IT'S A NUMBERS GAME: Q&A With Jon Wertheim, Co-Author Of "Scorecasting"
“Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won” is a book co-authored by childhood friends Tobias Moskowitz, a finance professor at the University of Chicago’s School of Business and Jon Wertheim, a senior writer for Sports Illustrated.
In the book, Moskowitz and Wertheim use economic principles to validate or challenge some of the conventional wisdoms of the sporting world. (Hence the frequent comparisons to breakout hit "Freakonomics.")
The topics covered include whether NFL coaches should implement what I would call the “Maddon Strategy” and go for it on fourth down more often. They also look at a better way of measuring the effectiveness of shot blockers in the NBA and whether a .299 hitter in baseball is more valuable than a .300 hitter.
We had a chance to conduct a Q&A with Wertheim recently, where we discussed the ideas of challenging conventional sports wisdom, enhancing the fan experience, why basketball and football lag behind baseball in statistical analysis and how new owners adjust to the sports world and integrate what made them successful in business.
* * * * *
Business Insider: The first thing I always wonder with books like Scorecasting, is who is your ideal audience? Obviously you would like everybody to read your book, but what segment do you really hope will read this book?
Wertheim: Our ideal audience is probably the critical sports fan/coach/player/GM who likes to look at a few numbers and questions conventional wisdom. We'd be thrilled if all of those people read the book. But, we also hope that even the casual sports fan will enjoy it as many of the concepts speak to broader issues in life and business.
BI: The average fan that still loves the traditional stats may be your most important target, but they may also be too stubborn to purchase a book like this. How do you sell your book to that fan?
Wertheim: Good question. I think if we can convince the traditional fan that they may be missing something big if they don't look a little more carefully or a little differently at the games they love, then it will be an easy sell. We aren't trying to change everything about the way they look at sports, but rather enhance what they are already doing. We don't think our stats based approach is the only way to look at sports either. Both traditional and new aspects of sports can make watching as a fan more enjoyable. The more we learn about our favorite sports, the better in my opinion. We're just bringing (I hope) a new perspective to the games that may uncover things we never knew before. I hope that would get even the most stubbornly traditional fan to take a look at it. Hey, if nothing else they can argue and complain how we got it all wrong!
BI: I have heard Scorecasting referred to as "The Freakonomics of Sports," and I can certainly see the similarities. Was that something you had in mind when you guys started this project? And what was the ultimate goal of Scorecasting? Are you trying to change conventional wisdom or just giving the fans another way to look at the games they are already watching?
Wertheim: Yes. Certainly Freakonomics, which was co-authored by my close friend and colleague Steve Levitt, was an inspiration and something we emulated. But, we also think the book is quite a bit different from Freakonomics. Sure, some of what we do is topple or at least challenge and try to understand current conventional wisdom in sports. But, we also tried to break new ground on topics---such as referee bias, the lure of round numbers, the economics behind steroid use, etc. We hope to give fans a new or fresh perspective on how we watch sports and part of that is to challenge and understand conventional wisdom.
BI: I have taught university classes, and I have always thought that the most challenging aspect of teaching is to challenge your best students without losing the other students. I thought you did a great job of presenting the analysis in a manner that the average fan could absorb without "dumbing it down." How did you weigh stimulating the critical thinking abilities of those that are already sold on things like sabermetrics to the average fan that may be more traditional when it comes to stats?
Wertheim: Thanks. Yes, it's nearly impossible to satisfy both types of fans fully. And, as a professor myself, you always face this problem that it's hard to satisfy the best students and the other students simultaneously. You hope that you can hit the right tone that at times challenges and certainly doesn't bore the best students and yet doesn't lose the other students. We tried to strike that balance with the book---offering some statistical analysis at times and some more lighter anecdotes and stories at other times. Surely, some stats guys will want more analysis and more numbers and others will want less than we currently do, but we tried to balance out these two sides. Hopefully we succeeded.
BI: One of the themes that is found throughout the book is the idea that coaches and teams will make decisions that are less than optimal. This might happen because that is just the way it has always been done (e.g. maybe NFL coaches should go for it on fourth down more often) or because they are overly obsessed with nice round numbers (e.g. a .299 hitter may be more valuable than a .300 hitter). But one thing I have struggled with is weighing these decisions in the bigger sports picture. In the stock market, while more is better, any profit can be looked at as a success. In sports, there is only one winner each year. So while teams might find more value in a player that will hit 29 home runs (just below that sexy number 30), the guy that hits 40 home runs is going to help win more games. At what point is it ok for teams to ignore the Return on Investment and overpay for the player that is more likely to help them win a championship?
Wertheim: That's a good point. I think it comes down to how you define "return on investment." If return is based on just wins and losses, then you're right that you may be missing the bigger picture of improving your chances for a championship. On the other hand, you can define return as "chances of winning a championship." In a lot of what we do, we analyze both and sometimes these objectives don't deliver the same implication. So, yes you may want to "overpay" for a player based on wins if it means it will improve your chances of winning a championship.
