Hockey Analytics...A Closer Look at the Relative Age Effect

[Blackbeard]

Scroll down to "A Closer Look at the Relative Age Effect" and click on the link just below the text for that heading for an intriguing albeit somewhat heavy read.

Could this "invisible" factor affect recruiting/scholarships in the women's game?


http://hockeyanalytics.com/

[Ted Knight]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

Scroll down to "A Closer Look at the Relative Age Effect" and click on the link just below the text for that heading for an intriguing albeit somewhat heavy read.

Could this "invisible" factor affect recruiting/scholarships in the women's game?


http://hockeyanalytics.com/

###A report found that 36 per cent of players drafted by NHL teams between 1980 and 2007 were born in the first quarter of those years, or from January to March, compared to 14.5 per cent of draftees who were born in the fourth quarter.

[Trillium]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

Scroll down to "A Closer Look at the Relative Age Effect" and click on the link just below the text for that heading for an intriguing albeit somewhat heavy read.

Could this "invisible" factor affect recruiting/scholarships in the women's game?


http://hockeyanalytics.com/

Of course it does. At least in Canada, it's always been a well-known fact that rep hockey teams at every level tend to skew fairly heavily towards kids with birthdays in the front half of the year.

[Rightnut]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

Of course it does. At least in Canada, it's always been a well-known fact that rep hockey teams at every level tend to skew fairly heavily towards kids with birthdays in the front half of the year.

I'm not so sure with regard to women's hockey. You can look at it a couple of ways. The immediate reaction is to say early birthdays are older and thus more advanced at younger ages and thus play more and get better relative to later birthdays. That is probably true very early but I would argue that over time as girls age, that difference becomes much less, especially since girls mature earlier than boys - there is essentially more time for the late birthday to catch up if everyone has "matured" around 14-15. You have several years of a level playing field from a physical standpoint. This argument also ignores the fact that late birthday women are likely to be a grade behind their same birthyear teammates and thus really competing for scholarships with the next birth year. So, in that sense, a late birthday may be an advantage.

[PuckRush]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

I'm not so sure with regard to women's hockey. You can look at it a couple of ways. The immediate reaction is to say early birthdays are older and thus more advanced at younger ages and thus play more and get better relative to later birthdays. That is probably true very early but I would argue that over time as girls age, that difference becomes much less, especially since girls mature earlier than boys - there is essentially more time for the late birthday to catch up if everyone has "matured" around 14-15. You have several years of a level playing field from a physical standpoint. This argument also ignores the fact that late birthday women are likely to be a grade behind their same birthyear teammates and thus really competing for scholarships with the next birth year. So, in that sense, a late birthday may be an advantage.

The "research" referred to below is detailed in depth in the book "Outliers". Explains (in a way that's difficult to do here) why birth month matters for hockey success.

Here is excerpt about hockey from Wiki:

The book begins with the observation that a disproportionate number of elite Canadian hockey players are born in the first few months of the calendar year. The reason is that since youth hockey leagues determine eligibility by calendar year, children born on January 1 play in the same league as those born on December 31 in the same year. Because children born earlier in the year are bigger and more mature than their younger competitors, and they are often identified as better athletes, this leads to extra coaching and a higher likelihood of being selected for elite hockey leagues. This phenomenon in which "the rich get richer and the poor get poorer" is dubbed "accumulative advantage" by Gladwell, while sociologist Robert K. Merton calls it "the Matthew Effect", named after a biblical verse in the Gospel of Matthew: "For unto everyone that hath shall be given, and he shall have abundance. But from him that hath not shall be taken away even that which he hath."[7] Outliers asserts that success depends on the idiosyncrasies of the selection process used to identify talent just as much as it does on the athletes' natural abilities.[7]

[Rightnut]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

The "research" referred to below is detailed in depth in the book "Outliers". Explains (in a way that's difficult to do here) why birth month matters for hockey success.

Here is excerpt about hockey from Wiki:

The book begins with the observation that a disproportionate number of elite Canadian hockey players are born in the first few months of the calendar year. The reason is that since youth hockey leagues determine eligibility by calendar year, children born on January 1 play in the same league as those born on December 31 in the same year. Because children born earlier in the year are bigger and more mature than their younger competitors, and they are often identified as better athletes, this leads to extra coaching and a higher likelihood of being selected for elite hockey leagues. This phenomenon in which "the rich get richer and the poor get poorer" is dubbed "accumulative advantage" by Gladwell, while sociologist Robert K. Merton calls it "the Matthew Effect", named after a biblical verse in the Gospel of Matthew: "For unto everyone that hath shall be given, and he shall have abundance. But from him that hath not shall be taken away even that which he hath."[7] Outliers asserts that success depends on the idiosyncrasies of the selection process used to identify talent just as much as it does on the athletes' natural abilities.[7]

