Monday, July 28, 2008

Maltby And Draper Are Replaceable

The important thing that comes from hockey sabermetrics is the ability to draw conclusions about the game. Sometimes these conclusions are not widely held by the hockey establishment. I listed some of these conclusions in the thread my sabermetric philosophy. One of my conclusions was:

Kris Draper and Kirk Maltby are no longer as valuable players as most fans believe. Their even strength performance could be replaced by the Detroit Red Wings without too much trouble (and likely with a cost savings).

This led to a long argument in the comments with David Johnson of hockey analysis. When the discussion was completed, I believe we agree (for the most part). I have not laid out my entire argument to support this conclusion at any point in this blog, so I thought I would do it today.

One way to assess how successful a player is in their role with their team is by looking at their adjusted +/- rating. I showed two ways to do this: one as a rate stat and one as a counting stat. The methods have some fundamental differences, but for the most part tend to give consistent results that usually are consistent with what can be seen by watching the NHL games. I listed the 20 worst players for the rate stat adjustment. Kirk Maltby is 15th worst and Kris Draper is 16th worst. I also listed the 20 worst players for the counting stat adjustment. Here, Kirk Maltby is 8th worst and Kris Draper does not appear on the list (he is 37th worst in the NHL in the 2007/08 season). This is not a particularly good showing for either player. Maltby obviously has a worse showing. Both are on the ice for a good number of goals against and few goals for when compared to teammates. This might be explainable if they consistently faced extremely tough competition. However, behind the net shows that is not true. It is the frontline players Nicklas Lidstrom, Pavel Datsyuk, Henrik Zetterberg and Tomas Holmstrom who face the top quality of opposition (and all but Holmstrom are among the best adjusted +/- ratings.

When I look at the Detroit Red Wings, I see a team that is hard to improve. They are the defending Stanley Cup champions and favorites for next year. Players with poor +/- ratings would be good candidates to be replaced in an attempt to improve the team. This is especially true when these players are not big scorers. Last year, Kirk Maltby scored only 10 points (he has not topped 11 points in a year since before the lockout). Draper was a little better offensively. His totals fell to 17 points last year.

It is true that Draper and Maltby have reputations as good checkers. Kris Draper won the 2004 Selke Trophy. Draper and Maltby both represented Canada in the 2004 World Cup. That was yesterday. Maltby turns 36 this year. Draper is 37. They are well into the decline phases of their careers. A lot of fans are slow to recognize this decline because it is not easy to statistically show that their defence is in decline (although their offence is easily shown by looking at their point totals).

Draper and Maltby earn almost $2.5 million between them for this work. Its not huge salary money, but in a capped NHL where every bit counts, it would be nice to get more performance for this money. It isn't too hard to do this. There are many examples of players who made bigger contributions than either of these two who earn around their salaries (or less). There are players out there who could do this. It wouldn't be that difficult for Detroit to have found some this summer, had that been their desire. Worst case, they would get players who don't score and put up poor adjusted +/- ratings against moderate opposition but cost less. That savings could be used to upgrade elsewhere. There is a good chance, if the Red Wings select well, that they could find players who make a much bigger contribution than Draper and Maltby.

All of this is not saying that Kris Draper and Kirk Maltby are pathetic NHL players. However, they failed in their roles last year. They were by no means the biggest failures in the NHL, but they did not do well. Given their age, it is unlikely that will change. This would be a direction I would be considering if I was in Ken Holland's position as the Detroit Red Wings GM.

That said I understand why he isn't doing it. Detroit is a place that NHL players want to play. This is shown, in part, by their signing of Marian Hossa this summer for below expected market cost. They go this way in part by winning and in part by being loyal to their players who have been loyal to the franchise. As an example, last year they brought back Darren McCarty in his comeback despite the fact he had little value. Kirk Malty and Kris Draper have been very loyal to the Red Wings and the Wings are loyal to them. This loyalty helps to get them free agents at below market cost. Players know they will be treated well in Detroit. Accepting subpar performances from Draper and Maltby (who can always warm the press box if necessary) is worth it to be able to have increased ability to sign free agents at below market costs.

Detroit has players like Darren Helm and Ville Leino in the system who might take their jobs. They could have added another low priced free agent or two to challenge them.

