A couple of weeks ago, I looked at the 2nd overall forward taken in the last 18 drafts to see if there was any statistical evidence to indicate whether or not making the jump directly the the NHL after being drafted can be detrimental to a player's development. What I found was that for those highly touted forwards, spending more time in juniors after being drafted didn't seem to have much statistical impact on either the player's first season nor did it seem to effect the length or quality of his career.
That was, admittedly, a rather narrow look as I was only taking the 2nd best forward from each draft. So, I went back to HockeyDB and ran a new test. This time, I looked at all junior forwards drafted in the top 30 from 1990-1999. I didn't look at more recent drafts because I wanted to make sure I had a good representation of a player's career potential; although many players drafted in the 90s are still active, they've generally hit their full potential by now.
Here's what I did. For those 10 drafts, I looked at every forward drafted in the top 30. 30 was my arbitrary cutoff; I originally was including more, but I found that the bulk of the players taken after this point took a much longer route to the NHL and spent considerable time developing in the AHL. I ditched any player who didn't have at least 100 career NHL games and those who never played 20 in any one season. I also excluded a handful of 20-year old junior draftees (like Scott Parker) as I wanted to ensure that everyone's development arc was as similar as possible. For those 66 players, I looked at their performance over their "rookie" season (for simplicity, I just used the first NHL season with at least 20 games), their performance over their "best" season (again, for simplicity, I used most points as the marker here) and, finally, the performance over their career. Again to keep things relatively simpe, I used points-per-game as a performance measure. It's not perfect, but I think it works for this study.
| # | Avg Picked? | Junior Years Post-Draft | Rookie GP | Pts | PPG | Best GP | Pts | PPG | # of Years | Career GP | Pts | PPG | ||
| 19 | 4.8 | None | 66 | 27 | 0.41 | 78 | 68 | 0.88 | 6.5 | 828 | 511 | 0.62 | ||
| 24 | 11.1 | 1 Season | 62 | 26 | 0.42 | 75 | 56 | 0.75 | 6.8 | 668 | 386 | 0.58 | ||
| 23 | 16.9 | 2 Seasons | 54 | 21 | 0.39 | 72 | 47 | 0.65 | 7.8 | 538 | 246 | 0.46 |
The data above shows the average rookie season, best season (and number of years it took to get there) and career numbers. Rookie PPG numbers are almost identical and are on-par with with my earlier study. The "best season" numbers certainly don't get any better by having a player wait; the only thing it seems to ensure is that the player will be a little older when they hit their career zenith. And, finally, it doesn't appear that the length or quality of a player's career is improved by having a player spend an extra year or two before being thrown at the NHL. Indeed, it would appear that the longer a player waits to be exposed to the NHL, the shorter (and poorer) his career will be. However, we have to look at the Avg Picked column - on average, the players who made the jump were drafted much earlier in the draft. It is likely that the dropoff as you move down is directly related to the quality of the player (obviously, better players are taken earlier in the draft with the exception of players drafted by the Islanders, of course).
To test this, I ran the same numbers above, but this time focused on only players taken in the top 10 of a draft. Collectively, the skill level should be comparably over the 10-years of drafts. Here's those numbers:
| # | Avg Picked? | Junior Years Post-Draft | Rookie GP | Pts | PPG | Best GP | Pts | PPG | # of Years | Career GP | Pts | PPG | ||
| 17 | 3.8 | None | 67 | 28 | 0.42 | 78 | 69 | 0.89 | 6.5 | 822 | 515 | 0.63 | ||
| 12 | 5.6 | 1 Season | 61 | 27 | 0.43 | 75 | 55 | 0.73 | 6.3 | 669 | 387 | 0.58 | ||
| 6 | 6.7 | 2 Seasons | 66 | 34 | 0.52 | 74 | 52 | 0.70 | 6.2 | 532 | 287 | 0.54 |
With this narrower scope, you don't see a change in the rookie year between players that jumped immediately and those that waited a season. You do a see a slight bump (about 7 points over 82 games) for players who spent two years in juniors, but that appears to be the only benefit. Based on this, players who make the jump immediately have, on average, better and longer careers. Even with this narrower look, we still have a situation where the players making the jump are coming from a slighter better talent pool. Because of this, I don't know that we can say definitively that joining the NHL at 18 helped in player development, but I do think we can say that making an immediate leap doesn't hurt in the short run or the long run.
Nor does there appear to be a great risk of ruining a player by exposing him too early. Joe Thornton, for example, had just 7 points in 55 games in his rookie season. That abysmal .13 PPG is the 2nd lowest of the 19 players who stepped immediately into the NHL. If that season has hurt his career, I'd hate to imagine what he was really capable of, as Thornton has the 2nd highest career PPG of all 66 players at 1.01 PPG. Owen Nolan had just 13 points in his 59-game rookie year (.22 PPG) but he's gone on to average .76 points per game over his 1,127-game NHL career (longest among the junior players drafted in the 90s). Conversely, the 2nd and 3rd best rookie seasons from players leaving juniors after the draft were from Pat Falloon and Alex Daigle.
Every year, teams who have drafted a player out of juniors in the first or second round are faced with a decision: play him or send him back to juniors (junior league draftees can't play in the AHL until they turn 20). This year, Matt Duchene and John Tavares seem to be on a collision course with the NHL. Evander Kane is reportedly on the bubble in Atlanta. Scott Glennie has already returned to his junior club and Nazem Kadri and Ryan Ellis are reportedly headed there soon as well. Whenever these moves are debated, the question of how such a player will have his development effected is invariably raised. In my mind, the answer basically boils down to "not much".