Jump to content
The Official Site of the Vancouver Canucks
Canucks Community

Drafting D-Men - A statistical analysis


TGokou

Recommended Posts

On 2/4/2018 at 10:26 AM, TGokou said:

**Update**

 

I decided I got a little bit more curious about the chances of drafting a #1 star D man. While I won't go into too many details here as it was only based on anecdotal evidence I noticed a really high correlation when drafting a LATE birthday D-man and getting your star D-man. I had to go a little further back in my analysis to get this information

 

Duncan Keith 2nd round July Bday

Shea Weber 2nd round August Bday

Dustin Byfuglien 8th round March Bday

Alex Edler 3rd round April Bday(While I doubt anyone would say he's a star defenseman he was pretty good for a while, and I threw him in here because he's Canucks property)

PK Subban 2nd round May Bday

Roman Josi 2nd round June Bday

Shayne Gostisbehere 3rd round April Bday

Erik Karlsson 1st round May Bday

Oliver-Ekman Larsson 1st round July Bday

 

So basically all the super-star D-men drafted in the 2nd round and later are ALL late birthdays. There is also a lot of players I cut off this list that I would consider really good players who were also late birthdays. Therefore in my original post where the chance of drafting a super-star D (or even a #1 D for that matter) of only being 4% can be greatly increased by ONLY drafting late birthday D-men in the second round onwards. In fact I would argue anecdotally there is a greater than 50% correlation of drafting really good D by selecting late birthday D-men...therefore increasing your odds from 4% to about 10%.

 

Most of these D-men I put here are 2nd round or later but assuming this applies to 1st round talent as well I would argue that Adam Boqvist definitely has #1 star potential based solely on his late b-day. Also based on Evan Bouchard's early birthday I would pass.

Interesting work.  I have one piece of criticism, as I don't think it has been reflected here.  Even if you end up agreeing with me, it doesn't invalidate the rest of your effort... keep up the good work :) 

 

My caution: be careful about using birth month.  In Canada, births are more likely in spring and especially summer than in colder months, so there are more candidates in those "late" birth months.  It is even more glaring a difference in typical European hockey countries (Swe, Fin, Rus, Cze).  I'm guessing weather/latitude has some influence here, since the differences are less noticeable in the US.  UN Data Source.

 

Before embracing or condemning birth month, I would recommend comparing draft success within months (or seasons).  It's possible you may find more draft failures also occur in late birth months, and the odds of success by month aren't all that different from any other.  As a spot check, I looked at a couple of the earlier years in your data range, and there were a number of failures (0-50 NHL games) with late birth months, so there are some grounds to dig deeper here if you think it worthwhile.  

 

Another factor involving birth month is how being late or early factors into one's development.  That can work either way though.  Kids a few months older than their peers can have higher confidence and develop better.  Or, on the other hand, they could appear to be better players playing with youth but struggle when it comes to the show.

Link to comment
Share on other sites

1 hour ago, Kragar said:

Interesting work.  I have one piece of criticism, as I don't think it has been reflected here.  Even if you end up agreeing with me, it doesn't invalidate the rest of your effort... keep up the good work :) 

 

My caution: be careful about using birth month.  In Canada, births are more likely in spring and especially summer than in colder months, so there are more candidates in those "late" birth months.  It is even more glaring a difference in typical European hockey countries (Swe, Fin, Rus, Cze).  I'm guessing weather/latitude has some influence here, since the differences are less noticeable in the US.  UN Data Source.

 

Before embracing or condemning birth month, I would recommend comparing draft success within months (or seasons).  It's possible you may find more draft failures also occur in late birth months, and the odds of success by month aren't all that different from any other.  As a spot check, I looked at a couple of the earlier years in your data range, and there were a number of failures (0-50 NHL games) with late birth months, so there are some grounds to dig deeper here if you think it worthwhile.  

 

Another factor involving birth month is how being late or early factors into one's development.  That can work either way though.  Kids a few months older than their peers can have higher confidence and develop better.  Or, on the other hand, they could appear to be better players playing with youth but struggle when it comes to the show.

Haha, for sure you have some valid points if I want to make this into a full blown experiment but I was only trying to highlight that there is a statistical difference. However in relation to your point, in a later post I did remove all players prior to March birthday and only counted the success rate for players with late birthdays. As for your point that there were players with late birthday failiures, that is definitely still going to happen but the chance of success does actually increase.

Link to comment
Share on other sites

On 2/4/2018 at 11:32 AM, TheGuardian_ said:

The Nucks will be diddled until the next CBA.

 

Pretty sure there will be a new system of drafting that will insure bottom teams get the players and still prevent tanking.

Maybe the league will allow "transfer" payments, then some teams could sell their draft picks. They can already in the form of taking on bad contracts and retaining salary. And the Canucks should go that route, find out who a team may want, lets say Guddy is acceptable, so the Nucks sign him to a 4 year 4.5 mil contract and pay him a 16 mil signing bonus, the team getting him gets the cap hit for the floor but pays him peanuts each year for his salary.

