July 03 2013 09:21AM
What constitutes a deep draft? Are there any patterns in the strength of draft years? Do some teams dominate the draft to a greater extent than others? Are there any historical trends that can be found by looking at the overall draft history? What teams find success at the draft and are there any patterns that can be gleaned from history?
In order to try and find some answers I collected data on the NHL Entry Draft from the first NHL Entry Draft in 1979 to 2008.
The cutoff date for draft data is self-imposed and reflects my belief that it can take up to five years for prospects to establish themselves as potential NHL players.
My criteria for determining a successful draft pick was a player that had managed 200+ NHL games and thus established himself as a useful player for more than two-seasons’ worth of NHL games. For the draft years 2006 to 2008 I had to extrapolate somewhat based on trending players and development to-date.
One might argue that a player taken 4th overall who only manages to register 250 NHL games can hardly be labeled a draft success and they would be right. However those instances are few and far between and are generally statistical outliers whose impact on the data will be weeded out by the sheer volume of the more normalized draft experiences. For every Lars Jonsson (8th overall, 2000) there is also a Henrik Zetterberg (210th overall, 1999). Over the course of the long-term the two data sets cancel each other out.
It also needs to be emphasized that at this stage this is a quantitative analysis. I am not at this time parsing the data out to find the dominant players or the generational talents nor am I discriminating against the blue-collar, bottom-six players and reclamation projects that circulate through the league every year. If you want to argue that team x is statistically bad at the draft podium but regularly hits home runs when they do connect, that is all fair but for that sake of this exercise we will begin by gauging the proverbial on-base percentages in order to keep the data-set quantifiable and objective. A qualitative analysis will follow in the final article of the series.
What I am looking for is a baseline for draft success in so far as it suggests whether a draft pick, in any round and year, has an overall likelihood of emerging as an NHL player. There are obvious flaws in the data, and I am working on drilling deeper down to find how teams have done in particular years and in particular rounds, but that is a project for another time. The data that has been collected herein is mostly useful in gauging the difference between teams that draft well and those that do not, draft years that are deep as opposed to those that aren't, as well as determining a kind of league-wide baseline of average success and putting it all into an historical perspective.
Another caveat is that I am not judging the management of any of the teams from the perspective of Hockey Operations. This separates the scouting staff under the General Manager from that GM's hockey operations decisions. I haven’t found a metric yet for bad management, despite repeated attempts.
Rather than post a massive table of data, I am going to try and break down some of the notable points into graphs and, when necessary, provide charts on summary data.
For the time being the data gives us an idea of draft probabilities, particular organizations’ success or failure, and perhaps a hint at some cycles that could greatly inform those who follow the draft.
I’m going to begin episodically, looking at the draft by decades because the amount of information is significant. Then, in the final installment I’ll have a complete overview of the history of the Entry draft from 1979 to 2008 as well as a few smaller items that have caught my eye.
The First Decade of the Draft
If we were to exclude the first two years as statistical outliers and limit our analysis to the drafts from 1981 to 1989 we can see a pattern of variance between draft years of approximately 7% to 8%, which would come out to just under one player per team. Keep this variance in mind as we progress through the series.
For starters, the 1979 draft is something unlike anything I think any of us will ever see again. Granted, there weren’t many selections made, 119 in total, but the fact that 53% of those selected managed to play 200 or more NHL games speaks to an outstanding draft class. This was the first year that the league had moved to the open Entry Draft format from the Amateur Draft. Essentially the rules were changed to allow professional players to be selected as well as lowering the age of eligibility from 19 to 18. The format change may have opened some doors for some players who might not otherwise have had an opportunity.
While this did result in a deeper-than-normal draft class of eligible players, the results are still far and away beyond anything else in NHL draft history, both by quantitative and qualitative standards. The following two years show something of an echo effect, before settling into the historical average that hovers around 20%.
Reading through the ’79 draft class, there are many players whom a younger audience might not recognize but who were stalwart NHL players for many years, on a number of teams. The great number of NHL players that emerge from a small sample size like the 1979 draft suggests that while there were fewer roster spots available to NHL players, the league was also drawing from a somewhat smaller talent pool, especially overseas. In fact, of the 126 players taken only fourteen were not listed as Canadian (eight Americans, five Swedes, and one Czechoslovakian). In the years following 1979, NHL teams were relying on many of these players later in their career for roster depth when younger players were becoming available.
Alongside the easily recognizable names from that draft class like Mike Gartner, Ray Bourque, Kevin Lowe and Dale Hunter are lesser known players who also managed to play 1000 games or more including Dave Christian, Brent Ashton, Mike Ramsay, Thomas Steen and Jay Wells.
These aren't names one finds on heritage jerseys, but they nevertheless managed long careers in the NHL.
The first draft was limited to six rounds, but was expanded the following year to ten rounds and it quickly became a strategic focal point for GMs as there are several instances of teams loading up on picks early in the process, the Chicago Blackhawks in 1980, the North Stars in 1982 and 1983, and the Islanders and Canadiens in 1985. All had 14 or more selections in those years.
As we’ll see, the Blackhawks would continue this practice well into their more recent history, monopolizing the draft in particular years, usually with disappointing results.
