Why is it rare for hockey players to start playing at a later age?

[Filthy Dangles]

The ice skating part, it's not something you can easily master picking up later on.

 

[LuckyDucky]

That is irrelevant to both this subtopic and of course the thread topic.

You don't get to suddenly be like "yeah but those aren't the good players". It doesn't matter! They played in the NBA, by far the highest most prestigious most elite basketball league in the world.

You're right. I was blending what others have said with our conversation. My apologies.

With that said, we are discussing an extremely small subset of players in the NBA. The questions I'd like to pose are: What makes guards in the NBA successful? How large is the talent pool for those guards?

 

[lorwood]

Skating. Hockey is a game of two sports played simultaneously. A hybrid of power and figure skating and the game of hockey. What I always tell folks with kids who want to play, have them skate, with private lessons if possible, for three years before ever picking up a stick. Youth hockey tryouts always come down to the same thing, who can skate the best. The better the skater the easier to coach and the quicker the game will come to them. Doesn't matter if your kid can rip a shot from the blue line if he or she is not in the top percentile on the skates he or she will not play on the top level teams.

 

[LuckyDucky]

Based on the reddit post quoted above, it would appear that the "startlingly high population" at the time of the study was... one person.

In any case, what's happening here on a statistical level is focusing on the extreme upper end of the height range (the upper 9%) to make the case that the talent pool as a whole is non-competitive.

But what about the other 91% of the league?

The great majority of NBA players stand between 6'3" and 6'8". Only about 3% of the general population falls into that range. In the NHL, the equivalent distribution occurs between 5'9" and 6'2", or about 50% of the general male population. From that baseline, the NBA is clearly operating from a smaller pool of candidates.

But, hockey is geographically limited. When we're talking about the general population, we're talking about the general population of Canada, plus certain demographics contained within certain US states, plus a handful of European countries. There are something like 30 million American men aged between 20 and 40. That talent pool alone is roughly six times larger than the pool of counterparts in Canada. It's not even necessary to point out that the ratio gets wildly out of proportion when we talk about European hockey vs European basketball, or the fact that basketball players come from places like Asia and South America and Africa where there is virtually zero hockey being played.

There's no metric for 'likelihood that dad pushed him to be a basketball player', but if that metric existed I have to think it would show that the overwhelming majority of American men over 6'3" have at least played basketball routinely and in many cases are pushed be competitive at it simply because of their height and any basic athleticism. A 6'8" man who has no basketball experience would be a curiosity. Is that the case with young men and hockey in Canada? Less so every year. Hockey is falling back as a youth sport of choice, as families increasingly opt for soccer (and, yes basketball). Within its pool, basketball operates much more efficiently than hockey at uncovering and then developing potential elite players. Access issues are at the root of it, with basketball being much less selective for financial and physical resources.

Now, let's go back to the extreme ends of the spectrum. Something like 10% of the NBA is between 5'9 and 6'3". The level of competition for those 10% of roster spots is completely insane, with gyms and parks all over the world full of 6' teenagers practicing their crossover with a dream. The NHL equivalent is guys between 5'4" and 5'8". Are there hockey rinks all over the world full of guys that size trying to make the NHL? Sort of, but let's be honest here... not really. A very small fraction of the male population with that body structure is seriously working toward a future in professional athletics, and that number drops to zero outside of about 5 regions on earth. Again, the basketball talent pool may begin with a disparity, but more than makes up for it with efficiencies and the sheer weight of numbers.

The bottom line is that the NBA is filling the vast majority (90%+) of its roster spots from a talent pool far greater than that of the NHL. It's impossible to put a precise number on how much greater, but intuitively I think we're talking something in the order of 10x globally. Only at the very upper extreme of their height distribution do you see the competitiveness drop sharply due to an emphasis on pure height for its own sake. Outside of that extreme, basketball is by far the more athletically selective sport.

Far better than I put it.

 

[majormajor]

You're right. I was blending what others have said with our conversation. My apologies.