The bottom 7 teams should compete for the top seven draft spots and the 8 to 15 should compete for those spots...all playoff teams fall into their original spots.

 

For the 15th place team to bump the last place team from picking 1st is unfair to the bottom teams; therefore, make the bottom teams compete only for the bottom 7 spots.

This will discourage tanking somewhat.

Link to comment
Share on other sites

So what if Juolevi sat for a couple of games. As soon as he joins the Canucks, all of that stuff will be long forgotten. I think he'll turn out to be a reliable, composed player for us. Canucks still need help with their D prospects though.  Not enough future NHL D in our pool. Getting Tryamkin back, if he comes back, will be like an extra 1st rounder. Would love to add Dahlin, Boqvist or Bouchard to this list.

 

Juolevi

Tryamkin*

McEneny

Brisebois

Chatfield

Rathbone

Brassard

Link to comment
Share on other sites

On 2/6/2018 at 9:00 AM, NUCKER67 said:

So what if Juolevi sat for a couple of games. As soon as he joins the Canucks, all of that stuff will be long forgotten. I think he'll turn out to be a reliable, composed player for us. Canucks still need help with their D prospects though.  Not enough future NHL D in our pool. Getting Tryamkin back, if he comes back, will be like an extra 1st rounder. Would love to add Dahlin, Boqvist or Bouchard to this list.

 

Juolevi

Tryamkin*

McEneny

Brisebois

Chatfield

Rathbone

Brassard

I would agree except i would add Bode Wilde to the list

Link to comment
Share on other sites

  • 2 months later...
On 2018-02-03 at 10:38 PM, TGokou said:

Just be forewarned I am gonna get geeky with numbers here. If you don't feel like getting into the numbers just skip to the last two paragraphs for a summary.

 

After hearing a lot of talk about how 1st line D-men can be drafted in the 2nd round/3rd round I wanted to find out how statistically possible it is to do so. In the process I decided to do a deep dive into numbers from 2007-2013 (I didn't include statistics from 2014 onwards as I don't feel a lot of late-bloomers would have enough time to make an impact). Please note that my numbers can be highly subjective but I'll try to explain who I included and why. I feel that any D-man who can play a large number of games in the league and put up mediocre numbers is a highly valued commodity and thus were included in this analysis. Conversely I feel that 3rd line forwards can be drafted or signed in free agency and so if they did not put up a certain number of points they did not deserve to be included in my analysis. So without further ado here are the numbers (taken from eliteprospects)

 

From 2007-2013 there were 76 D-men chosen in 1st round, 71 in second, and 55 in 3rd.

Of those 1st round: 56%  2nd round: 21% 3rd round: 20% have gone on to have significant careers 

Of the total numbers, in 1st round: 16% 2nd round 4% 3rd round 4% are what we would call either superstar D-men or at the very least 1st pairing offensive D-men.

 

From 2007-2013 in 1st/2nd/3rd round there were 145/143/141 forwards drafted in each round respectively

Of those 1st round: 48% 2nd round: 12% 3rd round: 8% are what we would call 2nd-line or higher forwards.

 

At this point anyone can form their own opinion of what these numbers might mean but I will try to give you my view point.

Generally speaking it is way more advantageous to draft D-man in 2nd AND 3rd rounds as there is a higher percentage chance (20%) that you will draft a very good servicable D-man for your organization. You also never know because you might get lucky (4% chance) and draft yourself a number 1 D-man.

 

In regards to the 1st round, here is where it gets a little dicey. While the difference between drafting a D-man vs forward doesn't seem so different (56% vs 48%) It was obviously notable that most of the top 1st line forward talent was drafted in the top 10 picks. I tracked what percentage of D-men drafted in the top half of the 1st round went on to become 1st pairing D-men and that number was at about 33% with no extra correlation even if you drafted in the top 4. While my bias is towards drafting a D-man in the 1st round this year, the smart thing to do in my opinion is to draft a forward or find a way to trade down to get 2 1st rounders in the top 15. On the other hand, only having a 4% chance of drafting a 1st pairing D-man in 2nd or 3rd round is ultimately a matter of luck and you may have to go 10 drafts before you ever draft one, assuming you only draft D-men in 2nd and 3rd round. Therefore, sometimes you do have to ultimately take a chance and draft a D-man high to have a decent chance of scoring yourself a #1 D...but ultimately it is a game of numbers.

