A Blog by Jonathan Low

 

Nov 5, 2016

How the Cubs Just Ended Sports' Analytics War

Data delivers. JL

Rany Jazeyeri reports in The Ringer:

An objective, data-driven view can change the world. It can laugh at omens. It can spit in the face of curses. It can whistle past the graveyards of games past. It can beat Cy Young winners in the playoffs. It can get up off the mat after a 3–1 series deficit in the championship round, just like it once did after a 3–0 series deficit in the ALCS. It can overcome lucky bounces, and two-run wild pitches in Game 7 of the World Series.
“First they ignore you, then they laugh at you, then they fight you, then you win.” — Mahatma Gandhi, maybe


The Great Analytics War ended at 48 minutes after midnight on November 3, 2016. The terms were unconditional surrender. The losing generals were not able to offer their surrender in person; most of them had long since departed the front for the unemployment line. The victors were Theo Epstein, the analytics movement, and the game of baseball.
At the conclusion of the most anticipated game in baseball history and one of the five greatest, Epstein accomplished the mission entrusted to him five years ago, when owner Tom Ricketts handed him the keys to his franchise and asked him to do for his Cubs what Epstein had already done for the Red Sox: bring a championship to fans who mostly had never experienced one in their lifetime.
If it wasn’t clear enough when Epstein ended Boston’s title drought 12 years ago, it should be abundantly clear today: An objective, data-driven view can change the world. It can laugh at omens. It can spit in the face of curses. It can whistle past the graveyards of games past, whether it be a Game 6 in 1986 or a Game 6 in 2003. It can beat Cy Young winners in the playoffs. It can get up off the mat after a 3–1 series deficit in the championship round, just like it once did after a 3–0 series deficit in the ALCS. It can overcome lucky bounces, and two-run wild pitches in Game 7 of the World Series. It can do what a previous Cubs team couldn’t, and bounce back after blowing a three-run lead in the eighth inning.


For well over a century, however, those in baseball ignored the very idea of using objective data to inform decisions, with rare exceptions (most notably Branch Rickey and his statistician Allan Roth). Then they laughed at the idea; when Bill James revolutionized the way a generation of young baseball fans thought about the game with his abstracts from 1977 to 1988, the most famous quote about him from someone inside the game came from then-Tigers manager Sparky Anderson, who called James “a little fat guy with a beard who knows nothing about nothing.” And then the analytics movement became too widespread to ignore or mock anymore, with the rise of the internet and sites like Baseball Prospectus (disclaimer: I was a BP cofounder) and finally the publication of Moneyball in 2003. And as people adopted it throughout the game, that’s when the war really started.

Theo Epstein (Getty Images)

The defenders of the status quo fought back by pointing out that the face of Moneyball, Billy Beane, hadn’t actually won anything substantial in the playoffs as general manager of the A’s, and in a sports culture that values postseason success above all, this was a tricky point to refute. Or at least it was until 2004, when the analytics-first, Bill James–reading, Micro League Baseball–playing Epstein, the twentysomething who landed the Boston GM job after Beane declined, delivered Red Sox fans the championship that had eluded his predecessors for 86 years.
You would think that ending one of the longest championship droughts in pro sports would reverberate throughout the game, and it did — slowly. Just a handful of teams had embraced analytics in 2004, but that soon became a half-dozen, and then a dozen, and then two dozen, and, within a decade, every single team in baseball had at least one full-time employee devoted to analyzing objective data.
But analyzing the data is one thing, and actually using that data to inform and influence organizational decisions is another. In less than 15 years, the culture inside major league front offices has changed so profoundly that where once teams were mocked for using analytics, they’re now mocked for not using them. Most front offices embraced the change wholeheartedly. But even then, some front offices remained more interested in window dressing.
The Phillies under Ruben Amaro Jr. were the last team to hire a dedicated analyst. Throughout Amaro’s tenure, his statements on analytics alternated between mockery and sheer ignorance — for God’s sake, the man didn’t know that walks don’t count as at-bats. Amaro, who was promoted to GM days after the Phillies won the 2008 World Series, presided over an almost perfectly straight downhill decline for the franchise, culminating in a 63–99 season in 2015 that finally got him fired.
Terry Ryan was hired as the GM of the Minnesota Twins in 1994, with a strong focus on traditional scouting and player development from within. After six more losing seasons in a row, his player development pipeline succeeded in building a division winner by 2002, the first of six division titles the Twins would claim in the next nine years. But as baseball was changing around them, the Twins stuck to their philosophy. When Ryan retired in 2007, his assistant GM, Bill Smith, took over the job; when Smith proved incapable and was fired four years later, the Twins rehired Ryan. They’ve lost at least 90 games in five of the last six years, and in the midst of going 59–103 this season — the franchise’s worst record since the team moved to Minnesota in 1961 — Ryan was finally fired.
And then there’s Arizona, where in 2014 Diamondbacks ownership got the bright idea of naming Tony La Russa, who had been a brilliant manager but was retired and had no experience running a team, chief baseball officer. They also let La Russa hire Dave Stewart as GM. His main qualifications for the job were that he (1) had been the ace of the La Russa A’s team that won the 1989 World Series and (2) had worked as a player agent after retiring. It took barely two years for La Russa and Stewart to prove that knowing how to build a winning team in 1989 is as useful today as an Apple IIe. They signed Zack Greinke to a $206.5 million contract and traded Ender Inciarte and no. 1 overall pick Dansby Swanson, plus other nonheadliners, for Shelby Miller in a deal that was universally savaged the moment they made it and only got worse over time. After losing 93 games in 2016, La Russa was neutered and Stewart got axed in favor of a new GM.
Those front offices, almost to a man, have been broken and are scattering in full retreat. That three of the front offices most hostile to analytics have gotten axed for poor performance in the last 18 months is proof enough that the war against analytics was coming to a close well before the 2016 World Series. But no one remembers the front of the war; they remember the decisive blow, and Epstein’s Cubs delivered that by winning a world championship.