BI: Another issue that I think is unique to sports is the idea of ownership. Owners in sports are not like owners in other businesses. In general, they are not building franchises from scratch and quite often, owners are not veterans of that particular sport. This isn't the owner of a successful bread company buying another bread company. Often, the owner feels the smart thing is to defer to those that are underneath him/her and to the people that have been in the sport for a long time. But those sports people may also have some of the bad habits that you write about. And this keeps the owner from implementing the strategies they used in their other successful endeavors. One notable exception to this is Stuart Sternberg's group with the Tampa Bay Rays. But we have also seen owners that meddled too much. For the most part, these owners are not dumb, and many are smart business people. Is there a way that these owners can be more effective at bringing their expertise to the sports teams they run?
Wertheim: This is an excellent point and a difficult issue. Owners need to balance their business expertise with the expertise of their coaches and executives as well as the psychological balance of their staff feeling they are being meddled with too much or neglected. My view is that owners and veterans of the sport bring different skills and knowledge to the table and both are valuable. The hard part is figuring out how to extract the value out of both groups' expertise and get them to understand each other. If both groups agree on the same decision, then I would think that would make both groups feel even better about the decision since two different perspectives arrived at the same conclusion. The tricky part is when the two perspectives diverge. In that case you try to understand why. But, at its simplest level, if both sides agree that there is value in the other's perspective, then that is a major first step. The disasters occur when one side feels the other has no value, which can't be right.
BI: One of the chapters I loved most was one that explains why some blocked shots in basketball are more valuable than others. This is not necessarily a foreign concept, but I had never seen anybody attempt to measure those values. And while most basketball people and fans know the blocked shot that leads to a fast break layup is more valuable than a blocked shot that goes out of bounds, at the end of the day, we remember the big blocks more. A lot of this might be how the deadball situation gives the player a chance to "style" and we see all his teammates congratulate him. But another aspect of this that was really interesting is the idea of advanced metrics in sports other than baseball. There are a lot of great sources for baseball stats. But SABR-like metrics in football, basketball and other sports are still relatively rare. Is this a lack of interest or do you think the sports just don't lend themselves to this type of analysis?
Wertheim: Great question. I think there are two reasons and you mentioned them both. First, the taste for metrics in other sports like the NBA and NFL are just catching on. So, baseball had a head start and now basketball is catching up. Football seems to be lagging more, but more and more teams are embracing the metrics/stats-based approach. Second, I think you're dead on that some sports lend themselves much more easily to this type of analysis than others. Baseball is the easiest to analyze because most plays are in isolation. That is, usually only one or two players is involved in almost every play and so isolating and measuring the effects is relatively easy. In basketball it's a little harder because you have 10 moving parts on the court at all times. On the other hand, there's only one player with the ball. So, basketball metrics have gotten pretty good at measuring shooters and defenders in isolation, but where it is tricky is measuring combinations of players on the floor and interactions between players.
Does Kobe Bryant play better with Andrew Bynum and Lamar Odom on the floor or with Bynum and Pau Gasol? Does it matter who the other team has on the floor? These are trickier to measure and basketball is just beginning to look at this. In football, however, it is probably hardest to measure because every play involves almost every player---22 of them. And, there is a lot of stuff going on in any play that is hard to measure and quantify (blocking patterns, receiver routes, decoys, play-fakes, …). So, it's also the case that the nature of the game in football doesn't lend itself as easily to these metrics. That doesn't mean they shouldn't or won't be used, it's just that it's harder and not as obvious how to do it.
BI: I listened to the recent podcast you did with Jonah Keri and you mentioned something that floored me. I had heard that all baseball teams have what we might call a "sabermetrics department." And while I knew that some teams obviously are going to use that information more than others, I was shocked to hear that some teams employ these people, but never use them at all. This to me seems even worse than just blindly giving your Cy Young vote to the guy with the most wins. This means these teams are paying people for absolutely no return. Is this just because teams feel they need to do this, but don't know how to use the info or are they just appeasing the fans, or do you think it is something else entirely?
Wertheim: It is amazing that teams would ignore an entire department providing what could be very useful information. Part of the explanation could be that the department exists just to appease fans, part of it may be option value. "We don't think we need these guys, but we could be wrong. So, we mostly ignore them but just in case this stuff proves useful we'll be glad we have them." Think of this almost like and R&D department in a corporation. We'll fund lots of research and development that mostly won't work and we'll throw away and ignore, but every once in a while they may come up with something big we can use. It may be worth it, then, to keep them around even if we mostly ignore them. I think some teams may indeed have this attitude.
* * * * *
Wertheim and his co-author, Tobias Moskowitz, with both be speaking at the MIT Sloan Sports Analytics Conference in Boston this weekend.
ncG1vNJzZmivp6x7o8HSoqWeq6Oeu7S1w56pZ5ufonyyrYyjpqdlp5q%2FtbTEoqRmm59irrbAx6ipZqeWYsCku9GempqrpJ67qHmRaWhqZWM%3D