Unconventional Wisdom - Relatively Speaking

Jamie Sabau

We are going to take a look at something different this week, as we compare relative values that we don't normally consider.

The Pittsburgh Penguins season draws to a close this weekend, so we are back again to take a look at the players' performance in various categories of fancystats. I recently had a hard drive die on me, so I lost my normal spreadsheets of accumulated data I had been working on throughout the season, and as such having to start from scratch decided to try something new.

In order to cater to the visually inclined I tossed the data into a bubble chart. I ran the numbers for and included all 37 skaters that have played for the team this season, however some of those played a small enough number of games that we have to worry about sample size issues. So players whose names appear in red are those that played less than 25% of the season, which means those that cannot reach 20 games played after this weekend. Even then there is some concern for that being a small sample, so players that fell below the NHL's standard 25 game cutoff have their names in orange.

5-on-5 Close Shots For Percent

Having recently examined the Penguins season by looking at possession and score effects, I decided this time to look at 5-on-5 Close Shots For and Shots Against, something I normally overlook in favor of the newer possession metrics. In order to determine how they are performing compared to their teammates I went ahead and calculated the relative values, which is on-ice values minus off-ice values.

The X-axis is Shots For Relative, the players' offensive contributions. The further to the right, those with positive values, are players that the team sees more shots on goal with them than they do without them, whereas the negative values on the left are those who have less shots on goal when they are on the ice.

The Y-axis is Shots Against Relative, the players' defensive contributions. The lower down the player is, down in the negative values, the fewer shots against the team faces while they are on the ice, whereas those with positive values higher up the chart are players that the team faces more shots with them than without them.

The bubble size is Shots For Percent Relative. Players with a yellow bubble have positive results, the team out shooting their opponents when those players are on the ice, while the white bubbles are negative numbers which indicate players that see the team getting out shot while they are on the ice.

Overall, players in the bottom right quadrant are the most effective at driving possession as they both create and prevent more shots while they are on the ice. The players in the top right quadrant are solid offensive contributors who are perhaps a bit of a defensive liability while those in the bottom left quadrant are solid defensive contributors who are somewhat lacking in offensive skill. That means the players in the top left quadrant as the least effective possession players as they both create less shots and face more shots against while they are on the ice.

Some of these players got cut off, as I opted to zoom in to focus one those players that have a significant sample size available. For the SF% chart this includes 7 players. Drazenovic is a large yellow bubble below and slightly to the left of Maatta and Niskanen. D`Agostini was a large yellow bubble below Dumoulin. Zolnierczyk is the small white bubble below Adams that you can just see the edge of. Sill is a large white bubble far to the left of Glass. Payerl and Goc are large white bubbles higher to the left of Jeffrey. Leblond is an enormous white bubble in the extreme upper left.

5-on-5 Goals For Percent

Using Goals isn't really a strange thing, but it is important to look to see if the players who did well based on shots are also still performing well when it comes to finding the back of the net. I had intended to look at overall goals, but when plugging in those numbers I saw that the highly successful PP and PK has skewed the overall for and against numbers. So while it is preferential for determining who wins the game, it doesn't work quite as well for comparing individual players. Close, on the other hand, cut us down to a bit too small of a sample size, even if it would be preferable when comparing how players are used without skewing the data with situational changes. So we are left with the usual 5-on-5 data.

The X-axis is Goals For Relative, so positive values on the right are those that see the team score more often with them on the ice, while those with negative values on the left are those that see the team score less often with them on the ice. The Y-axis is Goals Against Relative, so negative values lower down are those that cause the opponent to score less often whereas those this positive values higher up allow more goals when they are on the ice. The bubble size is Goals For Percent Relative, so those with yellow bubbles see the team outscore their opponent when they are on the ice whereas those in white are getting outscored.

Again we see the most effective players are those in the bottom right quadrant, as they are capable of both scoring more often and allowing less goals against when they are on the ice. The players in the top right then are those that are offensive contributors but perhaps a bit deficient in their own end, whereas players in the bottom left are better defensive contributors but are lacking in offensive skill. The players in the top left, however, are those that may be a cause for concern as they both score less often and allow more goals against when they are on the ice.

This time in GF% we have 5 players whose bubble got cut off as we zoomed in on our core options. Sill is the large white circle that you can just make out the edge of to the left of D`Agostini. Samuelsson is a large white bubble further down and to the left of Sill, while Leblond is a large white bubble lower than Samuelsson. Payerl is a large white bubble to the left of Pyatt and Ebbett, while Drazenovic is up above Payerl.

Overall "PDO"

Because the overall performance is important to the team's success, I wanted to find a way to look at it without being overly skewed by situational effects of PP and PK time. So I figured looking at a derived percentage value would balance it out. This isn't PDO in the sense that most people refer to it, but it is derived in the same manner, Sh% plus Sv%. However, I wanted to still be able to utilize the possession metrics, so we are going with my fallback of Fenwick Shot Percent and Corsi Save Percent.

The X-axis is Fenwick Shot Percent Relative, with the players that have positive values further to the right seeing a high Sh% when they are on the ice whereas the players with negative values to the left see a lower Sh% when they are on the ice. The Y-axis is Corsi Save Percent, with players higher up in the positive values seeing an increased Sv% when they are on the ice whereas those below in the negatives have a lower Sv% when they are on the ice. The bubble size is PDO, with yellow bubbles being higher PDO values when they are on the ice whereas white bubbles see lower PDO values when they are on the ice.

This time the top right quadrant are those that are the most effective, as the teams sees both an increased Sh% and an increased Sv% when they are on the ice. Players in the bottom right quadrant are those that are successful offensively but a little bit less capable defensively, while those in the top left quadrant are successful in their own end but don't bring much to the table when it comes to scoring. The players in the bottom left quadrant are the ones to be concerned about, as they both score less often and allow more goals when they are on the ice.

One issue we wind up with is that Martin and Jokinen's bubble overlap so as to make it difficult to read, with Jokinen slightly further right and a tiny yellow bubble while Martin is slightly further down with a tiny white bubble. 8 players got cut off as we zoomed in to capture our core players. Dumoulin is a tiny yellow bubble below and to the left of Malkin. D`Agostini was a large white bubble much further to the left of Glass and Sutter. Sill is a large white bubble a little further up and to the left of D`Agostini. Payerl is a large white bubble much further left of D`Agostini. Samuelsson is a large white bubble up above Payerl while Leblond is above Samuelsson. Ebbett is a large white bubble to the left of Pyatt and Jeffrey. Drazenovic is a huge white bubble way in the bottom left corner.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join PensBurgh

You must be a member of PensBurgh to participate.

We have our own Community Guidelines at PensBurgh. You should read them.

Join PensBurgh

You must be a member of PensBurgh to participate.

We have our own Community Guidelines at PensBurgh. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9355_tracker