With that said, we are discussing an extremely small subset of players in the NBA. The questions I'd like to pose are: What makes guards in the NBA successful? How large is the talent pool for those guards?

Yeah I'll admit it is very different for guards. Those really are probably some of the most athletic humans on the planet.

 

[Laodongxi]

Joe Mullen learned to skate on conventional roller skates (long before rollerblades), playing roller hockey in Manhattan. Very unique story. Not sure how many players have made the NHL learning on rollerblades before ever skating on ice?

But it's still skating from a young age.

Joe Mullen: Raised on roller hockey

 

[Fig]

You see a lot of NBA players who haven't started playing until they were like 16-18 or even 20; Siakam, Rodman, Hakeem, Mutumbo, Embiid, etc. Granted those are big dudes but I'm sure if you went down a little you might find guards that Steve Francis who didn't start until 13.

What about with hockey players? Can you really "fast-track" develop your skills? How often does a hockey player start at the age of 14 then out of nowhere becomes extremely talented in a 3-5 year span? Does hockey really require an insane amount of early development? Or can you skip that in the later ages and fast-track it?

The body mastery required for other sports like football, basketball and soccer are different than that of hockey.

Hockey starts with a steeper learning curve of unnatural movements and equipment like balancing on a literal razor's edge, yet also having to do so by adjusting the balance/tilt of your body to utilizing a stick that is always on only one side of your body and on/off the ice. Then, do it super fast while dealing with puck tracking, team systems, opponent systems and 1:1 talent.

If some of the other sports are kind of like body master gaps between walking and running while juggling on land, hockey is more like going from walking to juggling/working on a commercial fishing boat in rough seas. Exceedingly little of your previous body mastery is transferrable to that next skill. In fact, you at times have to fight your normal body mastery and recalibrate everything your body wants to do to get your sea legs under you.

 

[tarheelhockey]

We're in total agreement regarding the NBA players 6'2 and under (likely to be more at the athletic extreme than your typical NHLer) and regarding the guys 6'9 and above.

But you mention the 6'3 to 6'8 pool as 3% of the general population, whereas the NHL pulls from heights that are 50% of the general population. It's not quite as obvious to me that the number of developing basketball players is more than 17 times higher than the number of developing hockey players. It might be close, it might be more, it might be less. I'd want to see things tightened up on that point. Keep in mind that 75-80% of NBA players are American so the numbers should be re-weighted based on national origin, just like you would with the Canadian population being half of the NHL.

The IIHF has a survey of registered players, which gives us a better idea of the population of U-20 amateur hockey players. They find just under a million boys. This study finds 26 million basketball players in the U.S., and from their proportions it seems that 10 million or less would be U-20 boys. If you add another 20-25% for the international proportion of the NBA, then you can see we're inching towards a number close to but still short of a 17 million.

We'll likely never get to a clear-cut answer on this, but I tried to get some useful data on to estimate many people are in the professional basketball pipeline as compared to hockey.

I tried to find some kind of listing of total pro teams in each sport, but the numbers are harder to track down than one would think. A lot of leagues operate in a weird semi-pro space where it's hard to draw the line as to who's seriously competitive (who knew there was "pro" hockey in Turkmenistan?).

But broadly speaking, we can easily point to the size of the international organizations for a start: there are 58 IIHF member nations, versus 213 FIBA members. Bear in mind the IIHF membership list includes some questionable entries like Spain and South Africa, and I'm sure there are equivalents in FIBA.

I can't find an exact count on the number of professional leagues out there, but based on Wikipedia links I count 157 pro basketball leagues globally. As far as I can tell (and it really is difficult), there appear to be around 35 pro and semi-pro hockey leagues globally.

Trying to make this sort of estimate at the amateur level is so complicated that the answer would be near useless. But at least on a limited basis we can point to 144 men's hockey programs in the NCAA, compared to something like 400 men's basketball programs. In Canada those numbers are 34 and 48 in the U Sports program.

