I decided to look at Poile's track record in selecting players in the 1st Round of the NHL Draft. To do this I decided we have to evaluate every player's success in the NHL that was drafted since the Predator's came into the league in 1998. Doing this to more than Round 1 would take way too much time.
Please keep in mind I am evaluating how well a team identifies "good" players. Not necessarily how well the player did for the team that drafted him.
Example: A player like Scott Hartnell which was traded to the Flyers after a few years with the Predators would reflect 100% on the Predators' ability to identify talent, even though his best seasons have been with the Flyers. He would reflect absolutely 0% on the Flyers.
A quick note:
This all comes from a man (me) that's been educated in Stats (College Stats 1 & 2, and some classes for my job), but not so good at them to think this is the best evaluation possible. However, since I have kept it simple for everyone's ability to understand, this works well.
I have put a summary of my process and talked about reasoning/assumptions used at the bottom of the post.
I have also included at the bottom the charts that show how each draft position in Round 1 has fared over these years.
In the Below Chart:
Avg P/Game 98-2009 is: Avg total point production of players drafted in the draft positions that team had during this time period.
Example: Detroit had Picks 25, 29, 19, 27, and 30. Players drafted at those positions totaled 1.09. Then divide by the 5 picks and get .218..rounds to .22.
Actual P/Game 98-2009 is: Actual p/game production of the players actually picked in those positions.
Example: The players Detroit picked at 25, 29, 19, 27, and 30 actually produced 1.21. Divide by the 5 picks and you get .242 which rounds to .24.
'09 Difference is: Difference between Column 2 and Column 1.
|Team||Avg P/Game 98-2009||Actual P/Game 98-2009||09 Difference|
What does this tell you? In short, it says the Philadelphia Flyers has fared the best and the New York Rangers has fared the worst.
What does it say about the Predators?
1) With his 1st Round Picks, Poile has managed to be just a little below what the league average says he should be. This is reflected Nashville's -.03 in the "'09 Difference" column.
2) Poile has positioned Nashville as the 20th best team in the league in picking players in the 1st round.
I guess it's possible that any given team could have under performed due to the players available in the draft, but over a 12 year period I would expect that to balance out.
Dang, I had another point I wanted to make, but can't remember it. If I do I'll be sure to make it.
Summary of How I've Done This:
As stated, I wanted this to be simple. So, I threw out the possibility of using advanced stats. I settled on pulling Games Played and Points and then generating a Points Per Game output for each player. After looking at this number and the stats generated from this number I felt it was a pretty decent reflection of how well a player has performed.
You could argue this reflects poorly on goalies and defenseman and you'd be right.
I have 3 points on why this does not concern me.
1) The first round is about identifying top level talent. Offensive ability drives a player's draft position more than anything else on teams' draft boards and we all know most of the top level talent comes from the 1st round. You will see a defenseman like Seth Jones rated highly in the draft, but he is there because he can give good offensive output from a defensive position while maintaining a focus on defense. Not because he sucks on offense and is stellar on defense.
2) All teams will likely take all positions over time in the 1st round. We'd all agree if you are good at selecting goalies then you don't need to take one high very often. Similar logic could be used for defensemen since you only use 6 in a game compared to 12 forwards. Essentially, if a team is having to take more than the average then it is to make up for poor drafting and it should get reflected in this analysis.
3) If a team takes a defenseman high and he pans out, then he will likely only reflect negatively on this analysis if he was taken in the Top 3. The average output of the 3rd overall pick is about .56 points per game. .51 is the median. Pick 4 is .46 and .53 points per game.
Those 3 points got too long.
As I was assembling this I realized up to 2010 had pretty decent data. To be on the safe side, I left 2010 out as well and used only '98 - 2009. I did gather and analyze '10-13, but it seemed wrong to include any of it. Players drafted in 2010 or later are at too many different stages of development.
I decided to look at Mean (Average) and Median to analyze the data. This keeps it simple without over thinking it, and the results passed the "sniff" test.
I then narrowed that down to Mean. I did this because the information essentially paints the same picture for each draft position. Which is good because it saved me a couple more hours of working on this.
Below are the charts on Draft Positions. Chart 1 is the Mean. Chart 2 is the Median. SV% doesn't tell much due to small numbers taken at goalie.