Realty One


Thursday, February 28, 2013


Parents worried about whether their child has the goods to make it into the NHL might now have another seemingly arbitrary factor working against them - the time of year their hockey prodigy was born.

A study, published Wednesday in the online science journal PLOS ONE, suggests that the NHL is guilty of an age bias because it weighs its draft selections more heavily in favour of players born earlier in the year.

The 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.

Rob Deaner, one of the study's authors, said it has been widely known that a so-called selection bias exists in various sports and educational pursuits, but it hasn't been demonstrated before in the NHL.

"It's never been shown that people are systematically underestimating the ability of the younger players,'' said Deaner, an associate professor of psychology at the Grand Valley State University in Michigan.

"We found the teams are consistently underestimating how good the guys are that are born from July to December.''

The authors looked at about 2,700 Canadian players drafted by NHL teams over a 27-year period, finding that those born in the third and fourth quarters were drafted more than 40 slots later than their productivity warranted.

Deaner said the selection bias can start at an early age, when a 12-year-old player born between January and March will be streamed into an elite level of hockey, ahead of his peers born later in the year.

He said younger players may be just as good, but might not be as big or as skilled and will likely end up in a lower league.

"On average, the players who are drafted in a lower league might turn out to be better and those players might more often be relatively younger,'' he said.

A spokesman for the NHL could not be reached for comment.

The authors contend that the strategy might actually work against a team's success because they found that players born later in the year and drafted later actually had more productive hockey careers.

Deaner said the study showed that men drafted in the second half of the year were about twice as likely to have successful careers in the NHL - reaching benchmarks like 400 games played or 200 points scored - than those born earlier in the year.

"If the team wasn't making this mistake, they probably would have been more successful,'' he said. "The guys born in the first part of the year are much more likely to be busts.''

© 2013 The Canadian Pr


Johnny Knowitall said...

I'm sure that NHL teams are passing on better players just to take guys that are born earlier. What a laugh riot, there are billions of dollars in play here, they draft the best at the time. Of course when you look back you will say that the guys drafted later over achieved and more guys in the early rounds are busts, there is way more expectations so it's easier to bust! A first rounder can be a bust beacause they couldnt be the goal scorer that they were in junior , maybe they dont wanna change their game or dont know how in a short period of time so they are looked down upon and sent to the minors (that will suck the life out of most guys that have always been studs and now have to try struggle and have their heads played with). The guys drafted later might start in the AHL and have lots of time to work on their game and their strength, then one day they get a call and they get the role they are used to , so they are looked upon as a guy who made it out of no where. These numbers are a joke, I bet Rob Deaner and all the other author's involved are born at the end of the year and are all small men with no skills except to be able to find a reason why they don't stack up in life.......juss sayin

Anonymous said...

I call B.S. on this one. The guys that tend to go bust are the guys that develop their entire time with a chip on their shoulder, make excuses, want to get special treatment, complain about the situation they've been put in, and their parents figure they've been owed something. The guys that tend to do well are the guys that shut their mouths, listen, play defensively, play a 200 foot game, work hard, and strive to improve. The guys that go bust are also guys that never should have been picked in the first place.

Keep in mind when these guys are drafted there are 700 players in the NHL that aren't to keen on giving up their jobs as these guys are coming into the camps.

Supposedly Luc Robitaille was told by scouts he couldn't skate until his Dad told him he isn't as smart as a Hockey Scout to make that judgement, but he does notice whenever there is a loose puck Luc sure seemed to always get there. Theoren Fleury was drafted in the late rounds, and Al MacNeil said people would say, "well who knew he'd turn out to be the player he became." MacNeil said, "Theo knew."

Y'er Welcome

Anonymous said...

Read Malcolm Galdwell's "Outliers" re: when people are born and the opportunities that this presents...interesting.

Mick Panko

Anonymous said...

Scouts are generally too fricking stupid to even consider anything but their own futures and political gai in the game to be thinking about this.

Mike said...

"He said younger players may be just as good, but might not be as big or as skilled and will likely end up in a lower league."

That quote is off. How could they be as good if they are not as skilled?

It is obvious that if you have a player born in January and a player born in December on the same team that the player born in January is at an advantage. He is almost a full year older - this makes a difference even for 17 and 18 year olds when drafted.

Perhaps the younger players have the same potential but when they are placed up against players that are that much older they are not as good.

I have planned with my wife to have our children in the first 5 months of the year. Not so that they have a better chance in making the NHL but it is proven that on average people born January to May are more successful in school and athletics.

Anonymous said...

No wonder I didn't make it!! Lol

Anonymous said...

There is a reason that the NHL draft cutoff is September 15 and that is to allow players born late in the year more time to develope. There is an advantage to this also. Let's use the 2013 draft for example. A player born on Sept 17, 2012 would be eligible, as would a birth date of Sept 14 2013, a calenders year difference. So where you put a cutoff also can create advantages for those disadvantaged by a situation.


