[Ashasx]

Originally Posted by Enoch Root

This is all well and good, but it actually tells us nothing because it doesn't consider how many goalies were selected in the various rounds. For example, if only 10 goalies were ever selected in the 1st round, then there would be a 100% success rate (which wold be pretty good). Conversely, if the same number of goalies were selected in each round, then this wouldn't look very good for drafting early (which appears to be your point and your assumption).

I ran some numbers form 1997 through 2006. I stopped in 2006 since the goalies drafted since then haven't had enough time to have played a representative number of games. For example, Bernier was drafted in 2006 and has not yet played 200 games, so it seemed unfair to continue with more recent years.

Round / number of selections / % that have played 100 games / % that have played at least 1 game / % that have never played a game:

...1 / 28 / 42.8% / 96.4% / 3.6%
...2 / 28 / 35.7% / 71.4% / 28.6%
...3 / 34 / 20.6% / 50.0% / 50.0%
4-5 / 53 / 7.8% / 31.2% / 68.8%
6-9 / 118/ 14.2% / 35.6% / 64.4%

This pretty clearly shows that if you draft a goalie in the 1st or 2nd round, you are far more likely to get an NHLer than a goalies drafted in the later rounds. Just like with players.

With each subsequent round, it becomes less likely that the goalie will play 100 games (200 games as well), less likely to play at least 1 games, and more likely to be a total bust. JUST LIKE WITH PLAYERS.

Just a few people who have done more analysis than I have and have come up with a similar conclusion, since I'm told nobody takes me seriously:

Jonathan Willis: http://blogs.edmontonjournal.com/201...-anybody-else/
Scott Reynolds: http://www.coppernblue.com/2010/5/28...ders-1997-2005 & http://gospelofhockey.blogspot.ca/20...kents-old.html
Peter Siamandas: http://houseofpuck.com/the-opportuni...ting-a-goalie/
Kent Wilson: http://flamesnation.ca/2012/5/30/you...e-poor-gambles