PART 2: A Look at College Hockey Demographics

Welcome back!

I got some great feedback on my last College Hockey Demographics post, so I felt some extra enthusiasm as I put this second post together.  I really did my best to incorporate as many of the great suggestions I got into this as possible, but apologize if I left some things out.

Time permitting, I hope to make this an annual thing so that we can really start to make some good comparisons from season-to-season.

Anyways, if you haven't read Part 1: A Look at College Hockey Demographics, I suggest you give it a read.  In Part 1 I took a look at the total number of players playing college hockey this season from every country, state\province, and city.  I also did a brief break-down of where a majority of the points scored by college hockey players this season have come from geographically.

In Part 2 I hope to take things a bit further.  Based on some suggestions I got, I will be breaking those listings down even further to show the results per capita to give us a more normalized look at the data.  I was also able to dig up some stats that Inside College Hockey (INCH) did on player demographics back in the 2004-05 season, so I was able to do a really good job of providing some insight into the changes in college hockey demographics over the last 10 years.

I also wanted to dig a little deeper into where the scoring has been coming from because I have been very curious as to which states\provinces\countries have been producing the top scorers at the college level.  Probably won't tell us a lot to look at the data for 1\2 of a season, but as I continue to put these together perhaps some trends will emerge.  Hoping for some good insight there eventually.

To dig a little deeper into the player side of things, and to perhaps get some more insight into college hockey recruiting trends, I also did a player height\weight\and age analysis (both for the NCAA hockey as a whole and a team-by-team break-down).  Some pretty interesting stuff to look at here.

Finally, I decided to take a look at which leagues college hockey players are coming from pre-NCAA, and did a team-by-team break-down for some of the biggest college hockey filler leagues to determine which teams in which leagues are currently producing the most NCAA hockey players.

With that, let's get started!


Players by State\Province and Country Per Capita


The first thing I wanted to do was elaborate a little bit more on the demographics I listed in Part 1 of my demographics.  The point was brought up (and it was a good one) that these demographics would be a bit more meaningful if they were normalized by population.

If you look at the charts below, the rankings on the far left include the general rankings for number of college hockey players produced.  The first ranking is the overall ranking, and the second ranking is the ranking within Canada, Europe, and the United States specifically.

The group of rankings in the middle are the per capita rankings.  First the overall per capita ranking, then the ranking within Canada, Europe, and the United States specifically.

Finally, the rankings on the right (for goals and points) displays where each country, state, or province ranks in relation to goals and points produced per capita. The "+\-" fields to the right display the difference ranking for per capita goals and points produced in relation to their per capita ranking for number of players produced.

For example, in the chart directly below, New Brunswick ranks 35th overall per capita in players produced (having produced 1 player for every 751,171 people in New Brunswick), but they rank 62nd in terms of goals and points produced per capita.  The difference is 27 places in the rankings, or -27.

So essentially high numbers (Lativa is one) tell us that even though a region might not be producing a lot of players per capita, they are producing their share of players who have been scoring.  Vice versa, a low number (looking at you Vermont) tells us that even though a region may have ranked high in players produced per capita, those players aren't doing a lot of scoring in NCAA hockey this season.

Per capita NCAA hockey demographics for Canadian provinces

Per capita NCAA hockey demographics for European countries

Per capita NCAA hockey demographics for US states

As you can see in the chart below, there are some pretty drastic differences for some countries, states, and provinces when we rank them by players produced per capita instead of in general.  In Canada, Prince Edward Island, Saskatchewan, and Manitoba all produce a very respectable number of players for their population size.  Prince Edward Island's ranking went from tied for 32nd in the general rankings all the way up to 13th when we looked at player production per capita.

As for the European countries, the only country with a positive change was Latvia.  The rest saw negative moves when we looked at player production per capita.  Not surprising considering the still relatively low (but really growing) participation in NCAA hockey by European players and the relatively large population sizes.

Moving on to the United States, the big gainers when we looked at player production per capita were North Dakota, Alaska, New Hampshire, and Delaware.  Again, not surprising based on the strong hockey bases in these states and the relatively low population sizes.  Minnesota stands still for much the same reason, moderately low population size and strong participation level overall.  

The one decline that really jumped out at me was Michigan.  Always a strong hockey player producing state, it has seen it's numbers decline gradually over the last decade and it really shows up when we look at these rankings on a per capita basis.  For the health and future of USA Hockey, I hope to see Michigan's numbers jump back up where they belong in the near future.

