Scoring Chances: A Season in Review Part 1

The 2013-2014 Flyers Scoring Chance Study is complete!

Some minor housekeeping to get through before we dig into the data. Recall how a scoring chance is defined.


A scoring chance is any play directed at the net from within the scoring chance area. The scoring chance area is roughly defined as the homeplate shape shown above. Shots from outside of this area are not counted as scoring chances. There are some exceptions for screen plays or fancy puck movement, but this is the general rule of thumb. Chances do include missed shots, but do not include blocked shot attempts.

From Robert P of Jewels from the Crown

77% of all goals scored over the last 3 years have come from the scoring chance area. (Between the faceoff dots)
43% of shots come from that same area.
League average shooting % on shots from within homeplate is 13.3%

This definition of scoring chances is fairly successful at capturing shot quality.

On Ice Chance Differentials

A player is awarded a +1 for being on the ice for a chance for, and a -1 for being on ice for a chance against. These two numbers are combined to create chance +/-. This functions like like traditional goal +/- except with a much wider scale of events. Chance +/- is a good indicator if a player has had a net positive effect on the team's play.

Forwards(Even Strength):


If you've been following along all season, these numbers shouldn't surprise you. Claude Giroux, Jake Voracek, and Scott Hartnell are the main drivers of possession and scoring at even strength amongst the Flyers forwards. What's really worth noting here is that Jake Voracek is in a class all his own, with a staggering +79 mark on the season. He more than doubles his nearest competitor, Claude Giroux, and holds a 4% lead in Chances For%. That's kind of insane. Jake is an elite driver of on-ice results relative to his teammates.

Another surprising highlight this year has been the play of Michael Raffl. The 25 year old Austrian rookie has put up Top-6 caliber numbers and fans should be excited to see this kid get an expanded role moving forward. Berube has gone on record stating that Raffl will be a scorer in the NHL, and it looks like his proclamation isn't that outlandish. Waffles deserves a long look on the Flyers 2nd scoring line with Brayden Schenn and Wayne Simmonds.

As I said a few weeks ago, Vinny Lecavalier earned his demotion to the 4th line with his poor play.

Defense(Even Strength):


At 38 years old, Kimmo Timonen is still the Flyers clear cut number one d-man. He's been remarkably successful despite some of the obvious effects of age on his game. And thanks to a really strong second half of the season, Mark Streit looks to be earning his contract as the Flyers 2nd pairing puck mover. The story isn't as sweet for the rest of the Flyers defense. Coburn has posted very average numbers. Luke Schenn and Grossmann have been clear cut negative influences on the ice. Andrew MacDonald has been a gift to the other team's forwards in his first 19 games. Opposing teams are able to control over 60% of the chance battle with MacD on the ice.

Assessing Defensive vs. Offensive Talents

No two players achieve their chance differentials the same way. Some players are net positive influences on the ice because they suppress chances against, others because they generate a tremendous volume of offense, and some through a combination of both talents. We can drill down into the chance data and suss out these talents by looking at rate states. Different players get different amounts of ice-time. Creating stats per 60 minutes of ice time allows us to directly compare every player on the roster, regardless of their time on ice per game.

Sorting by Chances For Per 60

Chances For/60 should give us a reasonable approximation of the most offensively successful players on the roster.



Sorting by Chances Against/60

Chances Against/60 will approximate defensive talent. The lower the number, the better.



Retreading Old Ground

Roughly 2 years ago, our own Eric T. found a very strong correlation between scoring chances and fenwick, or unblocked shot attempts. We can look to see if that relationship has held true for the 2013-2014 Flyers will relative ease.

A correlation of .84 is a very strong statistical relationship and in lock step with Eric's findings.

Across those 18 teams, the correlation between a player's shot differential and his scoring chance differential is about 0.80-0.85 (depending on whether you set the cutoff for games played at something like 20 or something like 70).

Despite Eric's findings, the shot quality argument against using shot differentials to evaluate players still persists. The reality is that players who drive possession and shot differentials will ultimately drive scoring chances.

Individual Stats

In addition to tracking on ice +.-, I also devoted time to tracking direct involvement in the creation of scoring chance. This was accomplished by recording set ups and chance attempts. A set up is awarded to a player for any play that sets up a scoring chance. Think of these as potential primary assists. Includes passes, rebounds, etc. A chance is awarded to a player for physically directing the puck at the net from the scoring chance area. Think of these as potential goals. Includes shots, deflections, etc. The sum of all of these set-ups and chances is used to created the stat INV, or involved.

Forwards(Even Strength):

Defense(Even Strength):

As it turns out, set ups and chances do a reasonable job differentiating between shooters and set up guys. If you go down the list, most of the the guys we conventionally think of as shooters(Read, Lecavalier, Hartnell) get a high percentage of their scoring chances from shots. And passers(Voracek, Giroux) get a higher percentage of their chances from set ups. Perhaps most importantly, there is a strong statistical correlation with a players INV/60 and point production.

Once again, we've reinforced the conclusion that for individual players, possession is strong component of driving scoring chances, and ultimately point production at even strength. These concepts aren't new or ground-breaking but they are still met with a great deal of resistance in the wide world of analyzing sports. My hope is that I've provided yet another reference for folks who might be on the fence about data driven player evaluations.

Lets take a break!

There's still much more to come. In part 2, I'll be taking a closer look at special teams. A quick thank you to the BSH regulars who have encouraged me to continue with this project and see it through to the end. Your support was greatly appreciated.

Please don't be afraid to leave questions, comments, and critiques!!!

This item was written by a member of this community and is not necessarily endorsed by <em>Broad Street Hockey</em>.