The NHL Draft: The Call of a Lifetime

Maps, Visualization


Zane McIntyre was playing hockey in Thief River Falls, MN – a town with a population of barely over 8,000 people. Yet, he was considered one of the best goalies of his age in the state. In 2010, McIntyre would get a call to play for the Boston Bruins – an original six NHL team with a stadium that could fit the entire town of Thief River Falls twice over and then some.

NHL talent can be found in the smallest crevasses across the globe. When players get the call about being drafted, they answer it no matter how far away they may be. Thief River Falls is just over 1,600 miles from Boston but sometimes players travel further to compete in the world’s best hockey league. So, just where do these players come from? I’ve always been interested in answering this question every year the NHL Draft comes around. 

I could research each player individually to find out their story, but why not have all that data in one place? An effective visualization with this would also answer questions like…

  1. Do specific countries supply more of a position than others? 
  2. How have draft locations changed over time? 
  3. Where do teams normally draft from?


This has never been done before. The NHL Draft, or any third-party data analyst agency, has not created a map showing where players come from in pursuit of the ultimate hockey dream. There’s an incredible amount of potential that can come from making some visualizations around this data. We can find out what a typical coming-of-age story in the NHL is like and find individuals who were impacted by such a desired phone call. 

The closest map visualization I came across was this map of first overall draft picks. The map highlights where most of these picks come from but it lacks specific stories and data from those who weren’t selected first overall. The map also doesn’t work well at identifying where most draft picks come from. Each territory or country that has a first overall draft pick is highlighted in blue but that puts each location at the same level.

Map of first overall draft picks via imgur.

I later discovered this map that visualizes linguistic data across the globe. By showing different points, a user could understand that there are individual stories within larger groups or swathing data points. The map also color coats each point based on region. I believe this would be an excellent way to display different players on a map but it might be hard to see a larger picture of patterns. 

Linguistic data points across the globe via Carto.

Both maps show that there are some ways to visualize the answers to each of my research questions. However, I believe that testing out different map displays will be necessary.


Excel & Google Sheets

Although both tools are spreadsheet applications, I was forced to use both to gain proper latitude and longitude coordinates and combine datasets into one sheet. Later in this report, I will explain my use of both tools in my organization of data. 

Spreadsheets are my preferred method of collecting hard data in a tabular format. There are other file types that can store data but I was not familiar with any other options. 


My overall goal for visualizing the data and answering my main questions could be accomplished through a spatial mapping tool. Carto is a relatively free application that allows data analysts to show points across multiple types of locations around the world. The tool also allows data analysts to see spatial relationships and changing patterns over time. 


Collecting the data

The ideal dataset I needed had to include the player name, where they were drafted, who drafted them, when they were drafted, and where they were from. Unfortunately, there was no dataset that contained all of this in one form. Instead, I had to download multiple datasets and combine them into one sheet. 

This dataset from contained the most data I needed to create my ideal visualization. The only pieces missing were where the players were born and the latitude & longitude coordinates of the hometown and the team that drafted them. NHL Draft data.

The only dataset I could find that contained player names and hometowns was this dataset from The set included the hometown of each player which was extremely important in accomplishing my project objective. However, this set only included players up to those who played in the 2011 NHL season. Therefore, this project was limited to that draft year. In order to acquire enough data to solidify any patterns, I collected data from 1991 through 2011 NHL Drafts. 

NHL player data from

To finish my data collection step, I needed to also find the latitude and longitude coordinates for the players’ hometowns. I used a Google Sheets spreadsheet to obtain these points by using a script I found that would match the cities to the coordinates. 

Combining the data

Now that I had all the data I needed, I had to combine it into one spreadsheet. However, this was not a copy/paste solution. All the data was not in the correct order and would therefore be incorrect if I copy-pasted all the data into one sheet. 

I created a tab per each piece of data that I could later associate with a match by using a VLOOKUP formula in Excel. In order to match the data correctly, I used each player’s full name and matched that with their hometown, position, and year they were drafted. 

The final dataset in Excel was used in Carto visualization.

Creating the map

The first visualization that came up in Carto showed individual points across the globe signifying each player and where they were from. At first glance, we can see that players do come from across the globe to compete in the NHL. However, not every region provides NHL talent, especially those south of the equator. Most players come from either North America or parts of Western Europe. 

