Are Some Prospects More Likley to Fail than Others?
I went to Thrashers Prospect Camp today but I’m not going to post anything more about camp until after the final session on Thursday. Instead I have a post about prospects more generally.
Visiting prospect camp started me thinking about which prospects will succeed and which will fail. Baseball analysts have suggested that high school pitchers are more likely to fail to develop into major league players and therefore teams might be wise to think twice about drafting them high in the MLB draft. If we applied this approach to hockey, is one position more likely to produce busts than another? Are forwards more likely to flame out or is it defensemen who are more likely to become busts?
As I was thinking about this I began to suspect that among the three basic positions in hockey it would be defensemen who would produce the highest failure rate. Why? Evaluating the defensive abilities of a player are a bit more difficult than the scoring abilities of forwards and so I extrapolated that defensemen would probably produce more failed prospects.
But how to test this question? I was thinking about this question last night as I lay awake with a case of insomnia and finally I got up in the middle of the night to see if I could answer it to my own satisfaction since I wasn’t sleeping anyway. I pointed my browser to www.hockeydb.com and their wonderful collection of NHL draft lists. I decided to do a survey of all 1st round draft picks over a five year period (1995-1999) and calculate the failure rate of players by position.
What constitutes a “failed” prospect? For this little study any player who has played at least 250-300 NHL games and has a regular NHL job is a success and any prospect who does not meet that criteria is a failure at this point. (Note: goalies don’t get credit for games played when they are dressed as a backup, so I lowered the criteria for goalies and paid more attention to if they had NHL jobs or not.)
The 1st round picks between 1995-1999 add up to a total of 133 players, not a huge sample size but enough to give us some idea if any pattern exists. My quick and dirty late night survey of these five drafts found that my hunch is false. (It is not every day that I admit I’m wrong in print, so enjoy it.) Defensemen drafted in the 1st round between 1995-1999 were no more likely to fail to develop than forwards. The failure rate of defensemen was 49%, which was identical to the 49% failure rate for forwards. However, goalies did significantly worse than forwards and defensemen, with 62% of those drafted failing to attain NHL jobs as starters or backups. This high failure rate caught me by surprise a bit, but perhaps GMs should think twice about using a high draft pick on a netminder.
While I was going through these five recent drafts, I though it would also be interesting to consider where players were drafted from. Baseball analysts have compared college players to high school players, so I decided to compare draft picks from Juniors, NCAA, and European leagues.
The average failure rate for all prospects taken in the five drafts I looked at is 50%, and if we look at failure rates by league of origin the OHL, WHL and Europe leagues were slightly above average in terms of prospect failure rates (see the table below). The two areas that produced lower than average failure rates where the QJMHL and U.S. College hockey programs, (although the number of players taken from college is so small we would need to do a larger study before drawing any conclusions). Perhaps, during this five year period scouts and general managers tended to underrate the Q and the NCAA and thus only the best players were taken.
League Type…Prospect Failure Rate (number of players in the survey)
QMJHL………..39% (18)
OHL…………....54% (35)
WHL…………...57% (37)
NCAA……….....33% (6)
Europe………...53% (30)
All Players……50% (133)
Interesting stuff but I would want to do a much larger study if I were a GM and trying to figure out the risks involved of drafting players. Perhaps if I get an entire week’s worth of insomnia…anyway, I hope this hasn’t put you to sleep yet.
Visiting prospect camp started me thinking about which prospects will succeed and which will fail. Baseball analysts have suggested that high school pitchers are more likely to fail to develop into major league players and therefore teams might be wise to think twice about drafting them high in the MLB draft. If we applied this approach to hockey, is one position more likely to produce busts than another? Are forwards more likely to flame out or is it defensemen who are more likely to become busts?
As I was thinking about this I began to suspect that among the three basic positions in hockey it would be defensemen who would produce the highest failure rate. Why? Evaluating the defensive abilities of a player are a bit more difficult than the scoring abilities of forwards and so I extrapolated that defensemen would probably produce more failed prospects.
But how to test this question? I was thinking about this question last night as I lay awake with a case of insomnia and finally I got up in the middle of the night to see if I could answer it to my own satisfaction since I wasn’t sleeping anyway. I pointed my browser to www.hockeydb.com and their wonderful collection of NHL draft lists. I decided to do a survey of all 1st round draft picks over a five year period (1995-1999) and calculate the failure rate of players by position.
What constitutes a “failed” prospect? For this little study any player who has played at least 250-300 NHL games and has a regular NHL job is a success and any prospect who does not meet that criteria is a failure at this point. (Note: goalies don’t get credit for games played when they are dressed as a backup, so I lowered the criteria for goalies and paid more attention to if they had NHL jobs or not.)
The 1st round picks between 1995-1999 add up to a total of 133 players, not a huge sample size but enough to give us some idea if any pattern exists. My quick and dirty late night survey of these five drafts found that my hunch is false. (It is not every day that I admit I’m wrong in print, so enjoy it.) Defensemen drafted in the 1st round between 1995-1999 were no more likely to fail to develop than forwards. The failure rate of defensemen was 49%, which was identical to the 49% failure rate for forwards. However, goalies did significantly worse than forwards and defensemen, with 62% of those drafted failing to attain NHL jobs as starters or backups. This high failure rate caught me by surprise a bit, but perhaps GMs should think twice about using a high draft pick on a netminder.
While I was going through these five recent drafts, I though it would also be interesting to consider where players were drafted from. Baseball analysts have compared college players to high school players, so I decided to compare draft picks from Juniors, NCAA, and European leagues.
The average failure rate for all prospects taken in the five drafts I looked at is 50%, and if we look at failure rates by league of origin the OHL, WHL and Europe leagues were slightly above average in terms of prospect failure rates (see the table below). The two areas that produced lower than average failure rates where the QJMHL and U.S. College hockey programs, (although the number of players taken from college is so small we would need to do a larger study before drawing any conclusions). Perhaps, during this five year period scouts and general managers tended to underrate the Q and the NCAA and thus only the best players were taken.
League Type…Prospect Failure Rate (number of players in the survey)
QMJHL………..39% (18)
OHL…………....54% (35)
WHL…………...57% (37)
NCAA……….....33% (6)
Europe………...53% (30)
All Players……50% (133)
Interesting stuff but I would want to do a much larger study if I were a GM and trying to figure out the risks involved of drafting players. Perhaps if I get an entire week’s worth of insomnia…anyway, I hope this hasn’t put you to sleep yet.
posted by The Falconer at
1:57 AM
1 Comments:
No reason really, other than I needed to get SOME sleep. I'ld like to go back and do this with a much larger number of draft picks at some point.
By
The Falconer, at 11:57 PM
Post a Comment
<< Home