By 1981 the draft success rate falls by more than half of the 1979 rate, partially due to the expansion of the draft from 6 rounds to 11 as more selections meant that there were more marginal players taken and a greater chance of failure began to enter the equation. But also because the wealth of talent that had been made available by the transition to an Entry draft from an Amateur draft, which opened the floodgates of previously ineligible talent, was now exhausted and had settled into a rate close to the historical average.
*A note on 1984. The St. Louis Blues boycotted the draft over the contested sale of the team to Bill Hunter and a collection of investors looking to move the team to Saskatchewan. The team vehemently opposed the sale, and the league was strongly against it as well, but lacked a local buyer to counter-offer. As a result, the Blues refused to send a representative to the draft and their picks were all forfeit.
Team By Team Success Rates
I’ve included the modern names for organizations: Carolina for Hartford, Colorado instead of Quebec City, New Jersey for the Colorado Rockies, Dallas for the North Stars, and Phoenix for the old Jets. Changes in location did not often affect the hockey operations personnel as they were relocated from one city to another. There is no evidence that moving a franchise has had any impact on the strength of their draft record.
The breakdown of the draft over the course of this decade is above, and provides illustration of how each team did at the draft. Minnesota, Los Angeles and Philadelphia were the worst underperformers during this period, while the Rangers, Red Wings, Nordiques and Sabres were the furthest ahead. Interestingly, the Red Wings, and the Nordiques to a lesser extent, would use some of that draft success to eventually become powerhouse teams in the following decade.
Some may note that there does not appear to be any clear correlation between the teams that were most successful during this period (Edmonton, Calgary, the Islanders) and their relative draft success rate. This could be partially due to the lag effect a drafted prospect has on their respective franchise, usually around three to five years from draft day to impact player. This is certainly the case for the Islanders, whose team had been assembled during the 70s under Al Arbour.
It can also be skewed based on draft standing, as poor teams were drafting higher in the order (the Red Wings during this period, for example, who also boast one of the highest draft success rates over this time). Finding good talent in the draft is at least partially a result of the old real estate adage: location, location, location.
However, in this instance I believe it illustrates the difference between a quantitative study of the draft versus a more qualitative approach. The Oilers drafted the core of a championship team between the years 1979 and 1981. That they were barely above abysmal for the remainder of the decade was masked by the phenomenal talent they had accrued earlier.
The worst team at the draft podium during this period (Los Angeles) had almost exactly half the success of the best team (Montreal). This provides us with an excellent example of how to parse these results into an actual narrative by comparing the two teams directly.
Statistically speaking the difference between the two is dramatic and if we take into account a preferable draft placement for the Kings due to their poor finish against a less-desirable draft position for the Canadiens, then the result becomes even more so.
The Kings found 19 NHL players at the draft between 1979 and 1989 whereas the Canadiens uncovered 35 over the same time span, a difference of sixteen players spread over decade, or about 1.6 players per year. While that may not seem significant, having, on average, one to two internal prospects pushing for employment in the NHL every season is a key indicator of a successful team.
Another thing this direct comparison can illustrate is the difference in draft percentages between teams on a year-by-year basis, which can translate roughly to perhaps one body per draft per year. Taken in small amounts this does not have any great impact, but when taken in a longer historical context it provides significant reinforcement to the argument that the draft is now, and for the most part always has been, a key component of a franchise.
Overall, the decade from ’79 to ’89 saw approximately 2543 names called at the Entry Draft, 581 of whom managed to carve out careers of 200 NHL games or more. That gives one a success rate of 23%, which we will see is only slightly higher than the following two decades. While it is important not to confuse correlation with causation in this instance, it is plausible to suggest that the initial rush of talent was a result of the changes to the draft rules, opening up geographically protected areas and previously untapped pools of talent to NHL teams. That the NHL was able to absorb four WHA teams at the same time as the initiation of the Entry Draft format and still fill out those rosters with so many NHL players suggests that a rich vein of talent had been opened.
One main item that stands out in this initial stage of analysis is that there is no direct and necessary causal realtionship to drafting success and organizational success, although there is a stronger correlation to drafting failure and organizational ignominy. Teams like the Canadiens, Red Wings and Nordiques were among the best drafting teams and later became successful on account of that ability. At the same time the Sabres, Rangers, and Bruins were also among the top of the drafting class during the same period but in large part did not reap any significant rewards as a result of this ability.
Strong drafting, like goaltending, isn't everything unless you don't have it in which case it is.
Another item that stands out is that the difference between a good draft year for a team and a bad draft year can be as narrow a margin as one prospect, even a complementary one. Few teams dominate a draft to the extent that they completely eclipse the competition, but the margin of victory in the NHL is such that the addition of one or two bodies per year accumulated over the course of a decade can have a significant impact on the health of a franchise.
The decade that ushered in the first ten NHL Entry Drafts brought with it talented players who still dominate the NHL landscape today such as Mark Messier, Steve Yzerman, and Joe Sakic. These players have found few rivals in any of their successive generations thus far and as much as it seems that the NHL was finding its way during the first few years of the Entry Draft, it is unlikely that we will see another decade of talent like the 80s.
Next we will examine the draft from 1990 to 1999. The Dark Ages are ahead.