It gets harder when we throw junior hockey into the mix, because it kind of straddles the line between high school, college, and pro. There are 60 CHL teams, 150 Junior A teams, about 235 Junior B teams, 16 USHL teams, 23 NAHL teams, and an indeterminate number of Junior C teams -- a total of around 500 teams plus however many are in Junior C.

Per Statista, there are a little over 35000 players in boys' hockey in the USA, compared to 540,000 in the USA. I can't find comparable data for Canada and of course it's hopelessly complicated internationally. But those numbers would appear to imply about 1750 hockey teams and about 45,000 basketball teams.

Putting all of that in a box, it would appear that the total pipeline of talent leading to the NHL and NBA breaks down as follows:

-- Roughly 10-15 times as many basketball players engaged in basic recruitment and training on the youth level
-- By the time we get to college/junior age, the proportion shrinks to perhaps 5x as many players still engaged in a "pipeline" environment
-- At the pro level we see that 5x proportion again
-- At the international level (best of the best) it shrinks to more like 4x

What this means is that relative to hockey, basketball's promotional "pyramid" has a very wide base and a disproportionately narrow peak. If pro hockey's development pyramid was shaped like basketball's, there would only be about 10-15 pro hockey leagues in the world. That would leave us with only the NHL, KHL, AHL, ECHL, and the top tier of European leagues. We'd say goodbye to the tier 2 Euro leagues as well as the low minors like the SPHL and the Fed. The A-league in Slovakia or Denmark would be the lowest pro level still in existence.

I'm still mulling over your point of the efficiency of the selection for athleticism for each sport. I take your point that an athletic 6'6 kid would probably be steered towards basketball. And surely, they are a disproportionate number of those 10 million or so boys that play basketball. I'm not sure how to weigh that against the central issue that we're talking about - hard limits on the amount of potential athleticism within each subgroup. If you got every last scrap of athleticism out of 6'7 American boys, is that a lot or a little?

Again I think you're eating the pizza from the wrong end of the slice*.

Say there's this one guy who's 7'9" and athletic enough to play in the NBA. Now say that guy gets hit by a bus. What happens? His team isn't going out and finding an UN-athletic 7'9" guy to replace him. Instead they'll find an athletic 7'4" guy and move along with their season (mournfully).

Now let's say we have a very strange day and all the 7'+ guys get hit by buses simultaneously. There are none left in the talent pool. What happens? The teams replace them with 6'11" athletic guys, the average height drops, and they keep moving along with their season. It's not as though freakish height is a requirement to field a basketball team. It's simply a luxury that NBA teams enjoy when they are pulling from virtually the entire global population as a talent pool.

What I'm getting at here is that the NBA draws from the tallest 3% of the population because they can. There are, somehow, professionally-athletic guys at that height. It's not that the scouts are bound to the 3%, and have to ignore anyone shorter. It's that when you have half a million candidates for a roster spot, competitive forces are so intense that something as simple as a natural inch of wingspan makes the difference. If the 6'6" guy is equally skilled, there's nothing a 6'2" guy can do to close the gap. So you take the taller guy.

Now, this same height-driven effect does happen in hockey -- with goaltenders. At this point, a goalie has to be an absolute phenom to make the NHL at under 6'. Yet there are dozens of sub-6' skaters out there, some of whom aren't especially good players (all due respect but Nathan Gerbe has averaged 12 goals a season for over a decade at 5'4" and still gets contracts). What's the difference? There are only about 60 jobs for goalies. That's the equivalent of 3 teams of skaters. Imagine if there were 3 teams in the NHL, how hard it would be to get those jobs. It would be like making the Olympics for Canada or Russia. At some point, the competitive forces get so intense that the physical gaps become near-definitive. When that happens, the slightly taller (or heavier, or faster, or whatever) guy is going to come out on top. It takes someone truly special, a St. Louis or Fleury, to break through. That's what we see in the NBA, where you get the occasional Muggsy Bogues but it's an extreme rarity.

At the end of the day, the NBA just really is that much more competitive. The competition for any roster spot in that league is comparable to the competition for goalie spots in the NHL.




