|
Over the last two weeks, I have proposed a methodology by which to judge team’s success in the NHL draft, and have ranked each team based on this success. Here I will attempt to further validate that methodology by asking the ultimate question: does success in the NHL draft, as measured by the metric I have proposed, discounted GVT versus expected value, correlate with success on the ice?
The question is fraught with 'ifs' and 'buts.' The amount of time that a player stays with the team that drafts him depends on many factors, and the current CBA, which has made free agency available earlier to more players, will probably affect the outcome, although given the number of 12-year contracts being handed out in the NHL it’s debatable whether a players time with their original team is getting shorter or longer. In addition, teams usually get some value out of an impending free agent, as he is usually traded to a contending team at the trade deadline for draft picks and prospects, Jay Bouwmeester notwithstanding. I will therefore ignore all these effects and simply measure whether the discounted GVT predicts in any way a team’s future GVT.
Given that the first draft I analyzed was in 1993, I decided to begin my analysis with the 1994-95 season, which would have been when many of those players entered the league. The last draft analyzed was in 2007, so I have included every season up to the last one.
The correlation turns out to be much stronger than I would have ever expected:
Rk Team Draft GVT
1 Senators 300 354
2 Devils 231 586
3 Flyers 196 463
4 Avalanche 179 590
5 Sharks 72 57
6 Canadiens 65 -59
7 Blues 53 184
8 Wild 46 21
9 Sabres 46 268
10 Islanders 41 -529
11 Red Wings 35 1055
12 Hurricanes 28 -184
13 Maple Leafs 25 1
14 Stars 24 455
15 Bruins 17 23
16 Kings 14 -298
17 Thrashers -23 -535
18 Penguins -34 -27
19 Predators -42 -133
20 Ducks -52 -143
21 Blue Jackets -59 -416
22 Canucks -65 60
23 Rangers -73 -89
24 Oilers -75 -144
25 Blackhawks -123 -140
26 Capitals -124 -71
27 Panthers -145 -296
28 Flames -151 -116
29 Coyotes -198 -277
30 Lightning -206 -659
The correlation is very easy to see graphically: note how many teams there are in the upper-right (good draft, good teams) and lower-left (bad draft, bad teams) quadrants, as opposed to the other two. The extremes certainly work well, as the worst team of the last 15 years (Tampa Bay) is also the worst in drafting, and the top 4 teams in drafting are 4 of the 6 strongest franchises in recent memory. In fact, only Dallas and Detroit have managed to put together consistently winning teams without significant draft success in recent years. In Detroit’s case, it is partly a “hangover” effect from their earlier draft success: in the 4 drafts from 1989 to 1992, Detroit obtained a discounted GVT value of 415, more than any franchise has obtained in the previous 15 years. It is also due to their ability to keep their key players in the Winged Wheel uniform, from Yzerman, Fedorov and Lidstrom to Datsyuk and Zetterberg.
In Dallas’s case there is no previous draft success to explain the on-ice product. The Stars benefited from having deep pockets under the old system, managing to sign players like Brett Hull and Ed Belfour as free agents in their primes. They also have been shrewd traders: in 1996 they parlayed Kevin Hatcher into Sergei Zubov, a move that seems like larceny in retrospect. They have, like the Red Wings, managed to keep some of their key players as Stars for their entire careers, Mike Modano, Brenden Morrow and Jere Lehtinen being the most notable examples.
What is amazing about the correlation is that it seems to overexplain success: the best-match slope between draft GVT and team GVT is 2, meaning that for every 1 discounted GVT a team drafts, it obtains 2 on the ice. I can think of three explanations for this:
- I have used too high a discount rate, and teams are often able to get full value for their players through trades or compensation even if those players end up playing for another team at some point in the future.
- Having good players may create a snowball effect: free agents want to sign with teams that are successful, and good players on good teams will take a “hometown discount” in their contract negotiations to stay.
- Lastly, drafting is but one aspect of NHL organizations, and may simply reflect organizational competence. Teams that draft well do other things well, and vice-versa.
I don’t think the first reason is correct, but I believe the other two are at least partly. We know that good teams don’t struggle to keep their players, and have thoroughly mocked one particular player who signed with a team assuming that they gave him the best chance to win the Stanley Cup. Conversely, we know that Ilya Kovalchuk may very well want to leave Atlanta, and Dany Heatley wasn’t asking for a trade two years ago when the Senators reached the Stanley Cup finals.
As to the last reason, that drafting is merely a reflection of overall quality, I believe it is also true. While it may seem unfair, it matches well with our intuition: after all, who really think the Coyotes scouting organization is among the NHL’s elite? Who was really surprised to see the Devils near the top of the draft rankings?
While it may seem obvious that draft success has direct implications for success on the ice, we have now managed to quantify it. More importantly, we have even observed that on-ice success is magnified by good drafting, as players seek other quality personnel to surround them. NHL teams should pay less attention to their gut feelings and more attention to Iain Fyffe.
Tom Awad is an author of Hockey Prospectus.
You can contact Tom by clicking here or click here to see Tom's other articles.
|
0 comments have been left for this article. 
Conn Smythe Watch: Lundqvist Leads