Diving into the advanced stats can tell us which teams’ fates are tied to good or bad luck so far this season.
The first two weeks of the NHL season are for “It’s way too early.”
Now that we’re more than halfway through October, we can get away with “It’s way to early, but…”
Most teams have played six or seven games. Averaged out at 6.5, that constitutes 7.9 percent of the NHL season. It’s a tiny sample, but it’s not nothing. At some point we have to acknowledge that the Toronto Maple Leafs are a superpower, that the New Jersey Devils are better than we thought, that the Montreal Canadiens really are in trouble…
…maybe. What do the underlying analytics tell us? If we look at Corsi, save percentage, shooting percentage and PDO, we can separate the good from the lucky and the bad from the unlucky.
All analytics numbers below are courtesy of corsica.hockey.
FAST-STARTING TEAMS WE CAN TRUST
Columbus Blue Jackets
The Jackets, playing lockdown defense under coach John Tortorella, allow 2.17 goals per game, placing them third in the NHL. A another reason why, of course, is the stellar goaltending of Sergei Bobrovsky. He’s helped the team to 5-on-5 save percentage of .941, good for seventh in the league. Considering Bobrovsky’s year-long 5-on-5 SP was .939 last year during his Vezina-winning season, we know it’s not a fluke when he starts fast again. He’s simply maintaining elite play. The Jackets rank a so-so 13th in shooting percentage, meaning their PDO is merely good and not great at ninth (PDO combines a team’s shooting percentage and save percentage to approximate “luck”). They’re a top-four team in Corsi percentage and rank second in Corsi Against per 60, so nothing about the Jackets looks lucky. They’re for real.
Toronto Maple Leafs
The Leafs are the inverse of the Jackets in that their PDO is heavily tied to a sky-high 5-on-5 team shooting percentage of 12.96, which ranks them second in the NHL. That number should come down – though not as much as one might think, as the Leafs have a ton of skilled shooters. High-skill teams can sustain higher shooting percentages year round because of finishing ability. What works in the Leafs favor: with Frederik Andersen off to a mediocre start in net, their team save percentage ranks bottom-third in the league 5-on-5. Andersen opened slowly last year only to correct his game. His numbers should improve, so opponents will convert fewer of their shots, offsetting any drop in Toronto’s shooting percentage. The numbers also tell us Toronto’s team defense isn’t as sloppy as it’s reputed to be. The Leafs actually allow the ninth-fewest shot attempts per game 5-on-5 while generating the eighth-most. That balance helps the Leafs rank seventh in overall 5-on-5 Corsi. Trust in the hot start.
FAST-STARTING TEAMS WE CAN’T TRUST
Chicago Blackhawks, New Jersey Devils, Vegas Golden Knights
I’ve lumped all three of these teams into one category because they all profile similarly as hyper-lucky, ranking among the league leaders in shooting percentage and save percentage early on, jacking up their PDOs.
Oddly enough, of that trio, the team that looks the least believable under the hood is…Chicago? The Hawks have the No. 1 shooting percentage and No. 2 save percentage in the NHL 5-on-5, masking the fact they’re 26th in Corsi. They’re a bottom-four team in terms of preventing shot attempts. Once the puck-luck regresses to the mean, might the pre-season concerns about Chicago’s depth on defense float to the surface?
St. Louis Blues
In terms of shot attempts hurled at their net versus what they generate on the other team’s net at even strength…the Blues are getting owned. They have the 30th-best overall Corsi mark, ahead of only Ottawa, another team playing over its head but that didn’t have Erik Karlsson to start the year. The Blues are 29th in Corsi For per 60 and 25th in Corsi Against. They’ve won five of their first seven games, but linemates Vladimir Tarasenko and Jaden Schwartz have scored 10 of their 18 goals. The Blues are a one-line team right now.
SLOW STARTING TEAMS WHO SHOULDN’T PANIC
It’s not just that Leon Draisaitl is concussed or that Cam Talbot isn’t looking like the goalie who finished fourth in Vezina Trophy voting last year. The Oilers are the most dominant team in the game on the offensive side of the puck, leading the NHL with a staggering 5-on-5 Corsi For per 60 of 79.25. Since they grade out well defensively too, they have the best overall possession numbers of any squad. Who cares if Connor McDavid jacks up those numbers? He counts. He’s on the ice playing the game and controlling it. The Oil’s shooting percentage is a microscopic 3.92. This team has been extremely unlucky. Don’t freak out about Edmonton’s slow start.
The Habs are a disaster to start the year, with calls for GM Marc Bergervin’s firing all over Twitter. But while this team makes a lot of questionable personnel decisions in the off-season and mid-season, the analytics actually suggest Montreal is the least lucky team in hockey to start the year. The Habs rank dead last in shooting percentage and save percentage and PDO. The pessimistic viewpoint would suggest the poor shooting percentage reflects a dearth of talent at forward, but the Habs sit at 3.43. The worst mark by any team across 82 games in 2016-17 was almost double that at 6.21. Max Pacioretty, a career 11.3 percent shooter, sits at 4.3 and has one goal. That’ll change soon. Montreal also has the league’s worst SP despite owning Carey Price. Come on. We know better times are ahead. That doesn’t mean Montreal will emerge as a top-flight contender, but this team won’t be contending for the most ping-pong balls in the draft lottery, either.
SLOW-STARTING TEAMS WHO SHOULD PANIC
Buffalo’s 5-on-5 shooting percentage looks a bit unlucky, but 6.25 isn’t so low that it’s guaranteed to correct itself. The Sabres’ finished at 6.41 last year. The number thus might simply reflect a talent-thin roster. That lack of finish coupled with the league’s fourth-worst Corsi For per 60 means the Sabres aren’t just struggling to score when they do get chances…they simply aren’t getting enough chances. Crazy to think a Jack Eichel-led team could have so much trouble pushing the puck into the other team’s zone.
The Desert Dogs rate ho-hum in the possession stats, too, but what really hurts them right now a 5-on-5 team SP near the bottom of the league. It’s typically foolish to judge players on early-season sample sizes…unless those players are unproven commodities promoted to major roles. We were pretty sure Antti Raanta was starter material but didn’t know that. So far, he’s alternated shaky and stellar efforts. It matters more when a guy like Raanta struggles early than when the proven Price or Talbot does. It doesn’t help that Raanta’s platoon mate, Louis Domingue, has been especially bad, dropping all three of his decisions with an .862 SP.