Default map from Carto after uploading the dataset.

Next, I assigned an image to each point based on the NHL team that drafted the player. Carto default values did not allow me to assign more than 10 logos to the points. Instead, I switched to the CSS editor where I could pick each team logo and assign them to each point. This was an extremely time-consuming process since I had to add each logo one at a time to the CSS. 

Points were identified as teams that drafted each player.

Filter widgets

The map needed to be interactive for me to answer where teams drafted the most players and which regions developed specific positions. For this, I added both the player position and draft team columns to the widget section. I thought this would be a better analysis experience for users because instead of looking at multiple maps, viewers and other analysts could customize one map on their own and filter based on what they wish to look for.

Point map with filters by position and draft team.

Animating the data

This final map did not display an answer to just how regional draft patterns changed over time. A static map would not do well at all for showing the changes either. Carto’s animation capability was perfect to include in this effort. I created an animation based on yearly change and assigned a heatmap to each year. This experience would help viewers easily identify just how much change in draft locations was happening.

Animated heat map showing changes of draft pick locations year-over-year.


Where do teams normally draft from?

Upon filtering draft locations by each team, there doesn’t seem to be a strong effort by any team to focus on players from a specific region. Each team shows draft picks coming from similar regions in Western Europe and North America but no specific locations are heavily scouted more than others.

This finding shows that teams believe exceptional NHL talent can come from anywhere. Although there might be anecdotal evidence from team scouts explaining some regional leagues are better than others, the spatial data from this map does not show NHL teams follow such a premise. 

NHL Draft Interactive Point Map.

Do specific countries supply more of a position than others?

Filtering through the different positions that are drafted, there’s no strong correlation between position and region. The closest I could find was the difference between defensemen (D) and centers (C). These two maps show that more defensemen have come from Western Europe than centers.

Due to my own knowledge of hockey history, this isn’t much of a surprise. Both Sweden and Finland have shown a more defensive mindset in their game internationally and there have been plenty of All-Star defensemen from those two countries – Nicklas Lindstrom (SWE) and Sami Salo (FIN) just to name a few.

How have draft locations changed over time?

Change hasn’t occurred as much as I thought. Although hockey has been growing in countries besides those in Western Europe and North America, the data hasn’t shown any drastic movement between 1991 and 2011 away from those regions. Occasionally, there are spots across Eastern Europe and West Asia that pop up but not at a continual rate. 

NHL Draft Heat Map showing draft pick locations year over year.

What this actually turned into

This visualization strategy didn’t show correlations, causations, or anything that iq quantitatively significant. The maps became more of an experience for NHL fans to interact with and explore. It’s more of a tool to see just who is coming to the NHL, what team is picking them, and the position they plan to play. A viewer can begin understanding the story of each player as a singular node on the map. To add more context to this data based on the realization of what these maps really are, I added pop-ups to each node on the NHL Draft interactive Point Map. Now, a viewer can form a deeper qualitative understanding of who is coming to compete.


Mapping draft picks has never been done before, and now I really know why. There are clearly no strong patterns or shifts over time for were draft picks come from or who drafts them. Although, there is still a story that can be gathered from this spatial visualization. Analysts can map the beginning stories of a player and even follow them throughout their career. This map should be a building block towards effectively tracking a player’s journey and understanding that players in the NHL go through plenty of routes to compete.

I think this map could be better visualized with edge lines from each node. An ideal map to help truly identify drafting patterns would be to connect a player’s hometown with the team that drafted them. This could be an engaging experience during a future NHL Draft for fans and sports analysts alike. 

Something else I believe would help emphasize the story around this data is filtering based on players who actually play a game in the NHL. It’s quite rare that every player from each draft even skates during a preseason game, let alone an official regular-season game, with their draft team. What this will help show is that there are no guarantees once a player is drafted into the NHL. Sometimes, it can take years for them to finally crack a roster. 

Overall, these maps show that there are plenty of individuals who were skilled enough to be called upon by the National Hockey League. We see that talent comes from all over the world and team scouts travel far and wide to find those special players. This isn’t a story to help predict where the next best player could come from – it’s an epic of thousands of small stories of individuals who dedicated their lives to the game of hockey and were rewarded for it.