Maybe an update on that work is in order. Since that paper, I've continued scheduling the umpires through the company The Sports Scheduling Group, making the 2016 schedule the tenth schedule of ours that they have used. The problem has changed quite a bit. Perhaps the biggest change was a couple of years ago when MLB introduced video replay. That was a very good year to be a minor-league umpire. Two new MLB crews were created, increasing the number of crews from 17 to 19: eight minor-league umpires got promoted! That change also made the schedule more complicated, since it was necessary to have two crews in New York at all time to handle the video reviews.
But complicated is good for optimization: our systems were very useful to determine the impact of video replay on the umpires. And we have ended up with a schedule structure that the crews seem to like with crews handling video replays shortly before their vacations. Most stay on the east coast, and so have a relatively easy travel schedule leading up to their off weeks.
The methods have been updated since that paper, with more of an emphasis on large neighborhood search rather than the simple simulated annealing in the paper. But the problem still remains resistant to solution by exact methods. The Traveling Umpire Problem (with a new website at TUP Benchmarks | Home) shows that 14 team tournaments are now schedulable, up from the 10 or so in our early work, but the 30 team MLB problem remains well out of reach. But the heuristics consistently get good, usable schedules.
Operations research and analytics continue to have an impact on umpire scheduling, and an even stronger impact on league scheduling.
As for whether analytics can predict the Super Bowl? Maybe not the winner, but it does have a big effect on the schedule that leads there!
Original Message:
Sent: 02-02-2016 07:12
From: Scott Nestler
Subject: Can analytics predict the Super Bowl?
First, as the Editor of the "Analytics in Sports" volume, I would like to thank Laura McLay for calling it out in her most excellent blog, Punk Rock OR, which I have read for several years now. Also, thank you to Anne Robinson, the Series Editor, for asking me to put together this collection. This was a fun project because there is such a variety of quality material to choose from in the stable of INFORMS journals.
Now on to some specific comments about articles / podcasts / videos that Laura mentioned, and some others that interest me personally. Although one of my goals in putting together this volume was to focus attention on recent articles (from the past 2-4 years), I chose to include the 1971 article by Carter and Machol, and also the 1953 Letter to the Editor (of OR) by Mottley precisely because I wanted to show that people the OR field have been "doing sports analytics" for decades.
The majority of sports related articles appear in either Management Science (one of two flagship publications, the other being Operations Research) or else Interfaces, which has a more practical and applied bent to it than most journals. The article by Willoughby and Kostuk on strategic decision-making in curling (in Decision Analysis Journal) and the 2013 article by Doug Chung (in Marketing Science) are noteworthy in that they show diversity across a number of different INFORMS publications (which totals 14, for those who didn't know). With regard to curling, I have never played but always enjoyed watching it as a kid when we went to the local country club with family friends and when it is on during the Winter Olympics. Perhaps I like it because it is similar to "table (bar) shuffleboard."
The article on Scheduling Baseball Umpires ..." by Mike Trick, Hakan Yildiz, and Tallys Yunes is one of a number of classic applications of optimization, that date back a number of years. I included this one, over the seminal 1998 OR article by Mike Trick & George Nemhauser, "Scheduling a Major College Basketball Conference" primarily due to the "recency bias" in the intent of Editor's Cut. But, as this series is a living collection that can be updated, rather than a static list, we have the ability to add items to the volume. So, I'm putting that article in the collection too; look for it soon.
Bukiet, Harold, and Palacios' article on using Markov Chains to develop an optimal batting order in baseball has fascinated me for over a decade now. I started working on an improvement to it (with an expanded state space) with some friends when I was in graduate school at Maryland. But, it wasn't computationally tenable. I think we may be there now, both in terms of hardware and algorithms. If anyone out there is interested in collaborating on this, please send me a private note.
As a teaching faculty member, I have used some of the suggestions by Paul Kvam and Joel Sokol in "Teaching Statistics Using Sports Examples" to my students. Using sports as a "gateway drug" to math, statistics, and OR is a most excellent idea. Also, I'm currently teaching a Sports Analytics elective course here at Notre Dame ("Go Irish!), to both MBAs and undergraduates, and have found the entire volume to be a useful tool for stimulating discussions in few of my class sessions. I hope that you find it both entertaining and useful too.
Now, with regards to Anne Robinson's question about Super Bowl 50 (since they assume Americans can't read Roman numerals anymore?)... Like Yogi Berra, or somebody once upon a time said, "prediction is hard, especially when it is about the future." My corollary to that would be that prediction is especially difficult when people are involved. So, while analytics can inform us about the likelihood of future events, randomness due to injury, weather (even in CA), and just plain old luck often seem to rule the day.
In the chapter "How Competitive are Competitive Sports?" of their book "Scorecasting," Moskowitz and Wortheim talk about differences between various professional sports leagues (e.g. NFL, NBA, NHL, MLB). While there are similarities between these leagues (which are businesses), there are also differences (like the number or percentage of teams that make the playoffs, whether or not they have a salary cap, etc.) They close the chapter with the following. "Trying to predict who will win the next Super Bowl is a fool's errand, but trying to predict who will win the next World Series is far easier. Though you may not be right, you can limit your potential candidates to a handful of teams even before the season begins. Funny thing about sports: Distilled to their essence, they're all about competition. But as an industry, some are more competitive than others."
So what do others out there in the broader INFORMS community think? Will the Panthers or the Broncos prevail? (And was that end zone painting incident an honest mistake or not?)
------------------------------
Scott Nestler, PhD, CAP
Associate Teaching Professor
Mendoza College of Business
University of Notre Dame
South Bend, IN
Original Message:
Sent: 02-01-2016 17:11
From: Anne Robinson
Subject: Can analytics predict the Super Bowl?
In Laura McLay's latest blog post in Punk Rock Operations Research, she references the latest Editor's Cut: Analytics in Sports (INFORMS PubsOnline). This collection of articles, podcasts and videos highlights how much the prevalence of analytics in sports has grown, with articles spanning from the 1950s to current times. I particularly like the video from HLN about the football coach who never punts (http://pubsonline.informs.org/editorscut/sports/videos). He speaks about the need to convince players and other coaches about the results from the math models, reiterating that analytics and change management cannot be decoupled!
With Super Bowl 50 just around the corner, is anyone using predictive or prescriptive analytics to determine if the Broncos or the Panthers will prevail?
------------------------------
Anne G. Robinson
Executive Director, Supply Chain Strategy and Forward Operations
Verizon Wireless, Basking Ridge NJ
------------------------------