The NHL Entry Draft: Investment or Gamble?

In just over two weeks, NHL General Managers will attempt to convert the massive amounts of time, effort and money their organizations have spent on scouting, evaluating and courting the best of the current prospect pool into lucrative reality at the NHL Entry Draft in Pittsburgh.

The prospects themselves will be hanging on each pick, looking to cash in on their years of hard work, and fans will be equally expectant, hoping for that franchise-changing pick that will elevate their club to new heights.  Mock drafts have been proliferating for months, and as The Day grows closer, the fervor in the hockey world is ramping up to epic proportions.  When the draft unfolds, pundits will loquaciously expound on the virtues and vices of each selection.

Amid all this uproar, I wonder if it is all worth the trials, tribulations and hoopla that surrounds the draft.  Sure, we need some equitable system of bringing young talent into the NHL, since we long ago escaped the days when Montreal had the rights to seemingly every hockey prospect in North America locked up.  So, to that extent, the Entry Draft is an honorable institution, promoting fairness and industry among the NHL member clubs.  That being said – and considering the considerable investment that each club makes in each of the players selected — both directly in terms of salary and development cost, and indirectly in scouting and courting expenditures – how well does the system really work?   Does the Draft produce sufficient levels of quality to merit all of this activity and angst?

We can all point to spectacular failures, such as Boston selecting Lars Jonson with the #7 overall pick in 2000.  Jonson played just eight NHL games.  Similarly, there are some stunning successes, such as Nashville’s stealing of Pekka Rinne in the 8th round in 2004.  (Intriguingly, had Rinne been in the pool just one year later, he might not have been drafted at all, as the Draft was shortened to 7 rounds in 2005).  Still, you cannot properly judge the efficacy of the Draft as a whole by reference to these isolated extremes.  You need to look at the big picture.  Is the draft – and all that leads up to and follows it – truly an investment in a club’s future, or is it more akin to a venture into the casino, where hopes run high – but results seldom match?

Columbus Blue Jackets goalie Steve Mason (Dave Gainer/THW)

To answer this question, I looked at all 2,946 selections made from the 2000 Entry Draft (the first with the full complement of thirty NHL clubs) through the 2011 Draft.  However, in assessing what to look for, I had to consider what the ultimate goal of the draft is, and what common attributes among those drafted could be used to determine the extent to which those goals have been achieved.

You can’t use normal statistical measures – i.e. goals and assists, as those do not necessarily encompass the attributes of goalies or defensemen.  However, at the most fundamental level, what General Managers are looking for are players who can effectively compete at the NHL level.  Sure, they want stars who can score 80 goals, or post save percentages north of .950, but those are ephemeral qualities of “stardom” that can be fleeting.  Steve Mason was a “star” in 2008-2009, winning the Calder Trophy and leading the Columbus Blue Jackets to their first playoff appearance.  Today?  Not so much.  No, the essential goal of any draft is to find players who can compete at the highest level over the long term.  That ability to compete is also not necessarily confined to playing for the club that drafts the player.  NHL players contribute in myriad ways to a team’s success.  They can put up numbers on the ice, provide leadership on and off the ice  — or they can provide return value in a trade.  Some organizations are stockpiled at particular positions.

Does that mean they avoid drafting that position?  Not on your life.   With development times begin lengthy for all but the highest performers, attempting to fill immediate needs through the draft is a chancy proposition (as we will see).  However, if you can nab the best available player, and then parlay that player into existing talent at your areas of need via trade – well, that’s what being a GM is all about.   The value of that selection is not minimized in the slightest by the fact that the draftee is now wearing another sweater.

To measure the rate of success, I used what I will call the Player Participation Index, or PPI.  While it sounds technical, it is really a simple comparison of the number of NHL regular season games played by each prospect, as a percentage of the total possible regular season games available since the time that player was drafted.  This accommodates players who have moved on to other clubs, via trade or free agency, while simultaneously allowing for consideration of circumstances that reduce NHL games played, and accordingly impact the quality of the selection (i.e. poor performance, injuries or defection to the KHL, among other things).

After segmenting that information into six “buckets”, based upon percentage of NHL games played (from 90% + down to zero), I looked at the numbers by year, by round, and drilled down further into individual selections within the first round.  The “by round” analysis stands out as the most salient to the issue of overall efficacy, as there are material trends between rounds, while year to year differences are more likely attributable to influences having nothing to do with the draft itself.

The one truly noteworthy anomaly in the draft year breakdown comes in the 2003 Entry Draft, when the moon and planets aligned to provide a rare Mother Lode of talent in the draft pool.  An amazing 17 players from that draft had PPI values of .750 or better, representing more than 25% of all players reaching that value over the 12 years examined.  The next closest total in any single year is six — achieved a few times.

Here is the overall view, for all rounds of the Draft (Nine rounds for years 2000 through 2004 and seven rounds since):

[table id=2 /]

Some fairly stunning numbers leap out from this view.  Almost 65% of players drafted since 2000 have zero time on NHL ice.  Slightly less than 13% have a PPI of over .250 for their teams.    A measly 2.27% of drafted players have PPI ratings of .750 or higher, meaning that from a General Manager’s perspective, it is slightly more likely that a straight up bet on a number at the roulette wheel will pay (odds 38/1) than a player appearing in more than 75% of games (a little over 44/1).

Now, you might say “Wait a minute – the development time for younger players hauls these numbers down.” That’s true, but only to a point.  If we limit the view to only draft years in 2007 or earlier, the numbers change only slightly.  Under that scenario just over 58% of drafted players have played no NHL games, while the other numbers, in ascending order, are  23.8%, 7.22%, 8%, 1.81% and 0.43%  – not big differences.

