Friday, 22 March 2013

Math Means Better Travel

There are many things that go into a hockey season, but one of the biggest components is travel. Teams are always looking for ways to minimize the amount of time on planes moving from city to city, and the latest NHL realignment is supposed to minimize the travel... unless you're playing in the Northeast-Florida Division. Be that as it may, no longer will Winnipeg have to make long treks across the globe to play a divisional foe, and both Detroit and Columbus actually get to play road games in a time zone when most of their fans are still awake. In short, it seems good.

But is it really that good? What if there was a way to make it better by further reducing the travel by reshuffling the teams? What if you could reduce travel through math?

Ok, I know I've featured math on this blog before which may have bored a few of you, but hear me out on this one because it could significantly change the way in how the NHL looks. I was reading through today when a story about NHL travel patterns appeared as a headline. Entitled Algorithmically Realigning Sports Leagues, the article focuses on the recent realignment plan by the NHL and determines that total league travel miles could be further reduced by better realigning the teams so that they play divisional foes who are located closer than current or proposed divisional opponents.

Under the current league alignment, a team of mathematicians at West Point determined that the NHL's current setup delivers 1,185,123 miles of travel. However, the new alignment starting next season would actually add approximately 30,000 miles of travel to the total amount of travel done by the 30 teams! Teams may play closer rivals, but the opposite of what was promised is being delivered!

Of course, one could call into question the algorithm that is being used, but the mathematics behind the algorithm are actually quite sound. I won't go into the details here, but I will link to the science behind how this algorithm was conceived in order to determine the travel trends. Absolutely worth the read if you want to see how math can flush out the best-case scenario from hundreds of possible solutions.

According to the algorithm, the best scenario for the NHL would be six five-team divisions that were as follows:

  • NORTHEAST: Detroit, Toronto, Montreal, Ottawa, Buffalo.
  • ATLANTIC: Boston, New Jersey, NY Islanders, NY Rangers, Philadelphia.
  • SOUTHEAST: Washington, Carolina, Pittsburgh, Columbus, Florida.
  • CENTRAL: Chicago, St. Louis, Nashville, Dallas, Tampa Bay.
  • PACIFIC: Colorado, Phoenix, Anaheim, San Jose, Los Angeles.
  • NORTHWEST: Vancouver, Calgary, Edmonton, Winnipeg, Minnesota.
In this realignment, the 30 NHL teams would only travel 1,155,391 miles - 30,000 miles less than what the NHL teams currently travel in one season!
The team of mathematicians, though, looked at this realignment and realized that breaking up Florida and Tampa Bay made no sense, so they shuffled the alignment again to put those two teams back together in the same division. The result? Tampa Bay shifts to the "Southeast", Pittsburgh jumps to the "Northeast", and Detroit shifts back to the "Central". Difference in travel from the best-case scenario? A mere 578-mile increase!

Now you might be saying, "Teebz, who cares? Detroit and Columbus are in the east, and Winnipeg is in the west. This works better for everyone!", and you'd nearly be right. In the NHL's realignment, the NHL admitted that not everyone was happy with the proposal, but the majority ruled the vote. Democracy prevailed.

In the mathematician's best-case scenario, though, nine teams would see increased travel, but the league would see every other team traveling significantly less. In fact, the largest increase in travel would come from the teams in the American midwest as Chicago, Detroit, St. Louis, Nashville, and Minnesota all see increases of nearly 10,000 miles of travel. The other four - Pittsburgh, Washington, Philadelphia, an Columbus - see an increase of less than a 1000 miles combined!
So what does this all mean? Well, not much. The NHL and NHLPA have agreed on the NHL's proposed divisional and conference realignment, so there isn't much to say for the next few years. However, once this trial period is over, the NHLPA may want to use some of the statistics to ensure that the players are on the road less and resting more. The days of having road-weary teams may be coming to an end with this sort of advanced math.

If you want to read the full report that also examines the travel of the other three professional leagues, you can click here for it. It's in PDF format, so it should be readable on the majority of browsers.

The key here is that math and science can play a big role in how the NHL operates. If it can help teams save some money by traveling less, that's a good thing. And better-rested players turn in better performances. Personally, I only see this math as a win-win from any perspective.

Until next time, keep your sticks on the ice!


mp34 said...

Rivalries are much more important than saving in travel. As I've said in baseball, geographical closeness doesn't really mean anything - it's good teams & important games. Selig tried to pull that crap in Interleague Play; forcing geographical rivalries but really was more about getting regional fans into the ballpark. For every Mets/Yankees & Cubs/Sox series, there were 12 snoozers. After 16 years, it was deemed a failure.
The NFL realizes rivalries are more important with its divisions, and the NHL should embrace their Original Six history. I think the new NHL realignment stinks.

mtjaws said...

While it made sense for Winnipeg to move to the West, I wish Nashville and/or Columbus were coming east, not Detroit. But that's because I don't want Florida to play them too much!

And I really don't like having the divisions so huge with 7-8 teams. I like the current 5 team format because one doesn't need to follow so many teams as closely. And to not have equal divisions makes it even more confusing in the playoff race.

I enjoyed reading this disection of travel distances, and the various configurations of teams. Hopefully the NHL reexamines things in 3 years, and makes it as good as it is today.