What’s the best way to measure what New York Islanders player had the best game? Goals? Points? How about all of it. Introducing game score.
After every New York Islanders game, three stars are awarded out to players. Either they scored a boat load of goals, or they fulfilled some sort of milestone or narrative arch. Like scoring their first NHL goal, or blocking a certain goal from going in.
But as we all know, judging a player’s performance shouldn’t only be based on their production. Sure. Production is a huge part of it. Without goals, you don’t win. It’s as simple as that. But what players do to facilitate the creation of goals for their team should have similar weight in judging their effectiveness.
That’s where game score comes into play. It’s a stat “created” by the Athletics Dom Luszczyszyn. I say created because as he says as he introduces the concept back in July of 2016:
"The idea is not my own. Like most stats, the concept is completely ripped off from another sport where the question has already been answered.[…] Game Score is originally a baseball stat created by Bill James, but the basketball version was created by John Hollinger. It was meant as an extension and simpler alternative of player efficiency rating (PER) and aims to “give a rough measure of a player’s productivity for a single game”"
In Dom’s game score stats everything you can get from a regular box score plus a bit more is considered. Goals, assists (first and second), shots, blocked shots, penalty differential, both giveaways and takeaways, face wins and losses. As well as 5on5 Corsi differential and 5on5 goal differential. All of it is considered. What differs is the weight given to each individual stat.
Want your voice heard? Join the Eyes On Isles team!
To The Formulas!
Here are the two formulas that he came up with and that I’m going to use in establishing game score stats for individual New York Islanders players for the 2016-17 season.
This first one is for skaters:
"Player Game Score = (0.75 * G) + (0.7 * A1) + (0.55 * A2) + (0.075 * SOG) + (0.05 * BLK) + (0.15 * PD) – (0.15 * PT) + (0.01 * FOW) – (0.01 * FOL) + (0.05 * CF) – (0.05 * CA) + (0.15 * GF) – (0.15* GA)"
This, much simple one is for goalies:
"Goalie Game Score = (-0.75 * GA) + (0.1 * SV)"
As soon as I compile all the data I’ll release the other months, month-by-month. Which should form an interesting narrative as to who got hot as the season progressed.
Game score was made to be easily consumable. Scoring a ton of goals, or stopping a ton of shots will yield a high number. Since the 2007-08 season, Dom found that the highest game score was Sam Gagner at 7.1.
Since the 2007-08 season, Dom found that the highest game score was Sam Gagner at 7.1. That was thanks to an eight point night. The highest goalie game score was a 59 save shutout by Ben Scrivens in 2013-14 for a 5.9 score.
By the end of the experiment, I’ll roll it all up and graph to see just how Isles players impacted individual games from the start of the season to the end. It will be interesting to see if there was an overall uptick in individual production team wide, implying the system could be responsible for the success. Or did individual players just “wake up”.
This is going to be a fun experiment to conduct during the offseason. I’ll continue this during the regular season. Rather than rely simply on the eye test for who had “the best game” let’s add some analytics to it.