* not that there's anything wrong with that

 

[majormajor]

We'll likely never get to a clear-cut answer on this, but I tried to get some useful data on to estimate many people are in the professional basketball pipeline as compared to hockey.

I tried to find some kind of listing of total pro teams in each sport, but the numbers are harder to track down than one would think. A lot of leagues operate in a weird semi-pro space where it's hard to draw the line as to who's seriously competitive (who knew there was "pro" hockey in Turkmenistan?).

But broadly speaking, we can easily point to the size of the international organizations for a start: there are 58 IIHF member nations, versus 213 FIBA members. Bear in mind the IIHF membership list includes some questionable entries like Spain and South Africa, and I'm sure there are equivalents in FIBA.

I can't find an exact count on the number of professional leagues out there, but based on Wikipedia links I count 157 pro basketball leagues globally. As far as I can tell (and it really is difficult), there appear to be around 35 pro and semi-pro hockey leagues globally.

Trying to make this sort of estimate at the amateur level is so complicated that the answer would be near useless. But at least on a limited basis we can point to 144 men's hockey programs in the NCAA, compared to something like 400 men's basketball programs. In Canada those numbers are 34 and 48 in the U Sports program.

It gets harder when we throw junior hockey into the mix, because it kind of straddles the line between high school, college, and pro. There are 60 CHL teams, 150 Junior A teams, about 235 Junior B teams, 16 USHL teams, 23 NAHL teams, and an indeterminate number of Junior C teams -- a total of around 500 teams plus however many are in Junior C.

Per Statista, there are a little over 35000 players in boys' hockey in the USA, compared to 540,000 in the USA. I can't find comparable data for Canada and of course it's hopelessly complicated internationally. But those numbers would appear to imply about 1750 hockey teams and about 45,000 basketball teams.

Putting all of that in a box, it would appear that the total pipeline of talent leading to the NHL and NBA breaks down as follows:

-- Roughly 10-15 times as many basketball players engaged in basic recruitment and training on the youth level
-- By the time we get to college/junior age, the proportion shrinks to perhaps 5x as many players still engaged in a "pipeline" environment
-- At the pro level we see that 5x proportion again
-- At the international level (best of the best) it shrinks to more like 4x

What this means is that relative to hockey, basketball's promotional "pyramid" has a very wide base and a disproportionately narrow peak. If pro hockey's development pyramid was shaped like basketball's, there would only be about 10-15 pro hockey leagues in the world. That would leave us with only the NHL, KHL, AHL, ECHL, and the top tier of European leagues. We'd say goodbye to the tier 2 Euro leagues as well as the low minors like the SPHL and the Fed. The A-league in Slovakia or Denmark would be the lowest pro level still in existence.



Again I think you're eating the pizza from the wrong end of the slice*.

Say there's this one guy who's 7'9" and athletic enough to play in the NBA. Now say that guy gets hit by a bus. What happens? His team isn't going out and finding an UN-athletic 7'9" guy to replace him. Instead they'll find an athletic 7'4" guy and move along with their season (mournfully).

Now let's say we have a very strange day and all the 7'+ guys get hit by buses simultaneously. There are none left in the talent pool. What happens? The teams replace them with 6'11" athletic guys, the average height drops, and they keep moving along with their season. It's not as though freakish height is a requirement to field a basketball team. It's simply a luxury that NBA teams enjoy when they are pulling from virtually the entire global population as a talent pool.

What I'm getting at here is that the NBA draws from the tallest 3% of the population because they can. There are, somehow, professionally-athletic guys at that height. It's not that the scouts are bound to the 3%, and have to ignore anyone shorter. It's that when you have half a million candidates for a roster spot, competitive forces are so intense that something as simple as a natural inch of wingspan makes the difference. If the 6'6" guy is equally skilled, there's nothing a 6'2" guy can do to close the gap. So you take the taller guy.