Of course, individual team situations also impact the number . . . but not the validity of the analysis.  Obviously, teams with the lowest draft picks have fared relatively poorly on the ice, compared to their competition.  Thus, these clubs have more holes to fill, and have the pick of the crop, significantly magnifying the likelihood that a draft pick for those teams will see meaningful NHL ice time relatively early.   The Stanley Cup Champion team is unlikely to have too many gaps to fill, so the chances that a new draftee will see material playing time in the near future are relatively slim.  However, this does not undermine that overall scheme of things.  For the championship club, this player either needs to be good enough to crack the lineup, or good enough to

To be sure, the biggest differences are to be found between the individual rounds selected.  Of the 19 players over 12 years who have played 90%+ of the possible games, 18 were first round picks, and the other was a second rounder.  Overall, only 20% of first round picks have failed to make it to the NHL.  Still, over the 12 period, just under 40% have played half or more of the available games.  The statistics fall off a cliff after you pass the 1st round.  Only 19% of second round picks have played more than 25% of available games .  By the time you reach Round 3, your chances of playing in half or more of available games (4.28%, or 1 in 23) are worse than getting a blackjack in the first two cards of a single deck (1 in 21).

Clearly, if you want to find success in the draft, with any sort of consistency, it needs to come in Round 1.  Sure, the periodic “diamond in the rough” can be found in later rounds –frequently with goaltending, as goalies tend to mature and blossom later than other position.  However, as in the casino, eventually the pure odds are going to catch up with you.  However, merely having a first round pick is no recipe for success, either.  Consider the following chart, showing the distribution of PPI among the top 15 picks in the 1st round over the last 12 years.

[table id=3 /]

Historically, if you have the #1 or #2 overall pick, you are going to fare extremely well.  Nine of the 12 first picks (75%) have PPI rankings of .750 or higher, and all but one have played at least half of the games.  The number two picks have also fared well, with 58% at a PPI of .750 or more, and 83% playing at least half.  Look what happens, however, when you get to the #5 selection – only 1 in 4 play 75 % of the games, and just over half exceed a PPI of .500.  Not bad, but far short of a sure thing.  By the time you reach the #15 pick, none have played 75% of games, only 8% have played half of the games, and a full 42% have not appeared on NHL ice.  And that’s not even considering the lower half of the first round.  Overall, just 52% of the top 15 selections have PPI values of .500 or greater.

Malkin and Crosby NHL: Dec 12 Panthers at PenguinsSuch numbers add emphasis to the importance of the good fortune enjoyed by clubs such as Pittsburgh, who had four consecutive years with either the #1 or #2 overall picks, which they converted into Crosby, Malkin, Fleury & Staal.  The Atlanta Thrashers had the benefit of two #2 picks and one #1 selection between 2000 and 2002, which brought them Heatley, Kovalchuk and Lehtonen, albeit with considerably poorer results than the Penguins. Time will tell how Edmonton handles similar good fortune, as they prepare to use their third consecutive #1 overall pick.

Conversely, there are clubs who have had the misfortune of underperforming enough to be “lottery teams”, but not bad enough to notch one of the “sure fire” first two selections.  The Columbus Blue Jackets fit this picture.  They have only had one playoff season, but similarly (prior to this year), only had one pick in the top three – which they converted into Rick Nash.  So, clubs like the Blue Jackets suffer from a kind of limbo — not good enough to consistently reach the post-season, but not bad enough to receive the kind of significant help that a #1 or #2 draft selection can bring.  Of course, a big part of the equation is how well teams utilize the picks they do have, and that topic will be the focus of next week’s installment.

The numbers appear pretty dire, but other forces also act to keep the numbers low.  There are only 690 spots on active NHL rosters, and new draftees arrive at the clip of about 210 per year.  So, to accommodate that growth rate, approximately one-third of each active roster would have to disappear each year.  That simply doesn’t happen, of course, and the excess find their way to the minors – mostly the AHL.  Certainly, keeping the farm clubs well stocked is a laudable goal of the draft, as the minors bring live hockey to communities that don’t have NHL clubs.   However, wherever the players are assigned, to argue that the minor league “stockpile” is going to materially fuel the NHL club is largely illusory at any given point in time, statistically speaking.  The fact remains that the vast majority of drafted players will never see time on an NHL team.  It would appear difficult to justify the seven rounds of draft activity currently employed, given these numbers.

Of course, there is a significant – and well taken – assertion that the NHL should not necessarily be the only measure by which success should be judged.  Canada does a much better job than the United States of embracing junior and minor league hockey as substantial entities in their own right.  Culturally, the U.S. has traditionally attached negative connotations to the term “minor league”, making it almost synonymous with “failure.”   That’s simply wrong, and more than one U.S. –based NHL club has struggled with balancing its development and backup needs with the independent needs of the minor league club to win games, draw fans and achieve its own level of success.

While fueling the AHL is, of course, part of the output of the Entry Draft, it is not a selling point.  The Draft is portrayed as the massive entry point to the grandeur of the NHL.  That’s true – for a very, very few select members of an elite class.  Less than 15% of the players drafted will ever have more than a passing familiarity with an NHL club.   So, given that a GM’s goal is to bring in consistent NHL-contributing players, the numbers say the goal is going to be missed 85% of the time.  In my book, that’s not an investment, that’s a gamble.  And a bad one at that.

One Comment

  1. Pingback: Divining A Solution: The Quest of the NHL Draft | Overtime

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>