There has been a lot of talk about trading down, as well as, how likely a pick is to succeed in the NHL. I was curious about the actual numbers, so I looked back (again) at the drafts from 1995-2006. I picked these years because I believe 2006 is the most recent draft from which we can draw any conclusions (it may still be too recent). I, arbitrarily, divided each draft into groups of 5, then looked at two criteria. First, if a draftee had not played at least 2 full NHL seasons, 164 games (3 seasons for top 5 picks), then I considered the pick a bust. I gave the goalies a little wiggle room on this rule so goalies like Schneider, Bernier, Rask, and Neuvirth were not considered busts. Second, I determined if each player had played in an All-Star game. I looked at the first 75 draft positions in each draft, for a total of 900 players.
Of the 900 players, 526 of them were busts. This means that nearly 42% of all players taken with the first 75 picks in the draft played significant games in the NHL. This number was a little higher than I was anticipating and for a surprising reason. The top of the draft went as expected. Most players drafted in the top 10 played significant NHL games, (92% for the top 5 and 77% for 6-10). From 11-20, it fell to just over 50% (54%), then recovered to 70% for 21-25 before falling to 45% for 26-30. After that, things become strange. Instead of falling through the second on into the third round, it stabilized for the rest of the 75 picks. It starts at 23% from 31-35, climbs to a high of 30% at 41-45, but only falls to 27% by 66-70 and is still at 22.5% at 71-75. This means that through these 12 years, just as many players made the NHL from the beginning of the third round as the beginning of the second. I found this shocking, but certainly the All-Stars wouldn't be so evenly distributed in the second and third rounds.
Throughout these 12 drafts and 900 players only 74 appeared in an All-Star game, or just 8% of all players. Just the opposite of busts, most of the All-Stars appeared in the beginning of the draft. Almost 34% of All-Stars appeared in the top 5. This is nearly 3 times more than any other 5 positions. From 6-10 and 11-15 it was fairly stable, at 12.2% and 10.8% respectively. It then fell to only 5.4% from 16-20 before recovering to 10.8% of All-Stars (a similar pattern to the busts) at 21-25. The number of All-Stars fell to 7% at 26-30. For the next 45 players it never rises above 4 All-Stars per 5 places and only 36-40 and 66-70 produced no All-Stars during this period. If we consider that the first round lasts from 1 through 30 (it doesn't through this period due to expansion, but we'll pretend), then 81% of All-Stars that are taken in the top 75 go in the first round, 12% in the second, and 8% through the first half of the third (more than 100 due to rounding). Just like with the busts, the second and third rounds (the first half anyways) are indistinguishable with a slight nod to the third round.
Defensemen versus Forwards:
To the 74 All-Stars selected in the top 75 picks, we can add 27 taken in later rounds, for a total of 101 NHL All-Stars taken over this period (as far as I could discern). Of those 101, 57 of them are forwards, 31 are defensemen, and 13 are goalies. Among the forwards, 77% are drafted in the top 30, 40% are drafted in the top 10, and 32% in the top 5. For defensemen, 36% are taken in the top 30, 23% are taken in the top 10, and only 10% are taken in the top 5. In other words, All-Star forwards are more than twice as likely to be selected in the first round and three times more likely to be taken in the top 5 than defensemen.
As concerns forwards and defensemen, it played out about how I expected. All-Star forwards were taken at the top of the draft, with defensemen scattered throughout. I did find it interesting, however, that after the top 30 picks, both the likelihood of a prospect making the NHL, as well as the possibility of them becoming an All-Star was fairly constant through, at least, the 75th pick. This makes me wonder if we, as fans, are overvaluing the second round picks and, perhaps, undervaluing the third round, or is this just a quirk of a small sample and the mystery of numbers?
A Few Tidbits about the drafts:
The 2003 draft is just as good as everyone thinks. It produced 17 All-Stars (and Bergeron, who I am shocked has never been an All-Star) in the first 75 picks. The next draft for All-Stars was a tie between 2005 and 1997 with 9 each. As for the 2003 success rate, an astonishing 61% of the first 75 players have played at least two seasons in the NHL and 28 of the top 30! On the flip side, the worst draft for NHL success in this 12 year period was 1997. Ironically, it is the only draft in the sample where every player in the top 5 was an All-Star, but overall only 25% saw significant time in the NHL (less than 12% if you exclude the top 15) and not a single player between pick 27 and 47 played 162 games. The draft with the fewest All-Stars is a tie between 2006,2004,1999, and 1996 with 3 each. The lesson here is that not all drafts are created equal. The Ducks drafted both Perry and Getzlaf, the Preds got both Suter and Weber, and the Rangers got Jessiman, Baranka, AND Roche in the mighty draft of 2003.