Now, this same height-driven effect does happen in hockey -- with goaltenders. At this point, a goalie has to be an absolute phenom to make the NHL at under 6'. Yet there are dozens of sub-6' skaters out there, some of whom aren't especially good players (all due respect but Nathan Gerbe has averaged 12 goals a season for over a decade at 5'4" and still gets contracts). What's the difference? There are only about 60 jobs for goalies. That's the equivalent of 3 teams of skaters. Imagine if there were 3 teams in the NHL, how hard it would be to get those jobs. It would be like making the Olympics for Canada or Russia. At some point, the competitive forces get so intense that the physical gaps become near-definitive. When that happens, the slightly taller (or heavier, or faster, or whatever) guy is going to come out on top. It takes someone truly special, a St. Louis or Fleury, to break through. That's what we see in the NBA, where you get the occasional Muggsy Bogues but it's an extreme rarity.

At the end of the day, the NBA just really is that much more competitive. The competition for any roster spot in that league is comparable to the competition for goalie spots in the NHL.




* not that there's anything wrong with that

Yes, I've given up trying to find a precise answer to this, so I'll just make small comments.

- Goalies are getting taller and I don't think it has to do with the restrictive number of jobs available. There's just a bigger advantage to being tall. The game is increasingly chaotic and nobody can read plays that they can't see. Being a goalie is largely about being where pucks will just hit you.

- I wouldn't bother with trying to find an estimate of the international talent pool for basketball. Just multiply the U.S. talent pool by 5/4 and call it the NBA global pool. It will save you hours. (5/4 is the NBA player population divided by the number of Americans in the NBA).

- 35,000 boys in hockey in the USA, or 540,000? The IIHF number of registered players is closer to 250,000, by the way.

- I think we can leave talk of the seven footers aside because their role is different. The average NBA player is, what, 6'7? There is a much narrower pool at that population and that is worth factoring in. I'm precisely arguing that there is a height-athleticism trade off because the talent pool is smaller. Teams aren't opting for an equally athletic 6'8 player over a 6'2 player, just because they can, it's because the 6'8 guy is much better at basketball than a more athletic 6'2 guy! All else equal, you go for height, but you might also go for height when there's less athleticism, when the guy is slower and not as good at shooting baskets.

- Talent/athleticism looks like the right tail of a normal distribution / bell curve. The smaller the curve the less far to the right the tail goes. There aren't millions to choose from at that height, there aren't even millions of boys on the planet that end up 6'7. Maybe a few hundred thousand?

 

[lorwood]

Yes, I've given up trying to find a precise answer to this, so I'll just make small comments.

- Goalies are getting taller and I don't think it has to do with the restrictive number of jobs available. There's just a bigger advantage to being tall. The game is increasingly chaotic and nobody can read plays that they can't see. Being a goalie is largely about being where pucks will just hit you.

- I wouldn't bother with trying to find an estimate of the international talent pool for basketball. Just multiply the U.S. talent pool by 5/4 and call it the NBA global pool. It will save you hours. (5/4 is the NBA player population divided by the number of Americans in the NBA).

- 35,000 boys in hockey in the USA, or 540,000? The IIHF number of registered players is closer to 250,000, by the way.

- I think we can leave talk of the seven footers aside because their role is different. The average NBA player is, what, 6'7? There is a much narrower pool at that population and that is worth factoring in. I'm precisely arguing that there is a height-athleticism trade off because the talent pool is smaller. Teams aren't opting for an equally athletic 6'8 player over a 6'2 player, just because they can, it's because the 6'8 guy is much better at basketball than a more athletic 6'2 guy! All else equal, you go for height, but you might also go for height when there's less athleticism, when the guy is slower and not as good at shooting baskets.

- Talent/athleticism looks like the right tail of a normal distribution / bell curve. The smaller the curve the less far to the right the tail goes. There aren't millions to choose from at that height, there aren't even millions of boys on the planet that end up 6'7. Maybe a few hundred thousand?

Any program that is going to produce will be part of USA hockey. You may be able to get that stat there.