Of those 718 games, 375 (or 52.23%) were won by the team that won the opening faceoff.

Would seem conclusive, no? Well, not really. There's more to it than that. Remember, we wanted to explore not just if the team that won the faceoff won in OT, but also if the faceoff victory tied directly to an extended possession that led to the score. Unfortunately, the data seems to say otherwise. Even though more than half of the games were won by a team that won the OT faceoff, only 54 of those victories (7.52%) came off the first possession.

And this brings us to a whole other can of worms. How much do faceoffs matter? Work by
Craig Tabita
, Gabe Desjardins (
2011
), and Arik Parnass (
2015
, now with the Colorado Avalanche) demonstrated that faceoffs don't have as big an impact on shot rates (AKA possession) as we might think and, in fact, any benefit from a faceoff won has dissipated after 10 seconds of play.
I understand that this might make some of us shake our heads in disagreement. I get it. But for every faceoff won that leads to what we see as "good" there's also a faceoff scored like this. While the Blue Jackets won the opening faceoff in overtime against the Kraken, there was one additional faceoff before the game ended and it was between Morgan Geekie and Boone Jenner. Geekie was awarded the "win" here.

So, part of the problem we currently have with faceoffs is not just that their impact is short-lived, is that they can be scored to show which team immediately touched the puck but that in no way means they retained possession.
So, what does matter in overtime?
In 2021, The Athletic's Dom Luszczyszyn conducted
an in-depth study of every second of overtime from the 2019-20 season
. Here were the ingredients to overtime success that he identified:
This all makes sense as well, no? Let's juxtapose some of these criteria which we can measure against the albeit small sample size of OT games that the Kraken have played.

It is of course massively important to realize that two games does not a reliable sample size make; but, in the small amount of overtime that the Kraken have played, there seems to be more that factors into these outcomes than just faceoffs, especially if we use Luszczyszyn's metrics as an outline for keys to success.
And perhaps even more importantly, this doesn't make Brown's observation wrong! His point of sustained and controlled possession aligns with Luszczyszyn's findings as well and all parties (including me!) recognize that we can always learn more about what does and doesn't contribute to success in different game situations. The point is to always ask the question, dig a little deeper, and find out what we didn't know before.