Normal Goaltender Development, part 3: By draft round

This is part 3 of a series dealing with “normal” goaltender development after being drafted, and in this post I’m looking at the relationship between draft position and career path. You can find the earlier sections through the following links:

• Part 1

• Part 2

• Glossary and Notes

There are three “axes” of analysis in determining how players’ careers progress after their draft:

1. their career path towards the NHL

2. how fast they get to the NHL

3. how long they stay there & how “productive” they are once there.

I’m calling these Path, Speed, and Longevity.

And looking at these three things in relation to draft position gives us an admittedly imperfect tool for relating talent level to career path, at least at a macro level. Again, these averages elide a great deal of individual variation and should be taken as descriptions of groups of goalies rather than predictors or evaluations of specific individuals.

*****

I need to start off by correcting an error in this chart from part 1.

	# of Gs drafted	Avg # of NHL gms
Round 1 (1-30)	24	122
Round 2 (31-60)	29	51.93
Round 3 (61-90)	31	50.55
Round 4 (91-120)	35	13.69
Round 5 (121-150)	44	3.98
Round 6 (151-180)	48	12.42
Round 7 (181-210)	37	20.76
Round 8+ (211-300)	45	30.53

While the numbers are correct by draft position, it was a mistake to assume that draft position corresponded to draft round cleanly. For instance, while 31 goaltenders have been drafted between positions 61 and 90, not all of those draftees were actually taken in the third round of their respective drafts and not all third rounders were taken before number 91. The number of draftees per round varied from year to year, so each goaltender needed to be assigned to his actual draft round manually.

That has been done for part 3. And this corrected chart shows the general draft profile by draft round, as opposed to overall draft position. These are the averages taking into consideration all draftees, even those who have never played a game.

	Draft position range	No. of players	Avg no. of NHL gms
Round 1	1 – 29	25	117.16
Round 2	31 – 64	31	64.87
Round 3	62 – 96	34	32.62
Round 4	92 – 133	35	12.43
Round 5	121 – 166	54	8.26
Round 6	153 – 197	46	10.74
Round 7	182 – 226	37	26.49
Round 8 & 9	228 – 291	30	32.97

Note: from this point on when I say “Round 8” it includes Round 9.

The conclusions drawn remain the same. The most successful goalies in terms of games played are the first rounders, followed by 2nd & 3rd, then the late rounders, then the middle rounders. This is even more starkly apparent when we include the number of goaltenders who become starters for even one season (defined here as a year where a goalie played at least 60% of a season’s games).

Path:

For goalies with at least one NHL game, career paths look like this.

No. of GTs in each phase:	All		Rd 1		Rd 2		Rd 3
	104		22		17		13
Juniors	64	61.54%	15	68.18%	10	58.82%	9	69.23%
NCAA	23	22.12%	2	9.09%	3	17.65%	3	23.08%
Secondary	69	66.35%	14	63.64%	8	47.06%	7	53.85%
Primary	104	100.00%	22	100.00%	17	100.00%	13	100.00%
Callup	99	95.19%	22	100.00%	16	94.12%	13	100.00%
Callup&backup	94	90.38%	18	81.82%	16	94.12%	11	84.62%
Starter	31	29.81%	10	45.45%	5	29.41%	3	23.08%

No. of GTs in each phase:	Rd 5		Rd 6		Rd 7		Rd 8
	14		9		10		9
Juniors	9	64.29%	4	44.44%	6	60.00%	2	22.22%
NCAA	4	28.57%	4	44.44%	2	20.00%	3	33.33%
Secondary	11	78.57%	6	66.67%	7	70.00%	7	77.78%
Primary	14	100.00%	9	100.00%	10	100.00%	9	100.00%
Callup	12	85.71%	9	100.00%	8	80.00%	9	100.00%
Callup&backup	14	100.00%	8	88.89%	9	90.00%	9	100.00%
Starter	1	7.14%	2	22.22%	3	30.00%	4	44.44%

There are a few important things to notice here. One thing is that few high draft picks go into the NCAA, which should factor into our understanding of an NCAA goalie’s career progression. We should also note that every goaltender spent at least one season in a primary minor league (AHL, SHL, or KHL). None of them “leapfrogged” directly to the NHL, and the vast majority, regardless of draft round, had at least one “callup” season with games in both the AHL and the NHL.

For virtually all goalies, the path to the NHL leads through the AHL at some point. In fact, only two of the goalies studied have played zero AHL games: Henrik Lundqvist, who spent 4 years in the SHL before becoming a starter for the New York Rangers, and Alexander Pechursky, who played in the WHL and CHL (and one game with the Pittsburgh Penguins), but has spent the rest of his career in Russian leagues. Regardless of talent level or any other factor, everyone plays in “the A” at some time.

Speed:

How fast do players move through the phases? With some exceptions, I found very little overall variation related to draft round. For instance, the overall average length of time in juniors (among players who went into that phase) was 1.63 seasons, and no round averaged more than 1.8 years (round 2) or less than 1.25 years (round 6). There was no real pattern to the differences, though goalies drafted in the late rounds spent slightly less time in this phase and more in others (especially secondary minors). Early draftees did not spend significantly less time there than the overall population.

Thus I feel that it’s possible to say that there’s a fairly normal career path that includes the following milestones: about 1-2 years in juniors OR about 2-3 years in the NCAA, some time in a secondary minor league (see below), and about 4 years in a primary minor league, including at least one year in the AHL and the NHL. The average player gets his first game within 3-5 years of his draft, but there’s a great deal of variance in the timing of that, and early draftees see a slight advantage in that department.

