Draft Schmaft: From draft pick to NHLer

(Since it is draft day I thought I’d bump this article I wrote a couple weeks ago back up to the top of the stack.)

With the Carolina Hurricanes on the verge of winning the Stanley Cup (I think they will do it in game 5), it is time to start looking towards the off season and all the interesting stuff that it brings. One of the first things to happen after the Stanley Cup is awarded is the NHL draft set to be held in Vancouver on June 24th. You can find lots of information about the potential draft picks and what not elsewhere but as usual you will find something different here.

Back when Cliff Fletcher was GM of the Toronto Maple Leafs he traded away several draft picks which drew criticism from the Toronto Hockey Media which led to his famous “Draft Schmaft” comment and implying how he seemingly thought that draft picks were somewhat over rated and are just another commodity to be used however they can best be used to improve a team. So, how valuable are those draft picks? That was what I set out to find out.

To evaluate the draft I looked at all players drafted over the 8 year period from 1988 and 1995. Thanks to hockeydb.com, it was fairly easy to track down this information. I then ranked every player from those drafts based on how their NHL careers turned out and grouped them into four categories.

Non-NHLer – Any player who didn’t play in the NHL or only played a small number of NHL games (usually <50).

Depth/Utility player – This group contained players either only played for a couple seasons or maybe played for a longer time but in a very specialized role (i.e. the goon) but didn’t contribute a well rounded game as well as career backup goalies.

NHL Regular – This category consists of second and third liners, regular defensemen but not all-stars, and second tier starting goalies.

First Liner – This category contains the better players in the NHL including most first line players, top 2 defensemen and the better goalies in the NHL. These are the most valuable players in the league and are frequent NHL all-stars.

In total there were 2088 players drafted during the 8 year period, 1479 (70.8%) turned out to be non-NHLers, 306 (14.7%) turned out to be NHL Depth/Utility players, 229 (11.0%) were NHL regulars and 74 (3.5%) ended up being NHL First Liners.

The next thing I did was look at the chances of drafting an NHL player based on the position in the draft they were drafted in. I grouped players according the these draft positions: 1-5, 6-10, 11-20, 21-40, 41-80, 81-140 and 141+. The chart below shows the percent of players drafted who developed into NHL players according to these categories.

NHL Draft Analysis

As you can see, there is a fairly steady decline in the odds of drafting a NHL calibre player the deeper you go in the draft. Beyond the early second round (>40th overall) the odds of drafting an NHL First line player are quite slim and even having a draft pick develop into an NHL regular is quite poor (~15% chance at best). The flip side of this is that top 5 picks are almost certainly going to be NHL Regulars or better and top 10 picks do very well as well. Beyond that things get much less predictable as even mid-first round picks (position 11-20) have ~25% chance of being a complete dud and only about a 40% chance of becoming an NHL regular or better.

What else was interesting was that I didn’t see any significant difference between forwards, defensemen or goalies in their chances of developing into NHL players which means for the most part NHL teams have been good at correctly evaluating each position in relation to the other positions (i.e. there is no bias towards drafting forwards over defensemen).

The moral of the story is that when draft day comes around and all the hockey analysis talk the talk about how great a player the Montreal Canadiens will draft 16th overall you need to remember that the odds are that he won’t end up being much more than an NHL 4th liner.

This article has 6 Comments

  1. Nice breakdown. I linked to your post on the Thrashers Blueland Blog. I assume that when you said “To evaluate the draft I looked at all players drafted over the 8 year period from 1998 and 1995” you meant from 1988 to 1995. Correct?

    Keep up the good work.

  2. What criteria did you use for ranking players as “NHL Regular” vs. “Utility player” vs. “First Liner”? Was is just your opinion?

    Without these criteria, your analysis is irrelevant.

    It’s also flawed in that conditions existed then that don’t exist now, namely the fall of the Soviet Union and the influx of Eastern Europeans. Thus, your numbers are skewed.

    Finally, given the changes to the CBA, your numbers are made further irrelevant in that the way that teams draft now is completely different than the way they drafted then, and thus your numbers cannot be predictive.

    In this regard, you were fine until you cited your analysis to say that chances are the player the Habs pick at #16 will be a 4th liner. Had you left that out of your analysis, your exercise would have been analytical as opposed to predictive.

    Your study was not predictive.

    And if it sounds like I have done my homework, I have. I wrote my thesis on the NHL Draft.

  3. The criteria for categorizing players is somewhat based on statistics such as games played, goals, etc. and somewhat based on my opinion. The rating system is not perfect and I am sure that some of the players I have rated as a utility player should be a regular and vice versa but I can assure you that for the most part the first liner is better than the regular who is better than the utility player.

    You make an interesting point about eastern european countries but one might argue that with the increased openess that a teams knowledge of soviet and european players is probaby greater now and that makes late round surprises less likely than in the early 1990’s. Late round picks like Berezen, Titov, Timonen, Demitra, Vokoun, Konstantinov, Nabokov, Malakhov, Ulanov, etc. might have been drafted earlier.

    You talk about teams drafting differently now under the new CBA. What do you mean by differently and do you have any evidence that this is the case? I’ll assume that maybe they will draft more for skill and less for size and maybe this is the case but I have a hard time seeing that making a huge difference. Will the small skilled player who used to be a 3rd round pick now be a first round pick? No, I doubt the differences are that dramatic. Maybe the small skilled forward moves from a 3rd round pick to a second round pick but in the way I categorized players this may not have even changed which group the player was grouped in.

    But yes, ultimately my study is not a predictive study but if you assume that things haven’t changed all that much then my results can be used in a predictive nature. History is often the best predictor of the future.

    Care to share what you discovered in your thesis?

  4. Pretty much confirms the obvious. I think that if you placed point values on the respective picks, it would come out looking basically like the NFL Draft Value Chart that guides draft pick trades.

  5. Very nice work up.
    Yes, ‘maybe’ it states the obvious.
    But it may be stating an obvious conclusion about GENERAL MANAGERS as well? ie… first and second rounders get a better look, better assignments and better coaching, and quicker chance to get brought up and perform than do the lower picks BECAUSE the GM’s and scouts have a lot of FACE to protect. Self-fulfilling prophecy is another way to put it.

Comments are closed.