Anonymous said...

Doesn't matter where you put the cutoff the same problem exists. This sounds reasonable to me. Some of you seem to think the study is saying scouts are looking at the birthdays and they are being biased by birthdays early in the year. This isn't the case. Those born later in the year are not performing as we'll and thus they aren't rated as highly by comparison to those born early in the year.
You can't look at a small sample and come to any real conclusions but with a big enough sample the trend is there.

Anonymous said...

You can't fix this problem if you are drafting players a year at a time. The only way is to say have two drafts per year so they players can only be at most 6 months apart rather than a year a part.
Smart team should be looking at those players born later on the year who were overlooked in the draft.

Anonymous said...

You mean the same teams that consistantly draft well year after year after year in what crap teams refer to as a supposed crap shoot?

Johnny Knowitall said...

This article is a waste of time, and completely rediculous. Let's take a look at the this years top three picks and a few other guys you may remember from being selected first overall, then you can decide yourself on the merit of this article,

Seth Jones - Oct. 3rd
Nathan McCkinnon - Sept. 1st
Johnathan Druin - Mar. 8th
Nail Yakupov - Sept. 6th
Nugent-Hopkins - Apr. 12th
Taylor Hall - Nov. 14th
John Tavares - Sept. 20th
Stamkos - Feb. 7th
Patrick Kane - Nov. 19th
Alex Ovechkin - Sept. 17th
Marc Andre Fleury - Nov. 28th
Rick Nash - June 16th
Kovalchuk - Apr. 15th
Lecavalier - Apr. 21st
and two more fella's you may have heard of
Gretzky - Jan. 26th
Mario - Oct. 5th

If the scouts are picking guys according to when they are born then why do they keep drafting all of these late birthdays first?
and also how did Patrik Stefan not turn out? he was born sept. 16th, or Scott Scissons who was born Oct. 29th??? the list can go forever, and you can turn any angle you want with selective research.

Danny said...

Like Mick said, Gladwell detailed this in his book a few years ago:

Anonymous said...

Brilliant Johnny knowitall. I can see that you are a proponent of the scientific method...the way you prove a case is through isolated examples and not large samples that can identify trends.

Rob said...

This is Rob Deaner, the lead author on the study. Thanks for the interest in our research. I’ll try to clarify a few things here:

First, we don’t think NHL teams are deliberately selecting players based on their birthdays. We think they are drafting based on something ASSOCIATED with birthdays – there’s a big difference. For instance, teams might select far more players from the CHL leagues than the CJHL, yet, on average, the small number of CJHL players that are drafted might perform better in the NHL. And it could be that more of the relatively younger players play in the CJHL than in the CHL. This is because they would have been at a disadvantage several years earlier when players were drafted or selected for those junior teams. I should add that we haven’t tested this reasoning yet, but we will. Maybe it’s the underdog effect, meaning that relatively younger players have better work habits and this is crucial for success, yet scouts overlook work habits. This hypothesis is more difficult to test. But one of these hypotheses (or something similar that we haven’t quite thought of yet) must be correct. We know one of them must be correct because the evidence is overwhelming that drafting is indeed biased.

The average draft position of the draftees born in the third quarter of the year and those born in the first quarter of the year is almost identical. Yet those third quarter players are twice as likely to have successful careers. It’s not somehow because the third quarter players are absolutely younger. We know that absolute age has nothing to do with it because fourth quarter players are pushed ahead to the following draft due to the Sept 15 cut-off. So they are absolutely older than the first quarter players they are drafted with. Yet fourth quarter draftees are also twice as likely to be successful in the NHL. So relative age-- birth month, not birth year-- is a huge predictor of how successful a draftee will be.

Of course, relative age is only aspect of player development, and many individual players will deviate from the pattern. However, some cherry-picked exceptions (esp. only players who are in the NHL rather than those didn’t make it!) do not disprove our pattern any more than one snowstorm in June shows that summer months receive more snowfall than winter months do. And we showed in our paper that the underperformance of first quarter born players even occurred even when we only looked at first round draft picks.

If you’d like to see the data, just email me and I will send it to you. You can also download it yourself very easily from and similar websites. The effect is not subtle – it’s big, it’s been happening since 1980, and any open-minded person with a bit of statistical experience can verify it for themselves.

If you haven’t seen the graphs in our paper, I’d suggest having a look.

A few personal things.
I was born in January but never played ice hockey. I don’t believe my co-authors, Lowen or Cobley, played iced hockey and I don’t know their birthdays.
I disagree with Malcolm Gladwell’s point of view on many issues in the book Outliers, and I certainly didn’t intend to study relative age effects in hockey. I discovered the pattern when doing another study (about aggression in hockey). The relative age effect was too large and interesting not to pursue. We didn’t create the effect; we discovered it by luck, although anyone else could have too.

Thanks again for your interest and suggestions.