It has taken a while for sports to embrace advanced statistics. It really started in baseball when the book Moneyball came out in 2003. Moneyball was about the 2001 Oakland Athletics new approach to baseball where the GM Billy Beane would value players based on stats or sabermetrics rather than what scouts used to just look at players. Although Oakland still has never won a World Series, the Athletics’ went on to win 20 straight games at one point that year with this new method. Soon every team was following suit. Then Boston Red Sox GM Theo Epstein hired a bunch of analysts in 2004, most notably Bill James. The Red Sox went on to win the World Series that year. It has changed the game of baseball forever.
It’s taken about 20 years for hockey to start embrace advanced statistics. In fact, the NHL has advanced statistics on their site for a couple of years now. Also a fair amount of NHL teams have hired experts in the analytics field. Some teams even have departments for analytics. The most notable guys who have been in the forefront are Arizona Coyotes General Manager John Chayka and Assistant General Manager of the Toronto Maple Leafs, Kyle Dubas. Chayka was the CEO of statletes.com before he got the job. Arizona is one of the more promising teams coming. Dubas was a successful GM for the Sault-St Marie Greyhounds by using analytics as the main factor in their decisions. Dubas was so impressive that Toronto hired him to help rebuild their team. It is a process, but both Toronto and Arizona are in the right direction with those two guys in key roles. Even given all this, there are still doubters in advanced stats.
Advanced statistics are primarily used to try to predict the future of hockey game. This is ultimately what every fantasy owner tries to do for every decision they make. Whether it’s evaluating a player to draft, trade, or pick up on the waiver wire, owners are constantly trying to figure out what player stats are sustainable. Fantasy sports, in general, calculates stats to decide who wins. Although attributes like grit and heart are important skills to have in the game of hockey, they are not quantifiable. Fantasy hockey is all about stats, and advanced stats can help in that.
Here are some common advanced statistics for hockey: Corsi percentage (CF%) is the original and biggest statistic in the hockey analytics community. It is called shot attempts (SAT) by the NHL. CF% is the sum of shots on goal, missed shots and blocked shots over the shots against, missed shots against and blocked shots against at even strength. It is basically the +/- rating for players but it’s shots taken that are counted instead of goals. This is seen as a way to evaluate a players possession on the ice. A typical player has a CF% between 40 to 60. The SAT leader was Patrice Bergeron with a 61.1 SAT % last season. Boston and Los Angeles are two teams that have done really well in CF% as a team for the past couple of seasons.
Fenwick or unblocked shot attempts (USAT) by the NHL is a variant of Corsi but does not count blocked shots. This is seen by some to have a stronger correlation to scoring. Shooting percentage is the goals a player has over their shots. The league average for shooting percentage is .09. So when a player has a shooting percentage of 20, it is very likely that player will regress. PDO, or SPSV% by the NHL, is the sum of a team’s shooting percentage and save percentage. The theory is that every player will ultimately regress towards a sum of 100. It is not sure when it will happen but it will, eventually.
Although analytics have been embraced more recently especially by hockey bloggers, it still isn’t a perfect indicator of how sustainable a player can be. For one, it doesn’t account for how vital some roles in hockey are. Kris Russell is the biggest example of this. If one were to just look at Kris Russell’s statistics, they would think that he brings nothing to the table. He has a CF % of 46.5, which is not good especially for defensemen. Having said that, Russell is known as the best shot blocker in the NHL. He led the league with 213 blocked shots. Shot blocking is a very important skill for defensemen considering they are doing the goalie’s job. Because he blocks shots so often, Edmonton is more likely to not possess the puck when he’s on the ice. But at the same time, this is a part of his game, and it has seemed to work, more or less.
Shooting percentage is a great indicator if they can be sustainable or not. But it also doesn’t account for playmakers who pass more than they shoot like Mikael Granlund or Joe Thornton. As for other stats, power play specialists are also hindered because usually these stats are counted by even strength minutes. A fair amount of leagues have blocked shots, assists, special team points, and hits as categories. These stats obviously have value, but they are not necessarily accounted for if you just go by advanced numbers.
Advanced statistics can help fantasy hockey owners in some way. They can give you insight on how to evaluate players and help predict what could happen. But it’s not a perfect science. If they were, there wouldn’t be any fun in sports. Analytics can’t figure out exactly when a player is going to regress, and there is a chance these players are for real. This is why it’s important to also watch these players play so you can make up your own mind.