I think we are all pretty well aware of "Outliers" and that line of thought. My point, without any empirical research, is that I'm not sure, nor have I seen research, that this phenomenon applies to women's hockey. Also, because women's hockey is heavily impacted by the grade of the person in school, there is another variable thrown into it. The school grade of men is of little impact because most go to juniors and don't get to college until they are 20 yrs old. As of now, women are much more linear in their schooling meaning they graduate high school and go directly to college. So, if you are competing against a younger birth year because you are in the same grade as that younger birthyear, there may be an advantage similar to the "outlier" effect, of being a late birthday.

[Trillium]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

I'm not so sure with regard to women's hockey. You can look at it a couple of ways. The immediate reaction is to say early birthdays are older and thus more advanced at younger ages and thus play more and get better relative to later birthdays. That is probably true very early but I would argue that over time as girls age, that difference becomes much less, especially since girls mature earlier than boys - there is essentially more time for the late birthday to catch up if everyone has "matured" around 14-15. You have several years of a level playing field from a physical standpoint. This argument also ignores the fact that late birthday women are likely to be a grade behind their same birthyear teammates and thus really competing for scholarships with the next birth year. So, in that sense, a late birthday may be an advantage.

Having been a long time league/association exec, I can assure you that this is true in women's hockey through all age groups and teams up to the university level. I also just checked-- on my daughter's college team last season, more than 1/2 of the players had birthdays in the 1st four months of the year.

While the late birthday players may in theory catch up as everyone matures by around 14-15, this ignores that fact that all along, they have not faced the same level of competition and challenge to achieve their potential either, since they were not on the top teams in the age group at the younger age levels. They would have been playing all along at the next level down.

Furthermore, in Canada at least, late birthday kids are not a grade behind their teammates. Those born on January 1 through to December 31 are always in the same grade, except in the relatively rare cases where their parents opted on their own to hold the late birthday kids back a year.

[Rightnut]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

Having been a long time league/association exec, I can assure you that this is true in women's hockey through all age groups and teams up to the university level. I also just checked-- on my daughter's college team last season, more than 1/2 of the players had birthdays in the 1st four months of the year.

While the late birthday players may in theory catch up as everyone matures by around 14-15, this ignores that fact that all along, they have not faced the same level of competition and challenge to achieve their potential either, since they were not on the top teams in the age group at the younger age levels. They would have been playing all along at the next level down.

Furthermore, in Canada at least, late birthday kids are not a grade behind their teammates. Those born on January 1 through to December 31 are always in the same grade, except in the relatively rare cases where their parents opted on their own to hold the late birthday kids back a year.

The numbers may bear out that its true with women but I haven't seen any research on it. Further, since the larger percentage of women's college hockey players are American or go to high school in the US, the grade issue is significant I think. Since most states have a September cutoff, a good 25% of girls in any birthyear are potentially a grade behind and thus competing against younger birth years for the same scholarship. I could be wrong but it makes sense as an added variable and other than anecdotal evidence like looking at your daughter's team, I haven't seen the same research for women as Outliers offers for men's hockey.

[Trillium]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

The numbers may bear out that its true with women but I haven't seen any research on it. Further, since the larger percentage of women's college hockey players are American or go to high school in the US, the grade issue is significant I think. Since most states have a September cutoff, a good 25% of girls in any birthyear are potentially a grade behind and thus competing against younger birth years for the same scholarship. I could be wrong but it makes sense as an added variable and other than anecdotal evidence like looking at your daughter's team, I haven't seen the same research for women as Outliers offers for men's hockey.

One way of testing it would be to look at the rosters of USA Hockey U18 /U22 teams would it not? Don't these often include birthdays?

[Rightnut]

One way of testing it would be to look at the rosters of USA Hockey U18 /U22 teams would it not? Don't these often include birthdays?

I assume so. That doesn't necessarily get to the issue of women's college players. I suppose if someone simply did a survey of all the current rosters it would give us an idea whether the theory applies to women's college hockey. Not it!

[ARM]

Re: Hockey Analytics...A Closer Look at the Relative Age Effect

Furthermore, in Canada at least, late birthday kids are not a grade behind their teammates. Those born on January 1 through to December 31 are always in the same grade, except in the relatively rare cases where their parents opted on their own to hold the late birthday kids back a year.

That isn't how it works in the U.S. The exact date varies by state, but in general, kids start kindergarten at age five. Most states require turning five by Sep. 1, Meaning most September, October, November, and December birthdays started school with the next birth year.