Breaking down Subban's performance based on scoring-chance numbers
The strength of data-generated scouting is not the mechanism that generates the overall numbers. Any attempt to produce one number to reflect a player's impact on the game cannot help but be contrived. The strength of the system can be found in the data that produces this number. The overall numbers can only point us toward the areas where the game-changing data lives.
Think of a data-generated scouting report as a pyramid. The top of the pyramid is the player's risk/reward rating, and ratio; while the body of the pyramid is made up of the puck-possession micro-stats that are tracked. One way of breaking down a player's performance is to compare the ratings they have produced when the team has out-chanced the opposition at even-strength with the numbers they generated when they were out-chanced.
The more involved Subban has been forced to be in the play, the less success the Habs have enjoyed. In games where the Habs have out-chanced their opponent at even-strength, PK Subban's even-strength risk/reward rating, and event totals are actually lower than they've been when the Habs out-chance the opposition. This is a regular occurrence with defensemen, as the more a team's D is forced to be involved in the play, the more likely the bulk of that play is occurring in the defensive-zone.
Where Subban's events have occurred has had a direct impact on the team's success. Breaking the numbers down by zone, we see that Subban's defensive-zone risk/reward rating has been substantially higher when the team is out-chanced, while his offensive, and neutral-zone risk/reward ratings are higher when the Habs out-chance the opposition. This makes complete sense, as the more often a team's top player is forced to play in the defensive-zone, the less likely that player is to help his team create scoring-chances.
The more Subban has possession of the puck in the offensive-zone, the more success the team enjoys. As mentioned Subban's offensive-zone risk/reward rating is substantially higher in games where the Habs produce more even-strength scoring-chances than the opposition. Looking at both offensive and defensive-events occurring in the offensive-zone, we see that Subban's defensive-events in the o-zone have little impact on scoring-chances. It's his puck-possession events that are substantially higher when the Habs out-chance the opposition.
Subban's specific offensive (or puck-possession events) that increase the most when the Habs out-chance the opposition are passes, and dekes. Subban's success-rates within these metrics actually don't improve when the Habs enjoy success; it's the actual number of successful attempts per-minute that increase.
Subban's involvement in the offensive-zone cycle is more important than his shot at even-strength. Subban produces fewer successful passes to the slot when the Habs enjoy scoring-chance success, the specific passing plays that show an increase include cycle passes, and east/west passes. In games where the Habs out-chance the opposition at ES, Subban completes an average of 0.121 cycle passes per-minute; while completing 0.101 when they are out-chanced. Subban also completes 0.076 successful east/west passes per-minute when the Habs out-chance the opposition, while completing 0.040 when they're out-chanced.
In complete contrast to the powerplay, the number of successful passes to the slot, or shots that Subban completes have not resulted in the Habs producing more scoring-chances than the opposition. In fact, Subban gets 0.038 shots through to the net per-minute, with a 38.6% shots-through success-rate when the Habs out-chance their opponent. He gets 0.055 shots through per-minute when they are out-chanced with a shots-through success-rate of 55%. This reflects more on Subban's position, than on his play. Shots from defensemen are generally taken from further out; as such, they are rarely scored as scoring-chances.
Subban has attempted and completed more dekes in the offensive-zone in games where the Habs have out-chanced the opposition. This includes both wall-dekes and open-ice dekes. He has produced 0.114 successful dekes per-minute when the Habs have had scoring-chance success, while completing only 0.065 otherwise.
Subban's play at the offensive blueline translates directly into scoring-chance opportunities for both the Habs and their opposition. The more blueline holds Subban attempts, the more scoring-chances the opposition gets, while the more often Subban carries the puck into the offensive-zone, the more scoring-chances the Habs produce.
When the Habs out-chance the opposition at even-strength, Subban carries the puck into the offensive-zone 0.061 times per-minute played, when the Habs are out-chanced those o-zone carry numbers drop to 0.030 carries per-minute.
Subban has attempted 0.030 blueline holds, with a success-rate of 60% in games where the Habs are out-chanced by their opposition at ES. When the Habs out-chance their opponents, Subban has attempted only 0.008 blueline holds per-minute, with a 100% success-rate.
Subban's defensive-play in the neutral-zone is directly linked to how well Habs forwards are back-checking. The stronger the back-pressure, the more aggressive PK can be in the neutral-zone. In the neutral-zone Subban has been much more active defensively in games where the Habs out-cahnce the opposition at even-strength. He produces 0.076 successful defensive-events when Montreal is successful, while generating only 0.025 successful defensive-events in the n-zone when the Habs are out-chanced.
Subban has a much more active stick in the neutral-zone when the Habs enjoy success. The specific neutral-zone events that Subban has increased when the Habs out-chance the opposition include stick-checks and blocked passes. He produced 0.030 successful stick-checks when they out-chance their opponent, while contributing 0.005 when they are out-chanced. In terms of blocked passes, Subban has 0.038 blocked passes per-minute when the Habs are successful, while contributing only 0.015 per-minute when the opposition has out-chanced them.
The strength of the system can be found in the micro-stats used to generate the player's rating. The most important information can be found by using these overall numbers to point to specific metrics that impact both player and team success.