What also becomes clear when looking at these numbers on a per capita basis is that while the big gains by states like Illinois, Pennsylvania, California, Texas, and Florida are really, really great to see, it reminds us that these states (considering their populations) had a long ways to catch up and have great potential to make even bigger gains moving forward.

Change in ranking from general ranking to per capita ranking for NCAA players produced

Players Produced by City Per Capita

I also wanted to revisit the data that I compiled for the number of players produced per city to determine which city has produced the most NCAA hockey players this season on a per capita basis.  It amazes me how many hockey players some of these tiny cities near the top have produced.

There were over 900 cities who contributed players to NCAA hockey this season, so I had to cut the list down to only those cities that have produced at least two players (or else I would still be compiling the list of populations for all the cities).  This season Strong Field, SK comes in at the top having produced two NCAA hockey players (Brady and Chase Norrish of RIT) despite Srong Field only having a population of 47 people.  That means that 1 out of every 24 people in from Strong Field players NCAA hockey this season.  Amazing stuff.  Take a peek at the rest of the list below to see the Top 240 cities that have produced at least two NCAA hockey players this sorted per capita.

Top NCAA hockey player producing cities per capita (1-80)

Top NCAA hockey player producing cities per capita (81-160)

Top NCAA hockey player producing cities per capita (161-240)

Lots of Change in Ten Years

I was really fortunate to remember that Inside College Hockey did a regular piece on geographic participation of college hockey every season because it afforded me the opportunity to do a comparison with the numbers for this season.  I was even more fortunate to find a piece that they did for the 2004-05 season because that will give us a 10 year span of time in which to make some more meaningful conclusions.  Before I discuss some of the changes I noticed, take a look at the charts I put together below that lay out some of the changes between the 2004-05 season and this season.

Change in NCAA hockey demographics for Canada between 2004-05 and 2014-15 seasons

Change in NCAA hockey demographics for Europe between 2004-05 and 2014-15 seasons

Change in NCAA hockey demographics for the US states between 2004-05 and 2014-15 seasons

Some pretty noticeable changes between 2004-05 and today.  First off, let me say, congratulations New Jersey.   You nearly doubled the number of NCAA hockey players you've produced in just 10 short years.  Pretty cool.  Also great to see such big gains in California, Florida, Illinois, Wisconsin, Texas, etc...

You might also notice that Michigan had a pretty noticeable positive gain between 2004-05 and this season as well.  Perhaps this is just the beginning of that comeback that I was talking about.  I hope so.

Also pretty hard not to notice the big increase in players from Sweden and Finland in NCAA hockey over the last 10 years as well.  While I am sure we will continue to have European countries pop on and drop off this list in the near future while the European presence in college hockey continues to grow, I think it is safe to say that some of these countries (Finland and Sweden for sure) have made solid inroads and will continue to produce more NCAA hockey players in the future.

It's a whole different tale when you take a look at the numbers for the Canadian provinces.  A part from British Columbia, which has made a considerable gain in the past 10 years (thanks to the BCHL in my opinion, but more on that later) and Prince Edward Island (which went from zero players to one this season), all of the Canadian provinces saw a decrease in NCAA player production over the past 10 years.

I'm not sure if it is getting harder to pry Canadian kids away from Canadian Major Juniors, whether coaches aren't recruiting North of the border as often because the supply of high quality players in the states is growing and more European kids are deciding to play college hockey, or whether coaches see more Canadian players as flight risks in the midst of the recent NCAA vs CHL recruiting battle.  

I don't have the answer, but I think it is very intriguing.  It could be all of the above, a combination of the above, or none of the above all together.  Whatever it is, I think it is safe to say we are seeing some shifts in college hockey's demographics.  For a clearer picture of these changes I put together the chart below that summarizes just how much the demographics of college hockey are starting to change.

Overall change in college hockey demographics from 2004-05 to 2014-15

 
Top NCAA Hockey Feeder Leagues


The next questions I wanted to answer relate not to where these players grew up, but which leagues and teams they played for before coming to play NCAA hockey.  I am hoping this data will give us a good idea of where college hockey players are playing their hockey now days (in hopes that we can spot some trends over time) and which teams in those leagues are turning out the most players.  Below is a list of the leagues that have produced the most of this season's NCAA hockey players broken down by overall percentage as well as percentage of defensemen, forwards, and goalies.

