Sunday, September 14, 2014

Edmonton Oilers Young Stars; by the numbers

Through 2 games

This post will compare players on the Edmonton Oilers Young Stars team through the team's first 2 games of the 2014 Young Stars tournament.

Relative to replacement, or above replacement speaks to how that player performed when compared to an average player playing the same position. Averages are calculated from the numbers produced by the players on the same team. To calculate the value "relative to a replacement" we simply subtract the average value from each player's individual value. Two games is an absolutely tiny sample size. As such, these graphs are only meant to represent how these players performed in these two games. They are not meant as a statement on these players' long-term potential.

Winning the analytics race (if anyone ever does) won't just be about having the smartest analyst, it will be about having the best data. Exclusive puck-possession data for individual players, incorporated into other available stats is the only route to a true competitive advantage. 

DATA-GENERATED PLAYER SCOUTING REPORTS
 
A data-generated scouting report could be used to either compliment a perception-based scouting report, or to contradict it. The important thing to remember though, is that there is no predisposition to do either. Data is data; it's not about bad, good, better or best.

Perception-based scouting alone can be limiting; even deceptive at times. As such, it requires a litmus test. That litmus test can be produced through data-generated scouting. Think of perception-based scouting as a way to establish what a player is capable of doing, while data-generated scouting tells us what the player is actually doing. DGS paints a bigger picture; allowing us to determine players' strengths and weaknesses based on a variety of data, and not just the plays we remember.

These two approaches to scouting are not in opposition. In fact, any truly responsible scout knows that they are completely reliant on each other.

SCORING-CHANCES PER-POSSESSION PLAY; RELATIVE TO REPLACEMENT

This graph represents the number of scoring-chances each player has contributed per-possession play. A possession play is defined as any play made while in possession of the puck; these include shots, passes, dekes, dump-outs, dump-ins, etc. A player receives credit for a scoring-chance only when he is directly involved in the play that produced the scoring-chance.



 
SCORING-CHANCES for PER-MINUTE PLAYED; RELATIVE TO A REPLACEMENT

This graph communicates the number of scoring-chances per-minute played each player has contributed. Individual results are compared to players who play the same position. Once again, a player receives credit for a scoring-chance only when he is directly involved in the play that produced the scoring-chance.

 


SCORING-CHANCES against PER-MINUTE PLAYED; RELATIVE TO A REPLACEMENT

This graph communicates the number of scoring-chances against per-minute played each player was directly responsible for. Individual results are compared to players who play the same position. A player receives credit for a scoring-chance against only when he is directly involved in the play that produced the scoring-chance. Within this graph, LOWER IS BETTER!




EVEN-STRENGTH RATING; RELATIVE TO A REPLACEMENT

This graph communicates each player's rating relative to a replacement. Players' ratings are calculated by subtracting the number of failed events each player produced per-minute from the number of successful plays they produced per-minute. How involved in the player a player is directly impacts their rating; the more a player is involved the higher the rating he can produce. Individual results are compared to players who play the same position. A list of all the events tracked within this system can be found here.

 


EVEN-STRENGTH TAKEAWAY TO TURNOVER RATIO; RELATIVE TO A REPLACEMENT

This graph communicates each player's even-strength takeaway to turnover ratio; relative to a replacement. Players' T-to-T ratios are calculated by dividing the number of takeaways a player produced by the number of turnovers they make. A turnover is defined as any failed event with possession. They include passes, dekes, shots, etc. A takeaway is defined as any successful attempt to remove puck-possession from the opposition; they include stick-checks, body-checks, blocked passes, blocked shots, etc.

 


EVEN-STRENGTH RATIO; RELATIVE TO A REPLACEMENT

This graph communicates each player's ratio relative to a replacement. Players' ratios are calculated by dividing the number of successful events each player produced by the number of failed plays. Unlike ratings, ice-time, or involvement has no impact on ratios. Individual results are compared to players who play the same position. A list of all the events tracked within this system can be found here.
 



OVERALL O-ZONE POSSESSION PERCENTAGE; RELATIVE TO A REPLACEMENT

This graph communicates how much "time" each player spends with possession of the puck in the offensive-zone; relative to a replacement. Expressed simply, this graph shows the percentage of each player's overall events that were spent with possession of the puck in the offensive-zone. The calculation is done by dividing the number of events a player contributed with possession of the puck in the offensive-zone by the total number of events (both offensive and defensive) they contributed in all 3 zones.

Teams who produce a high percentage of o-zone events with possession generally out-chance their opponents.

 

  
OVERALL D-ZONE DEFENDING PERCENTAGE; RELATIVE TO A REPLACEMENT

This graph communicates how much "time" each player spent defending in the defensive-zone; relative to a replacement. Expressed simply, this graph shows the percentage of each player's overall events that were spent attempting to remove possession from the opposition in the defensive-zone The calculation is done by dividing the number of events a player contributed while attempting to remove possession of the puck from the opposition in the defensive-zone by the total number of events (both offensive and defensive) they contributed in all 3 zones. Within this graph, LOWER IS BETTER!

Teams who produce a high percentage of d-zone events while defending are generally out-chanced by the opposition.

 
  
OFFENSE-DRIVING PLAYERS PER-60; RELATIVE TO A REPLACEMENT  
 
One of the main objectives in hockey analytics is to establish a direct link between specific events that occur on the ice and scoring-chances. Obviously, the further away from the opponents' net (outlet pass) an offense-driving play occurs the less likely the play is to result in a scoring-chance. Similarly, the closer the event is to the opponents' net (pass to the slot), the more likely it is to result in a scoring-chance. Not surprisingly, the players who produced the most offense-driving plays per-60, were also among those players who produced the most scoring-chances per-60.

Only plays that push the puck up-ice while maintaining possession, or events in the offensive-zone that directly result in a shot on net are counted  as offense-driving plays.  A list of all the events I track can be found here.

Successful offensive-zone events included are: Offensive-zone passes off of the rush, offensive-zone east/west passes, offensive-zone passes to the slot, offensive-zone loose-puck recoveries (off of dumps, broken plays, and rebounds), shots, and deflections

Successful neutral-zone events included are: Neutral-zone open-ice dekes, east/west passes, and north/south passes, along with redline and offensive-blueline controlled-carries 

Successful defensive-zone events included are: Outlet passes, stretch passes and defensive-blueline carries.