An evaluation of the Oilers roster with ESTR?

Puck Prospectus -- ESTR?
Puck Prospectus -- ESTR?

For those who love the game of hockey, and are fans of the Edmonton Oilers, in particular, we’re constantly searching for various statistical metrics for evaluating our players.  In most cases, the ‘measure of a man’ is how well a player performs at even strength or ‘EVs’.  Currently, there are various statistical methods for getting an idea of how well a player actually does perform at even strength, though there are often flaws and limitations with each method.

Conventional methods for evaluating how well a player performs at EVs often are not limited to, but include (Goals For-Goals Against), plus/minus rating, or Corsi numbers.  Corsi numbers are an interesting stat.  The Corsi number is generally a measure of the number of shots directed towards the net while the player is on the ice (typically, the number directed towards opposing team’s net – the number directed to your own net).  Since there are far more shots in a hockey game than actual goals, the Corsi number has the potential of being a more accurate than a simple +/- number due to the larger sample size of data.  Typically, the Corsi number is 4x more accurate than the +/- number.  However, one of the limitations of the Corsi number is that it can be skewed by where a particular player starts on the ice.  For example, if a player often starts in his own defensive zone, it will adversely affect his Corsi number.

Recently, while reading the excellent blog, Puck Prospectus, I came across a new and interesting ‘advanced stat’ to measure player value at even strength.  Intrigued by this new metric, I emailed Timo Seppa (the author of that article) asking him if he could kindly provide me with this data for the Oilers roster players for last season.  Timo promptly send me a reply with the data and for that, I am very appreciative (thanks again, Timo).  I will now do my best to present this data.

As Timo said, “Say hello to ESTR”  (I wonder if he got that line from Scarface?).

Even Strength Total Rating (ESTR) is a more advanced version of Goal Difference per 60 minutes (GD/60), taking into account the quality of teammates and quality of opposition for all non-empty net Even Strength goals scored while a player is on the ice. ESTR is a sum of its components, Even Strength Offensive Rating (ESOR) and Even Strength Defensive Rating (ESDR), which are Goals For per 60 minutes (GF/60) above average and Goals Against per 60 minutes (GA/60) above average, when adjusted for teammates and opposition. To get ESTR and its components for each player, goal difficulties for all goals over the course of a regular season (or postseason) are calculated, weighted by the base GF/60 and GA/60 of the players on the ice. For each player’s GF and GA, the average goal difficulty the player was on-ice for is then applied against each player’s base GF/60 and GA/60 to get their ESOR, ESDR and ESTR.

In short, ESTR is the number of goals per 60 minutes of ESTOI that the player is worth at Even Strength, given NHL-average players as teammates and opposition.

  • Essentially, higher numbers of ESTR, ESOR, and ESDR reflect that a player is comparably more effective than an average player offensively (ESOR), defensively (ESDR), and overall (ESTR) at even strength.

To provide context for these ESTR numbers, the first table includes the top 3 ranked players for the 2008-09 season based on ESTR along with some other notable players

Table 1 (click link for more data)

2008-09 Top Ranked Players by ESTR***For players who played more than 500 min of even strength ice time (ESTOI)

Table 2 —  Ranking of the Edmonton Oilers roster for 2008-09 by ESTR

Ranking of '08/'09 Oilers by ESTR
Ranking of '08/'09 Oilers by ESTR

My analysis:

Well, in comparison to the first table, we don’t currently have a player that ranks in the top 20 in the league at even-strength with respect to the ESTR ranking.  However, what’s interesting, is that Dustin Penner at +0.93, based on this metric, is not only the best even strength player on our roster, but he’s not too far off of the surprising Rene Bourque at +1.18 (rank 20th).  Secondly, I would be remiss if I didn’t include Timo’s comment in the email which was that for all of our players who played less than 500 minutes (ESTOI), they were all negative.  Timo remarked that this reflected good coaching (MacT).