 

*EDIT: I added some additional stats in a post further down the page*

 

On 2018-02-04 at 10:26 AM, TGokou said:

**Update**

 

I decided I got a little bit more curious about the chances of drafting a #1 star D man. While I won't go into too many details here as it was only based on anecdotal evidence I noticed a really high correlation when drafting a LATE birthday D-man and getting your star D-man. I had to go a little further back in my analysis to get this information

 

Duncan Keith 2nd round July Bday

Shea Weber 2nd round August Bday

Dustin Byfuglien 8th round March Bday

Alex Edler 3rd round April Bday(While I doubt anyone would say he's a star defenseman he was pretty good for a while, and I threw him in here because he's Canucks property)

PK Subban 2nd round May Bday

Roman Josi 2nd round June Bday

Shayne Gostisbehere 3rd round April Bday

Erik Karlsson 1st round May Bday

Oliver-Ekman Larsson 1st round July Bday

 

So basically all the super-star D-men drafted in the 2nd round and later are ALL late birthdays. There is also a lot of players I cut off this list that I would consider really good players who were also late birthdays. Therefore in my original post where the chance of drafting a super-star D (or even a #1 D for that matter) of only being 4% can be greatly increased by ONLY drafting late birthday D-men in the second round onwards. In fact I would argue anecdotally there is a greater than 50% correlation of drafting really good D by selecting late birthday D-men...therefore increasing your odds from 4% to about 10%.

 

Most of these D-men I put here are 2nd round or later but assuming this applies to 1st round talent as well I would argue that Adam Boqvist definitely has #1 star potential based solely on his late b-day. Also based on Evan Bouchard's early birthday I would pass.

 

 

 

Bump worthy, especially considering where the Canucks are drafting and who is available in those positions. 

 

Good work, OP. 

Link to comment
Share on other sites

On 2/3/2018 at 10:38 PM, TGokou said:

Just be forewarned I am gonna get geeky with numbers here. If you don't feel like getting into the numbers just skip to the last two paragraphs for a summary.

 

After hearing a lot of talk about how 1st line D-men can be drafted in the 2nd round/3rd round I wanted to find out how statistically possible it is to do so. In the process I decided to do a deep dive into numbers from 2007-2013 (I didn't include statistics from 2014 onwards as I don't feel a lot of late-bloomers would have enough time to make an impact). Please note that my numbers can be highly subjective but I'll try to explain who I included and why. I feel that any D-man who can play a large number of games in the league and put up mediocre numbers is a highly valued commodity and thus were included in this analysis. Conversely I feel that 3rd line forwards can be drafted or signed in free agency and so if they did not put up a certain number of points they did not deserve to be included in my analysis. So without further ado here are the numbers (taken from eliteprospects)

 

From 2007-2013 there were 76 D-men chosen in 1st round, 71 in second, and 55 in 3rd.

Of those 1st round: 56%  2nd round: 21% 3rd round: 20% have gone on to have significant careers 

Of the total numbers, in 1st round: 16% 2nd round 4% 3rd round 4% are what we would call either superstar D-men or at the very least 1st pairing offensive D-men.

 

From 2007-2013 in 1st/2nd/3rd round there were 145/143/141 forwards drafted in each round respectively

Of those 1st round: 48% 2nd round: 12% 3rd round: 8% are what we would call 2nd-line or higher forwards.

 

At this point anyone can form their own opinion of what these numbers might mean but I will try to give you my view point.

Generally speaking it is way more advantageous to draft D-man in 2nd AND 3rd rounds as there is a higher percentage chance (20%) that you will draft a very good servicable D-man for your organization. You also never know because you might get lucky (4% chance) and draft yourself a number 1 D-man.

 

In regards to the 1st round, here is where it gets a little dicey. While the difference between drafting a D-man vs forward doesn't seem so different (56% vs 48%) It was obviously notable that most of the top 1st line forward talent was drafted in the top 10 picks. I tracked what percentage of D-men drafted in the top half of the 1st round went on to become 1st pairing D-men and that number was at about 33% with no extra correlation even if you drafted in the top 4. While my bias is towards drafting a D-man in the 1st round this year, the smart thing to do in my opinion is to draft a forward or find a way to trade down to get 2 1st rounders in the top 15. On the other hand, only having a 4% chance of drafting a 1st pairing D-man in 2nd or 3rd round is ultimately a matter of luck and you may have to go 10 drafts before you ever draft one, assuming you only draft D-men in 2nd and 3rd round. Therefore, sometimes you do have to ultimately take a chance and draft a D-man high to have a decent chance of scoring yourself a #1 D...but ultimately it is a game of numbers.

 

*EDIT: I added some additional stats in a post further down the page*

Please:  phone/email/text/carrier pigeon JB and TL with this analysis.  It's really good work, and those two NEED TO KNOW!

Link to comment
Share on other sites

On 2/4/2018 at 2:48 PM, bloodycanuckleheads said:

Juolevi is playing in a 5th tier league right now (Liiga) - where he just recently got healthy-scratched multiple times (and when he came back, he was on the bottom-pairing).  And, that's a leaugue filled with AHL rejects.  The chances that he becomes a 1st pairing NHL d-man are not very high at the moment.  He's looking nowhere near as good as Sergachev right now, he's looking more like another Virtanen.

tenor.gif

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...