Breakdown of NCAA hockey players produced by league for 2014-15 season

As you can see in the chart above, the USHL has a pretty commanding lead in the number of NCAA hockey players it has produced with 35.55% of this season's NCAA hockey players having played in the USHL prior to playing college hockey.  What is noticeable to me is how many players are also coming from the BCHL and NAHL (15.26% and 12.89% respectively).  Between these three leagues, that accounts for roughly two thirds of this year's players.  Pretty remarkable.


Another thing that came to mind when I looked at these numbers, and I eluded to this earlier when we were looking at the per capita numbers for regions, is that it is probably no coincidence that the numbers from British Columbia have grown so much over the last 10 years considering the number of players the BCHL has been sending to college hockey.  

It's also hard to miss the low percentage total of players that have come directly from US high school hockey directly to the NCAA.  I don't have the numbers for 10 years ago that shows the break-down by feeder league back then (wish I did), but I am guessing that it was much higher than the 2.55% that we see today.  

No surprise here as NCAA teams have more and more success with players having honed their skills in leagues such as the USHL, BCHL, and NAHL prior to coming into the college ranks.  There will always be a player here and there that is skilled enough to jump right into the college ranks, but for the most part we are seeing college teams ask their recruits to play a year or two of juniors before coming into college hockey.  I expect this trend to continue. 


However, I didn't stop there.  The next thing I did was to break-down the Top two feeder leagues (USHL and BCHL) by team to see which teams within these leagues have produced the most players.  I am sure these totals will fluctuate over time, but I still think it is interesting to see which teams are producing the most college hockey players we watch every weekend.  Take a look at the results below.

 

NCAA hockey players produced by USHL team for 2014-15

NCAA hockey players produced by BCHL team for 2014-15

While there seems to be a much more clear-cut leader when it comes to the BCHL (Pentincton), there are quite a few teams in each league that have been contributing players to the NCAA ranks.  Note that the USNTDP leads the USHL in players produced.

The final analysis I wanted to do in terms of the NCAA feeder leagues was to break-down the number of players produced by these leagues by position and by production.  I took a look at which leagues have produced the most defensemen, forwards, and goalies.  Then I took a look at which league has produced the highest scoring defensmen and forwards and most productive goaltenders at the NCAA level.  

Again, these numbers will probably become more meaningful as I track these numbers over time and start to pinpoint some trends, but the results are interesting none-the-less.  In the chart below the leagues are ranked first by points\player produced, then by pts\player for defensemen and forwards.  Finally, on the far right, I ranked the leagues first by goaltenders produced who have played at least 100 minutes so far this season for their NCAA school, then ranked those leagues by the average Save Percentage of goaltenders produced who have played at least 100 minutes for their NCAA school.

 

Further break-down of NCAA feeder leagues by position and production

Player Height, Weight, and Age Demographics

Throughout the course of the season I tend to hear a lot of discussion about which team(s) in college hockey are the youngest, oldest, largest, and smallest, so I thought it would be fun to include some demographics on the age and size of college hockey players and NCAA rosters in this analysis.  Before we start breaking the numbers down, let's start with the average age, height, and weight of your typical NCAA hockey player.

The average Height, Weight, and Age of your typical NCAA hockey player

As you can see in the chart above, the typical NCAA hockey player is 6' 0.06" tall, weighs 189.02 lbs, and is roughly 21 years and 326 days old.  Not sure about you, but the stat that really jumps out at me is the average age of your typical NCAA hockey player.  Again, I don't have the numbers for size and age demographics from 10 years ago, but I would bet that we have seen an increase in age in your typical NCAA hockey player over the past 10 years.

That should come as no surprise if you think about it.  Going back to the discussion above regarding the likely drop in players coming directly from US high school leagues, teams are wanting their recruits to hit campus with more experience.  Players are increasingly being asked to play a year or two of junior hockey before coming to campus, so it makes sense that the average age of college hockey players has likely gone up.

I also took the time to break these demographics down by team.  The charts below are the age, height, and weight demographics and rankings for all NCAA teams, separated by conference.  The rankings on the far left indicate the NCAA ranking for each team on a combination of age, height, and weight (from the largest and oldest NCAA team to the youngest and smallest NCAA team).  I also ranked where each team ranks within their conference using the same criteria (again from the largest and oldest team to the smallest and youngest).

