DNHL Best Scouts

[shooter47]

Without the contraction draft the performance sample is too small.  To include contraction results skews the whole thing, IMO.  Saying someone is the "best scout" because they drafted Zach Parise or Pekka Rinne is a little silly.

Including the keeper picks skews this too, IMO. Teams which are horrible in real life (like the Oilers, Panthers, etc.) are going to look a whole lot better then teams that are good in real life like Chicago, Anaheim, Boston. Keeping a top 5 pick isn't that hard to do...

[Rob]

Including the keeper picks skews this too, IMO. Teams which are horrible in real life (like the Oilers, Panthers, etc.) are going to look a whole lot better then teams that are good in real life like Chicago, Anaheim, Boston. Keeping a top 5 pick isn't that hard to do...

 

[GypsieDeathBringer]

I left the keeper players in because you still have to pick or not pick the players.  Not ever great NHL player comes from the first 2 rounds.  "Good scouts" will probe the depth.  I felt like leaving the keepers out because the NHL Oilers suck which makes it easier for the DNHL Oilers to have good talent was as statistically unbalancing as leaving out the DNHL Panthers because they had so many top picks making it easier for them to have access to good talent.  If a team drafts whatever player, it was their choice no matter how easy or hard it was to get talent at that pick. 

There are around:
470 players to track if including everyone at the moment
222 players have actually scored fantasy points out of those 470
33 have scored fantasy points if you take away keepers and the original contraction draft.
 
33 data points for 20 teams didn't really paint a picture.  Which is why, statistically speaking, keepers are needed. 

[shooter47]

I left the keeper players in because you still have to pick or not pick the players.  Not ever great NHL player comes from the first 2 rounds.  "Good scouts" will probe the depth.  I felt like leaving the keepers out because the NHL Oilers suck which makes it easier for the DNHL Oilers to have good talent was as statistically unbalancing as leaving out the DNHL Panthers because they had so many top picks making it easier for them to have access to good talent.  If a team drafts whatever player, it was their choice no matter how easy or hard it was to get talent at that pick. 

There are around:
470 players to track if including everyone at the moment
222 players have actually scored fantasy points out of those 470
33 have scored fantasy points if you take away keepers and the original contraction draft.
 
33 data points for 20 teams didn't really paint a picture.  Which is why, statistically speaking, keepers are needed.

I am glad that you have taken the time to do this analysis for the league as I like leagues that have added content to them.

I understand that you added in the contraction draft and keepers to get more data points but I think that the contraction draft and keeper picks don't really tell you how good a scout a GM is. IMHO long term I don't think your analysis will give an accurate view of the best scouts in the league if keeper picks are part of it.  I don't agree with this statement at all, "I left the keeper players in because you still have to pick or not pick the players." This may be true for keepers that are picked in round 2 or later but this doesn't hold up when you discuss First round players. Deciding to "keep" a top 5 pick from the NHL draft for the Oilers is not that hard of a decision and doesn't tell me anything about how good a scout they are. They also have no input on which top 5 player they get in the draft, they get whoever the real Oilers pick.

I also don't agree that the Oilers keeping a top player is similar to the DNHL Panthers having a lot of top picks. If the Panthers trade or acquire top picks in the supplemental draft they still have to do scouting and pick a player. The pool of players they are picking from is huge (any prospect who is not owned and less then 40 NHL games and any drafted players that are not kept). This is not at all similar to the Oilers choosing to "keep" a top 5 pick from the NHL draft.

I don't think the solution to seeing who is the best scout is as simple a process as adding up the total points of all players that are drafted/kept. I think that a more accurate (but also a lot harder) way to determine the best scouts in DNHL would be to compare how the player drafted compares to where they were drafted at. In the MLB a lot of work has been done to determine the average WAR that can be expected in a certain draft spot (a sample of some of hte studies done can be found here: http://baseballanalysts.com/archives/2009/06/draft_picks_and.php). Now I'm not a full believer in WAR's ability to completely measure the value of a baseball player but this study does do a good way of allowing you to compare draft picks across the board. Now with only a couple of years in our league there really is no way of determining value of draft picks like this. However this type of analysis is more accurate IMO to analyze how good a scout a person is. If I had the first pick in the supplemental draft and you had the 20th pick in the first round and we both picked a player who scored 150 points in a year who is the better scout/GM? In your ranking system we are equal. In the other system you would be seen as a much better GM because you got a similar player to the one that I got but with a much lower pick in the draft.

In the end I hope you don't take this as an attack against your research and analysis as I am glad that you did it and it is beneficial to the league. IMHO I just don't think that what you are trying to measure and what you are actually measuring are the same things.

[snugerud]

I like to picture Jesus in a tuxedo t-shirt because it says I want to be formal, but I'm here to party.

[GypsieDeathBringer]

Sure, by no means is this a definitive picture of who the best scout is and there are better ways to measure it.  All those other ways require extensive comparisons on a level that I'm not willing to do.  This was just the easiest metric available based on our fantasy criteria, because I can't use just the NHL stats they don't have the same meaning to this league.  Corsi and possession numbers don't mean anything for us, but fantasy points very much does. 

I always like some good criticism to bring about a better measure, and I do very much like the idea of maybe I find the average fantasy points scored in for example round 1's, and then every pick after the #1 pick would get a bonus in that round.  So if the average points scored by picks in round 1 was 100,  and team A's #1 pick scores 150 they would have 50 points above average.  Team B's #20 pick scores 150, but they get a bonus of *something* I'd have to play with the numbers to see what makes sense, but lets say it is 1 bonus point per pick after #1 so team B would have a 69 points above average.  And if a team had multiple picks in a round I'd add them up and average it out so there would be no imbalance due to quantity of players.  I'd also have to compare averages of goalie to goalie, d-man to d-man, and then non-dman to non-dman within each round. 

All in all I still think this portrays who are the top scouts.  Keeper players account for a very small amount of the points for most teams, and the contraction draft even if you took out the veteran players the top teams on this list drafted very well.  The top 5 to 7 teams are still going to be the top 5 to 7 teams. 

If someone wants the best scout title based on this and taking away vets it is the Ducks by a gigantic margin.  I'll attach my spreadsheet that shows each teams picks and the Ducks will blow your mind.   

[snugerud]

I hear the Toronto Maple Leafs were considering hiring Tyler as a scout., because he was really super good at fantasy hockey pools 

[Tyler]

I hear the Toronto Maple Leafs were considering hiring Tyler as a scout., because he was really super good at fantasy hockey pools

They offered me a deal, but I didn't want to be associated with that franchise.

[norrya66]

I like to picture Jesus in a tuxedo t-shirt because it says I want to be formal, but I'm here to party.