The secondary league pattern came closest to the kind of pattern I think most of us assume players go through. Players drafted in later rounds played appreciably more time in secondary leagues than players drafted in early rounds. This may have a relationship to age, because as I noted earlier, later draftees spent less time in junior leagues, which are age-restricted leagues.

Longevity/Production:

Here I measured longevity in the number of seasons in which a player played any NHL games, and production in number of NHL games.

Overall, the 104 goalies average 90 total games and three and a half NHL seasons.

	Rd 1	Rd 2	Rd 3	Rd 4	Rd 5	Rd 6	Rd 7	Rd 8
Gms played	133.14	118.29	85.31	43.5	31.86	54.89	98	109.89
NHL seasons	4.41	4.76	3.38	2.6	2.07	2.44	3.2	3.89
Avg GP/NHL season	21.78	19.68	17.22	10.02	6.71	17.22	17.1	18

As we’ve seen before, middle rounders tend to have less success in both the length of their career and the number of games they play.

There’s one more way to look at longevity, production, and speed of progression: the ratio of games played to years since a player was drafted. If a player gets to the NHL quickly, stays there, and plays a lot of games, he’ll score high on this list. Career backups and players with brief careers where they were starters will score better than recent draftees with only a few games, who will in turn score better than players who had a brief stint in the NHL before dropping out of the league.

In other words a player who plays 40 NHL games within 10 years after his draft will score a 4. A young goalie four seasons past his draft who has had a handful of NHL games will score around 1. No goalie with at least 200 games scored less than 16; no goalie with at least 100 games scored less than 8.

gms/yrs	Round1	Round2	Round3	Round4	Round5	Round6	Round7	Round8	All
0.01-1.00	5	0	3	6	9	3	5	4	35
1.01-10.00	7	9	7	1	3	4	2	1	34
10.01-20.00	4	4	0	3	1	2	1	1	16
20.01-30.00	2	3	1	0	1	0	1	2	10
30.01-40.00	3	1	2	0	0	0	1	1	8
>40.00	1	0	0	0	0	0	0	0	1

Less than 20% of NHL goalies have a ratio higher than 20; more than 65% have a ratio lower than 10. Early rounders account for 7 of the 9 players who score above 30. However, this score also shows how successful late round players can be. Although about half of players in rounds 7-9 score less than 10, 5 of them scored over 20.

The top thirty netminders by score:

Overall	Player	rd	GP	YRS SINCE DRAFT	GP/yrs since draft
1	Marc-Andre Fleury	1	467	10	46.70
205	Henrik Lundqvist	7	511	13	39.31
25	Cam Ward	1	431	11	39.18
5	Carey Price	1	310	8	38.75
72	Jonathan Quick	3	286	8	35.75
2	Kari Lehtonen	1	380	11	34.55
69	Steve Mason	3	239	7	34.14
44	Ilya Bryzgalov	2	425	13	32.69
258	Pekka Rinne	8	293	9	32.56
41	Ondrej Pavelec	2	231	8	28.88
1	Rick DiPietro	1	318	13	24.46
64	Jimmy Howard	2	234	10	23.40
77	Craig Anderson	3	318	14	22.71
63	Peter Budaj	2	272	12	22.67
214	Cristobal Huet*	7	272	12	22.67
271	Jaroslav Halak	9	220	10	22.00
161	Mike Smith	5	263	12	21.92
23	Semyon Varlamov	1	147	7	21.00
291	Brian Elliott	9	204	10	20.40
232	Martin Gerber*	8	229	12	19.08
99	Ray Emery	4	228	12	19.00
34	Michal Neuvirth	2	121	7	17.29
21	Tuukka Rask	1	138	8	17.25
171	Roman Cechmanek*	6	212	13	16.31
14	Devan Dubnyk	1	139	9	15.44
52	Corey Crawford	2	152	10	15.20
99	James Reimer	4	104	7	14.86
8	Pascal Leclaire*	1	173	12	14.42
60	Dan Ellis	2	184	13	14.15
207	Anders Lindback	7	62	5	12.40

Rounds 1-3 are well-represented, and rounds 7-9 make a respectable showing, but only 4 of these players comes from the middle rounds. Workhorses who’ve been around a while are at the top, followed by younger starters and long-term backups. Only four of these top thirty are no longer active in the NHL, indicating a productive but brief career.

I think that this is a good indication of how individual players’ career arcs related to the population of all drafted goaltenders as a whole. These top 30 are the guys who got to the NHL faster, stuck around longer, and were relied upon to a greater extent than 90% of goaltenders drafted between 2000 and 2010. They are, by definition, the high-end outliers in terms of career progression, and it’s important, I feel, to remember that.

***

This was originally intended to be as far as I would go with this project, but the question of age has kept coming up: age at draft, age at first NHL game. There are questions such as whether, overall, we are making assessments about potential at too young an age and whether there’s greater success among players drafted when they’re too old for junior leagues.

Thus, I’m planning on adding age as a final analysis as soon as I can get the new data gathered and analyzed. In the meantime, I’ll do a wrap up of the conclusions we can and can’t draw from this data for next time, in particular as this huge amount of information relates to my original question of how fast is too fast to push Andrei Vasilevski.