Advanced Stats Basics and Where the Flyers Rank

2 of 3

Corsi

Behold, the foundation of advanced stats. Corsi, in the simplest terms, is +/- but for shot attempts instead of goals. All shot attempts, not just shots on goal.

For the sake of round numbers let’s say the Flyers and Pittsburgh Penguins play a game in which Philadelphia attempts 55 shots and the Penguins take 45. In this scenario, the Flyers finished the game with a Corsi For Percentage(CF%) of 55%.

On a smaller scale, let’s say Sean Couturier was on the ice for 6 shots for and 4 against. This would give Couturier individually a CF% of 60%.

If you have ever seen or heard the term “play-driver” thrown around by the analytics community, this is what they are referring to. Players who help push play in the right direction to win the shot attempt battle. It helps to highlight just how impactful one specific player can be to a team’s performance.

In Sean Couturier’s Selke Trophy winning season he finished with a sparkling 56.25 CF%, this gives us an idea of what a high-end play-driver looks like in terms of Corsi results.

Sean Couturier, Philadelphia Flyers (Photo by Mitchell Leff/Getty Images) /

So why is winning the shot attempt battle so important? It’s pretty straightforward, more shots equal more chances to score goals. Corsi has also been found to be a great predictor of future results in the NHL. In 2018-19, of the 16 teams that advanced to the playoffs, 12 of them were also in the top half of the league in CF%.

So if your team is in a slump over a 10 game stretch, but is regularly winning the Corsi battle, odds are over the course of the next 10 games they will break their slump and post a positive goal differential (assuming they continue to drive play). Vice versa, if your team is piling up wins but regularly losing the Corsi battle, there’s a good chance there are cracks in the proverbial boat that will eventually sink it.

Expected Goals

By now you may have deduced that there is an inherent flaw with counting just shot attempts. Not all shots are created equal. A shot from the point isn’t as dangerous or impactful as say a shot from the middle of the slot. This leads us to our next stat, Expected Goals (xG).

Expected Goals helps account for the variable of shot quality, assigning a numerical value to each part of the ice to give weight to every shot taken. Let’s create another scenario.

This time the Flyers are going to lose the Corsi battle to the Penguins in this fake situation, getting out-attempted 60 to 40. By Corsi alone, the Penguins would be expected to dominate this game.

What actually happened in this faux case study was that the Penguins were held to the outside and took the majority of their shots from the point and half-walls, generating just 1.00 xG. The Flyers on the other hand created dangerous shots from the slot and net-front areas and generated 3.00 xG (4.00 xG generated in total between both teams).

Based on this info the Flyers exit this game with a 75 xG% and win the shot quality battle in a dominant fashion.

Alas, just as Corsi has its limitations so does Expected Goals. Without manual tracking, only the location the shot was taken from is accounted for in this metric, which could undersell the actual quality of a shot attempt.

For example, if a player takes a shot from the center of the slot xG may give that a weight of 0.1 xG. What that doesn’t account for is the type of shot taken or what kind of pass, if any, preceded it. If that shot from the slot was a one-timer preceded by a pass from below the goal line that could spike the value of that shot up to .25 xG, over twice as dangerous.

Thankfully the NHL is moving toward utilizing tracking technology to make this puck movement data more accessible and easier to accumulate, but in the meantime, xG does a good job of giving us an idea of if/when teams and players are winning the shot quality battle.

PDO

PDO is the analytics community’s closest way to quantifying “luck”. PDO is the numeric value of a team or player’s on-ice shooting percentage plus save percentage. Generally speaking, these two quantities add up to around 100.

Anything drastically above 100 suggests that there’s some good luck involved. Conversely, anything dramatically below 100 implies that a team has experienced some bad luck and should probably try to find themselves a couple of four-leaf clovers to turn their luck around.

Of course, good teams can hold a PDO over 100. Last season both the Colorado Avalanche and Boston Bruins both finished with season PDO’s over 102, far from a fluke considering the quality of those clubs.

On the other hand at the bottom of the PDO chart, the Detroit Red Wings with a PDO of 96.7. They earned that number with their poor play, but arguably the unluckiest team from 2019-20? The Vegas Golden Knights. They finished with a PDO of 98.9, yet still finished near the top of the Western Conference, suggesting that they could have been even better if some bounces had gone their way.

As for the Flyers, they finished last season with a PDO of 100.7 which suggests their strong play was not the result of flukiness or luck. They earned their record through strong underlying play.

Home/Analysis

Advanced Stats Basics and Flyers Ranks Thus Far

By Ryan Hodge | Feb 12, 2021

Corsi

Expected Goals

PDO