On the right side of each chart is a break-down of where each team ranks in the NCAA based on roster height, weight, and age individually.  Again, the rankings go from the oldest, tallest, and heaviest to the youngest, shortest, and lightest for each respective ranking.

Atlantic Hockey age and size demographics

Big Ten Hockey age and size demographics

WCHA age and size demographics

NCHC age and size demographics

Hockey East age and size demographics

ECAC age and size demographics

As you can see from above, Cornell (6' 1.22"), Merrimack (6' 0.86"), and Canisius (6' 0.77") have the tallest overall rosters this season, while Ferris State (5' 10.74"), St. Cloud State (5' 10.96"), and Mercyhurst (5' 11.22") have some of the shortest.

This season Western Michigan (199.76 lbs), Penn State (198.59 lbs), and Michigan (196.04 lbs) have some of the heaviest rosters in NCAA hockey, while Vermont (178.46 lbs), Quinnipiac (181.14 lbs), and American International (181.35 lbs) have some of the lightest.

The oldest roster in college hockey this season belongs to Alaska Anchorage (24.205 years) followed by Northern Michigan (22.767 years) and Alabama-Huntsville (22.690 years) and the youngest roster in college hockey this season belongs to Boston University (20.479 years) followed by Michigan (20.864 years) and then Boston College (20.865 years).

However, not wanting to stop there, I further broke-down the age and size demographics by position.  In the charts below (again split up between conferences), you will find each school's average age, height, and weight for their defensemen, forwards, and goaltenders.  The rankings you see for each category are the NCAA rank (smallest and youngest to oldest and largest), but the schools in each conference are again sorted by their overall ranking (smallest and youngest to oldest and largest.

Positional break-down of Atlantic Hockey age and size demographics

Positional break-down of Big Ten Hockey age and size demographics

Positional break-down of WCHA age and size demographics

Positional break-down of NCHC age and size demographics

Positional break-down of Hockey East age and size demographics

Positional break-down of ECAC age and size demographics

• This season Colgate, Alaska Anchorage, and Northern Michigan have the oldest defensive corps in college hockey, while Cornell, Minnesota, and Boston College have some of the youngest defensive corps.  

• If we were to measure defensive corps on height, Western Michigan, Minnesota State, and Rensselaer have some of the tallest defensive corps in college hockey this season while Ferris State, Mercyhurst, and Denver have some of the shortest.

• In terms of weight, Western Michigan, Nebraska Omaha, and Clarkson have some of the heavier defensive corps in college hockey this season while Mercyhurst, Ferris State, and Vermont have some of the lightest.

• Moving on to forward groups, Penn State, Alabama-Huntsville, and Mercyhurst have some of the older forward groups in college hockey this season, while Wisconsin, Michigan, and Boston University have some of the youngest,

• If big forwards are your thing, then Cornell, Merrimack, and Notre Dame's forward groups should interest you as they are some of the tallest in college hockey this season, while St. Cloud State, Quinnipiac, and Vermont's forward groups are some of college hockey's shortest this season.

• In terms of having some weight to throw around, Penn State, Notre Dame, and Cornell's forward groups have plenty as they are some of the heavier in college hockey this season, while Vermont, American International, and Quinnipiac's rosters are some of the lightest.

• Last but not least with goaltenders, Wisconsin, Alaska-Fairbanks, and Sacred Heart have some of the oldest goaltending groups in college hockey this season, while Ohio State, Harvard, and Notre Dame have some of college hockey's youngest goaltending groups.

• Big goaltenders are becoming a hot commodity in hockey these days, and if that is the case than Air Force, Boston University, and Harvard are in luck because they have some of the tallest goaltending groups in college hockey this season.

• In terms of weight, RIT, Mass-Lowell, and Providence have some of the heavier goaltending groups this season, while Lake Superior State, Rensselaer, and Army have some of the lightest.

Lost of info to digest, but a lot of interesting info to chew on if you're bored.

Closer Look at Where Scoring is Coming From

The last thing I want to do in this piece is take a closer look at where the top scoring in college hockey is coming from demographically.  You might recall that in my first segment I broke down how many goals, assists, and points came from each country, state, and province.  I also noted that I didn't like this approach because some states and provinces have a majority of the goals and points simply because of the vast number of players they produce.

