Counting Cards at the Draft

We all know that drafting is an inexact science. It is extremely difficult to project what a player is going to do down the road. In doing research on the draft it’s clear the top half of the first round (picks 1-15) have the highest probability of NHL success.  This chart from Derek Zona of Copper Blue outlines these odds very nicely.

Draft Number	Total	Top Players	Percentage
1	7	6	0.857
2-3	17	15	0.882
4-7	32	16	0.500
8-13	51	21	0.412
14-25	99	29	0.293
26-50	202	30	0.149
51-100	405	30	0.074
101-200	771	30	0.039
201+	572	18	0.031

He looked at drafts from 1997-2005. He defined a top player as having as having played a minimum of 200 NHL games having scored at least 0.5 points per game. This eliminates all the grinders and goons who can easily be had in free agency each season for cheap. For defencemen he used the same 200 games played minimum, but reduced the scoring rate to 0.15 points per-game and 18:30 time on ice, this again eliminates any goons but also eliminates players who are just power-play specialists.

As you can the drop from the top picks is pretty staggering.  After pick 7 you have a less than 50% chance of becoming a top NHL player. It’s for all intents purposes a coin flip.

The goal of a team’s draft should be to beat those odds. If you can find a top NHL player after pick 13 you have done very well for your organization. This is what scouting boils down too in my opinion trying to beat the odds. Everyone knows who the top 10-15 players are in a draft class the hard part is knowing the guys after that.

What I want to do is use numbers to study what players have done as amateurs and see if any trends emerge with the goal of beating those odds. This is much what Billy Beane and the Oakland A’s did in Moneyball. The odds were (and still are) stacked against them and they used stats so that they could stack the deck and be card counters at the casino. I want to be able to be the card counter at the draft. 

The best research I have found was the study previously mentioned in this blog from “That’s Offide” Defence, Defencemen andthe Draft. He of course found that among defencemen selected in the top three rounds, the ones who became NHL regulars scored at 0.6 points per game in junior. I have done some research trying to find other studies like this and so far to no avail. In the months leading up to the draft I hope to replicate this study using other leagues and positions. With this I hope to find out more ways to predict NHL success which I understand is not perfect and there will always be exceptions to the rule. But hopefully it will be a better understanding of how to predict future NHL players which can increase the odds of draft picks which really should be every team’s goal.