When a player plays a 3rd or 4th line role on a team and puts up a poor point total and a adjusted +/- rating against moderate (but not strong) quality of opposition, they are expendable. Given their ages, it is unlikely that Draper or Maltby will bounce back in the future.

Naysayers would argue against the adjusted +/- stat. The players are being compared to other really good players on a really good team and thus come out looking bad by comparison. How would they look on a worse team, where they are compared to worse players? For the most part, they wouldn't likely look any better. It is possible that their adjusted +/- would be better in the rate case, but more than anything that shows that the rate adjustment is not so good at finding weak performances on teams where the off ice +/- rating is always poor. In the end it doesn't matter much. There are many examples of non-star players on good teams who put up much better adjusted +/- ratings then Draper or Maltby in similar roles (and often with more points scored). This would improve Detroit on the ice to acquire one (neglecting more intangible things like leadership).

Kirk Maltby is the more replaceable of the two. He brings less to the Wings than Kris Draper does, but neither are strengths for their team when they are on the ice. I imagine the Red Wings know this and are planning a further phase out of these two this season. Neither of them brings enough to the team to provide a contribution that is hard to replace on the ice. Given their age, this should continue to get worse over time. Given the fact they make more than the minimum salary (Draper especially) for minimal contribution, replacing them would free up salary space for the Red Wings to use in the future. Done properly, replacing Kris Draper and Kirk Maltby this season would be a very good move for the Detroit Red Wings.

Comments:
You make a decent case that Maltby and Draper had relatively poor seasons in terms of +/-, but what about their work as penalty killers? That, after all, is their primary job on the Red Wings. If they do well on the PK but don't score shorthanded, that doesn't impact +/-.

And considering their combined salary of $2.5 million, that means they're actually making roughly 50% of the average player on a salary-cap maxed roster. Taking that into account, it's hard to call them overpaid.
 
Shorthanded they were nothing special. I will look into that in the future.

Its true they were not badly overpaid, but if I look at the Detroit roster, they are the two people I would be looking at as not providing bang for the buck last year.
 
ummm...faceoffs?

draper earns his check and more in the circle. i believe he was second in number of faceoffs on the wings, and first in percentage with somewhere near 63%.

invaluable to a possession team.

maltby doesn't do as much, but he doesnt make much.

fuck this noise.
 
It is true that Draper won about 58% of faceoffs last year (lower than the number you quote). 58% is a good number that makes him one of the better faceoff men in the league (where else is 58% something to be proud of?).

The problem is even with these faceoff wins, the Draper line puts up a poor +/-. Just how important are faceoffs anyway? Here is something I once wrote on this topic. There is a pretty poor correlation between faceoff records and winning hockey games. Draper may be good on faceoffs, but that's not really that valuable to the Wings.
 
Hartigan signed with Dinamo Riga, like 2 weeks ago. Draper is far more valuable than Maltby. It's all about face-offs (best in the playoffs) which our game starts with. He's worth the money right there, no matter what his plus/minus is.

Maltby needs to retire.
 
cowboy - that simply can't be true. If his team is giving up goals by the bucketfull while he's on the ice despite his faceoff success, then we have to conclude that his "worth the money" skill is not, in fact, worth much of anything. It is not creating goals nor preventing opposing goals.

It would be interesting to see how much of that +/- comes in situations where Draper has taken a face-off vs. times when he changes on the fly (or is tossed from the dot) and a goal is scored. It may be possible to prove that the team does better in those limited situations, and then shape his play such that he only comes on to take important faceoffs, and changes out at the earliest opportunity.

My suspicion is that it won't really change his overall numbers all that much. How often are goals scored immediately off face-off wins? It happens but it's not common. How often does a team win a faceoff and it leads to a lot of offensive pressure? More often, but of course many of those situations don't result in goals, so they won't show up in plus-minus.

The biggest effect of these things is in the perception that a lot of pressure eventually wears down an opponent. You may not score on the power play, but you spent the whole time on the attack - and then two minutes later you score. The players on the ice then get the plus - the power play guys get no statistical recognition. For that matter, other than the goal and assists, they don't get a plus if they DO convert a chance.

I've been following this sabermetric discussion the whole way, just lurking. I'm cooking up some thoughts that I'll probably share when they're done. My problem is that I don't have a lot of time to do the research to see whether my formulae describe real results or not.
 
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?