I also didn't like the idea of doing a goals and points per game per region using all players in college hockey because some of the bottom tier players from each region can drag those averages down.  I want to focus just on the top players in college hockey, and where they come from.

If you take a look at the charts below, I tried to break-down where all of the scoring, and then top scoring, in college hockey is coming from this season.  For the sake of this exercise I have split the regions (countries, state, provinces) up into two groups.  The top chart shows the group of regions that have produced at least 5 NCAA hockey players this season.  The bottom chart shows the group of regions that has produced less than 5 players.

It was important to split the regions into these two groups because I wanted to take a look at the production of the Top 5 players from each region to see how they compare and I can't fairly do that when some regions haven't produced 5 players.

So, I did an overall analysis for every region (under heading "Overall" below), but I only did an analysis of Top 5 points and goals for the regions that have produced at least 5 players.  Make sense?  Good, now let's move on.

In the charts below, I started by listing how many players each region has that have produced double-digit points up until this point in the season.  That is the ranking you see (for all NCAA teams) on the far left.  The ranking that you see just to the right of that is the ranking of the number of double-digit point players produced adjusted by the number of games played by players in each region.  For instance, New York has produced 11 players who have put up double-digit points so far this season, which is tied with the 11 that Illinois has produced.  However, overall the players from New York have played in 10.76 games while the players from Illinois have only played in 10.71.  The second ranking makes the adjustment.

Moving further to the right on the charts you will see a segment with the heading "Overall".  This segment lists the points per game for players from each region and then ranks them accordingly.  This is just a straight-up ranking on average points per player for each region.  This is the order in which the list is sorted below as well.

On the far right of the charts you will see segments with the headings "Top 5 Pts" and "Top 5 Goals".  These are the segments I calculated just for regions that have produced at last 5 NCAA players this season.  I took the Top 5 point producing players from each region and then ranked their average points per game and then did the same with the Top 5 goal scorers from each region.

The idea is that this will give us a look at just the high-end players that were produced by each country, state, province.  By taking the Top 5 point producers and goal scorers and averaging out their points and goals per game, we can compare each country, state, and province to each other to see where they stack up in terms of high-end talent.

Probably not a flawless approach, but I think it is a good start.

Closer look at which regions have produced college hockey's elite scorers.  Table shows only regions with at least 5 players.

Closer look at which regions have produced college hockey's elite scorers.  This table shows regions with less than 5 players produced.

So, what did all of this work tell us?  Glad you asked.  Let's compare the rankings for number of players produced (both in general and per capita) to what the rankings are for pts/game and goals/game for the Top 5 players in each region.

Even though Minnesota is ranked #1 in the number of players produced in general and per capita, the state hasn't been as successful in producing the top-end players that are in college hockey this season compared to other states near the top.  Minnesota ranks #6 in points per game and #4 in goals per game for their top players.

Moving up about as much as Minnesota fell in the rankings is Massachusetts (hello Jack Eichel).  Massachusetts was ranked #6 in players produced in general and #5 in players produced per capita, but are ranked #1 in points per game and #3 in goals per game for their top players.

Another state that really jumped out at me was California.  Ranked tied for #9 overall and only #34 per capita, California came in at #5 for points per game and #10 for goals per game for it's top players.

Finally, even though British Columbia was ranked #4 in players produced overall and #3 in players produced per capita, they were ranked #2 in points per game and #1 in goals per game for it's top players.  Both are ahead of Ontario (#3 in points per game and #2 in goals per game) which leads British Columbia in players produced overall, but also trails British Columbia in players produced per capita.  That ought to put a few smiles on the faces of some B.C. hockey fans.


Conclusion

I think this was a fun exercise, and I am really glad I took the time to do it.  Like I said, I hope to make this an annual thing so that we can start comparing the results year-to-year to see if we can spot any trends.  I realize that the methods used aren't perfect, but I do think they produced some interesting results.  If you spot something in the data I failed to pick up on or didn't mention, please let me know.  Would love to hear it.

Sorry I wasn't able to implement all of the requests that I got, but I did do my best to work as many in as I could.  Some were left out because they were too hard to accomplish with the limited stats I have, and some were left off because I am just low on time.  Hope you understand.

Anyways, hope everyone has a very happy, safe, and healthy holidays and good luck to all the college hockey fans out there the remainder of the season.