Consistent with the opinions of the Oiligosphere, the Edmonton Oilers current roster of players does not have a wealth of players who are particularly strong at even strength.   Last season, we only had 10 players on our roster who kept their head above water at even strength based on ESTR.  Four of our top ten players at ESTR were defensemen.  Based on this data, it’s understandable how much Lubo’s injury hurt our playoff chances.  I wrote a previous post on Lubomir Visnovksy which is consistent with the impact of his even-strength play on Hemmer’s game.

Much like Dustin Penner, Robert Nilsson is subject to plenty of criticism regarding his level of play.  Also like Penner, his boxcar numbers are less than inspiring (64gp, 9-20-29).  Admittedly, lately I’ve been among the masses who were willing to send Nilsson and his $2million cap hit packing.  In contrast to my own sentiments and many among the Oiligosphere, this data suggests that Robert Nilsson is a valuable player at even strength.

If you’re an Oilers’ fan, chances are that you love Sam Gagner and have high hopes for what he can bring to this team.  I myself was surprised to see Gagner ranking in the top 5 on our current roster of players based on ESTR.  Based on ESTR again, it’s perhaps less surprising that we’ve let go of both Kotalik as well as Brodziak.

Again, special thanks to Timo Seppa over at Puck Prospectus for providing me with this data.  To see more articles by Timo, click the link.  As well, Timo has his own blog site, Ice Hockey Metrics, that you can visit by clicking the link on that text.

***Update, Timo’s written his own blog on this topic:  Striking Oil:  Edmonton’s Best.



7 thoughts on “An evaluation of the Oilers roster with ESTR?

  1. From this data, it appears that our kids could swim last year! Cogliano, Gagner, and Nilsson all held their respective heads above water by maintaining a positive ESTR ranking. Moreau and Staios both performed quite poorly especially for their respective cap hits of ($2 million and $2.7 million respectively).

  2. Great Post Greg aka Hockey Noob, lol! It’s interesting to see that Pouliot ranked higher than Brodziak at ESTR given that he’s looking like the guy to take the 3C position and role this up coming year. Good find on this data!

  3. ACDC B… That reminds me of how I used to write multiple choice exams in school… lol. Well, not quite. Well, I think we’ll miss Brodziak with respect to the number of defensive zone draws that he took last year. I still firmly believe as do many that Malhotra would be a great player to slot into the 3C position. From the ESTR data, it seems to indicate that Brodiak was better defensively than Pouliot, but his total ESTR number was lower more as a result of his lower ESOR number.

  4. Very Interesting post. Unfortunatly statistics in hockey still don’t give a good assessment of a player’s worth. This is not baseball. It only goes to emphasize that these stats don’t show so much of a player’s contribution on the ice as what situations the coach used them in. “3rd liners” or “checkers” as well as “stay-at-home” defensemen are used against other teams best lines and therefore face more offense, but are able to hold down the fort and limit high quality opportunities. Other “offensive specialists” play when the time to strike offensively is there. Look at it in a way to Lacrosse. Not nearly as obvious but the match-ups are still there. I defy any Oil fan who watched more than 40 games last year to say they didn’t cringe when watching the likes of Penner and Nilsson hemmed in the defensive zone by a opponants top line. One of the few exceptions to this coaching bias would be Horcoff who can (and was) played in any and all crutial situations, and by the stats provided he faired quite well.

  5. JMC–Some very good thoughts… I agree with what you’re saying in that there’s certainly going to be a coaching bias reflected in a stat such as ESTR. To clarify, how well a player is used in terms of the fit with his own linemates as well as the type of opposition that he plays against is certain to have a very significant impact on a player’s stats. Even if the metric tries to mathematically correct in some method for qualcomp/qualteam, there’s still an intangible element that can’t be factored in. For example, if Hesky was miscast on a checking line with say Moreau/Reddox, it’s not hard to imagine that his stats would be terrible.

    I also agree that hockey’s certainly a more challenging sport than baseball in terms of being able to measure a player by a single or multiple metrics as hockey is a lot more of complex.

    In short, it’s always good as you suggest to look at these type of stats in the context of having acually watched these players.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.