The hottest team in the NHL right now is without a doubt the Washington Capitals. They’re 11-1-1 in their last 13 games and have stormed back from second last in the Eastern conference to take control of the Southeast division. They’re on the verge of being the hottest team heading into the playoffs – a team no one ever wants to play – but the question is; are they for real?
The biggest noticeable difference over the last six weeks or so has been Alex Ovechkin. He’s returned to his old form and is on pace for 53 goals in an 82 game season. But the biggest difference that has gone unnoticed is that the Capitals’ luck has turned around.
PDO is an advanced stat in hockey that tries to capture the luck factor. It is a combination of five-on-five shooting percentage and save percentage, which Gabriel Desjardins of behindthenet.ca says are primarily luck-driven stats in the short term. Over a large enough sample size — say a few seasons — you can use those stats to separate the good players from the bad players. But throughout the course of a single season these stats will fluctuate wildly and regress to the mean because they are primarily luck driven. Desjardins, who is one of the masterminds of analytics in hockey, says that if you were to get to know one advanced stat, he would recommend PDO.
By combining five-on-five shooting and save percentage, you get a PDO number that is above or below the average of 1000. Those well above it have been really lucky, those well below it have not. On February 25th, the Capitals were one of the unluckiest teams in the NHL. Their PDO was 979 and ranked 26th in the league. Since then, their luck has turned around and their PDO is now 1012, which is tied for ninth in the NHL.
So the question is, has the Capitals turnaround been fuelled more by luck or are they actually a good team that has started to click? According to Fenwick Close Percentage (FCP), they’re just on a lucky hot streak.
Fenwick is a stat that measures shot attempts (all goals, shots, missed shots – but not blocked shots like Corsi), which essentially translates into puck possession. The “Close” in “Fenwick Close Percentage” means that the stat is adjusted for the score of the game, because shot attempts stats can get skewed if a team is throwing everything on net because they’re down by a few goals or a team is sitting back on a lead. So FCP only measures shot attempt differential in close games, which means when the score is tied, or within two in the first or second period. Teams that have good FCP numbers — or in other words teams that control puck possession in close games — have a higher correlation to winning than any other stat.
So, if FCP is the best predictor of which teams are actually playing well and not just getting lucky, how are the Capitals doing? Well, even after their hot streak their FCP is 47.53, which ranks 22nd in the NHL. They’re playing like a non-playoff team, but since their luck has turned around and they’ve gone on a hot streak, they’re in the playoffs.
Even though most teams don’t want to play the team that’s hot going in to the playoffs, there’s not much reason to fear the Capitals. Just like their luck turned around for the better over the last month and a half, it will likely fall back down to Earth because they’re not really playing good hockey. The six seed in the East should still be the most coveted non-home ice playoff spot in the conference.