Panning For Gold:The NHL Draft Analyzed (Part II)

In Part I of this feature, I examined the NHL Entry Draft (Rounds 1 -3) from 2000 through 2009, showing how ephemeral the process can be, how a relatively small percentage of players ultimately contribute significantly at the NHL level and how the PPI (Player Participation Index) can reflect the true efficacy of both individual selections and overall trends.

In this closing installment (after further reflection, I elected to combine the final two installments into a single article), I examine the characteristics of those players with the highest PPI’s, look at differences between positions, and finally examine drafting performance at the team level over the ten year period selected.

As should be apparent by the statistics cited in Part I, the draft process is in once sense inexact, but it another way, quite predictable.  While the overall percentage of significant contribution is relatively low, that percentage moves in a predictable fashion, decreasing from Round 1, through Rounds 2 and 3.  So, within the confines of each draft year, the organizations do a consistently good job of evaluating the available talent pool.

One interesting exercise is to look at the players with the highest PPI’s (90% +) over the period:
[table id=12 /]The mere fact that there are only 19 names on this list over a 10 year period shows how difficult the task is.  The stars are certainly present on the list — Ovechkin, Crosby, Kovalchuk, Nash — but a few names that you wouldn’t necessarily expect also appear — Meszaros for example.  As might be anticipated, there are a preponderance of younger players, as it is easier to maintain a 90% + PPI over one or two years, as opposed to the ten years Hartnell has done it. 

What the data shows is that the top PPI guys are going to be either 1) superstars; 2) on a team with a lot of holes that needs immediate help; or 3) a combination of both.   Nash and Kovalchuk would fall in the third category.  But how much of a “star” do you need?  A first round draft pick should be pretty much a lock, right?  Well, not so much, as it turns out.  Check out this table, showing the PPI for first round picks, by their selection position within the first round:
[table id=14 /]This is some amazing stuff, when you look closely.  There is a huge gap between having a top five pick, and having a pick anywhere else in the first round, with the PPI dropping from 64.62% for the top five to just over 39% for slots six through ten. (PPI moves up to 70.1 for the top three picks).  Equally interesting is the fact that the PPI stays within a very narrow range from picks six through 25, then takes another dip for the last five picks.  Blue Jackets’ President Mike Priest recently observed that his team “had not been bad enough”  to garner the superstars that come in the first few picks.  A definite mixed blessing . . .

What about positions?  Is there any truth to the fact that defensemen are slower to come around than forwards?  How do centers fare, as opposed to wingers?  What are the numbers for goalies?  Check out the following:
[table id=13 /] While there is some reduced efficacy for defenseman, the difference is not statistically significant.  Once again, what does stand out is the precipitous drop from Round One to Round Two, and a lesser drop from Round Two to Round Three.   The law of diminishing returns.  As the numbers show, drafting goalies is a highly perilous proposition.

Finally, let’s take a look at how teams have performed as a whole over the ten year period, as that is really the ultimate indication of how a drafting strategy works.   Here is the data:
[table id=11 /] There are a number of interesting observations to be made here.  First, the disparity between #1 (Minnesota) and #30 (Tampa Bay), is astounding, with Tampa having a paltry PPI of under 8%.   That appears to tie in with the poor play of late, but does not account, by itself, for their Stanley Cup run.  Similarly, Minnesota has not been able to match their lofty status in PPI to playoff success.  This simply shows that the draft alone is not a valid hallmark of ultimate success.

However, the data also shows that while the draft may not be dispositive, it is a big contributor for some clubs.   The clearest example here is Washington, with 38 overall draft picks and 16 first round picks.  Their first round selections in 2004 — Alexander Ovechkin and Mike Green — would represent several years worth of draft success for some organizations.  Chicago has also made a big investment in the draft, with 38 picks and 12 first rounders, which contributed to their Cup title this year.

At the opposite end of the spectrum are San Jose (22 picks) and Detroit (23 picks), showing that volume is not necessary for success.  Both have decent PPI numbers, but rely more on veterans than youngsters, and because of their talent, have fewer openings for young talent to break into.  Pitiful Tampa was also near the bottom in draft picks (25), indicative of a predilection toward trades, rather than the slower process of building through the draft.

Some teams are just beginning the transformations initiated by their draft strategies.  Colorado and Los Angeles have been stockpiling young talent for awhile, with the Kings being a bit more aggressive in moving veterans out in favor of younger talent.  Just this past year, Colorado cleared the decks of some veterans, and enjoyed the best season the Avalanche have seen in a few years. 

I could go on for days about the data, and how it applies to various clubs and situations, but that’s the fun of the debate, which I will leave for you to ponder.   However, here is my take on what the data shows:

  • Building a team through the draft is a difficult process, but can be done — if you are patient.   The data suggests that the maturation process for a player that is going to contribute approaches 4 years.
  • There are few “sure things” in the draft.  Only those players taken in the first few picks come close to guarantees.
  • While teams do a good job of ranking the talent at a relatively high level (first vs. second round), the distinction gets much hazier when you move to selections within a round.
  • The draft is an important piece of the overall success of a franchise, but is not the only piece.  Teams have proven to be very successful without excess reliance on young players, but with Chicago, Washington, Pittsburgh and others, that trend may be shifting.

The bottom line is one word that all fans hate to hear — “patience”.   Rather than rushing that first or second round pick to the NHL, perhaps better to let them season for a few years, learning how to handle situations in a less highly charged environment.  So, when June 25 rolls around this year, and you start berating your local GM, stop and think about the homework and agonizing he is going through, just to give the team a reasonable statistical chance of having some meaningful contributions from these draftees.   Would you want your livelihood dependent upon a process with a relatively low chance of success?  Such is the stuff of the draft, and while it gives the GM’s ulcers, it provides terrific grist for the mill of public debate.