The most remarkable thing about the Cubs ending the longest championship drought in American pro sports history might be just how straightforward it was. It’s not hard to imagine that the first time Ricketts and Epstein sat in a room together after the 2011 season to discuss whether Epstein — who had just mutually parted ways with the Red Sox — could bring the same magic to Chicago that he had to Boston, Ricketts asked him what he thought was a realistic timetable for the Cubs, who had just lost 91 games as an aging, boring club with a weak farm system. And it’s easy to picture Epstein responding that rebuilding the team right would require getting worse before getting better — that the team would be absolutely terrible for two years, show glimmers of hope by the third year, and, if everything went right, contend by the fourth year and be ready to win a championship by the fifth.

Getty Images

And that’s exactly what happened. The Cubs lost 101 games in 2012 and 96 games in 2013. But by 2014 the improvement was so rapid that you could see it in-season: After starting 40–57, they finished 33–32, and after starting the 2015 season 52–47, they finished 45–18. And in 2016 they won 103 games, the most in baseball and the most by any Cubs team since 1910. A franchise that had floundered for more than a century committed to a plan, executed it to perfection, and did not allow itself to be distracted from the ultimate goal. And as a result, Epstein, still just 42 years old, young enough that he was in grade school when James was publishing his abstracts in the 1980s, can sum up a Hall of Fame résumé in the length of a tweet: He guided the Red Sox to their first championship in 86 years. He guided the Cubs to their first championship in 108 years.
If you could choose to be a fan of any team for any season in the history of baseball, you would choose either the 2004 Red Sox or the 2016 Cubs. Maybe you’d prefer the catharsis of beating your longtime nemesis and the unprecedented comeback from a 3–0 series deficit, or maybe the Cubs’ longer wait and a 3–1 series comeback in the Fall Classic capped off by one of the greatest games of all time is more your thing. But it’s really one team or the other. Somehow, the same man built both.
Yet the task of building the 2016 Cubs was even more Herculean than the task of building the 2004 Red Sox. The Red Sox were already a competitive team, having won 93 games the season before Epstein was hired as GM, and the fact that the vast majority of teams then didn’t use analytics at all gave the Sox free reign to go after underrated players with abandon, to the point where David Ortiz wasn’t even their first choice to be the new DH. The Cubs, by contrast, not only required a complete rebuild, but had to go about that rebuild in the face of much, much more analytically savvy opponents.
The Red Sox winning under Epstein in 2004, and again in 2007, was like Genghis Khan sweeping across the steppes to conquer Asia, using military tactics that his enemies had never seen before — particularly a huge edge in intelligence operations, what we might now call “data” — and had no idea how to defend against. The Cubs winning in 2016, on the other hand, is more like Napoleon’s Grande Armée conquering Europe, using the same basic military theories as everyone else, but with more discipline and more skill.
For 100 years the Cubs were the trust fund kid who keeps screwing up but maintains his popularity anyway, like if Homer Simpson had been born with a silver spoon in his mouth. They didn’t have to win to be popular, so they didn’t. If mediocrity has any virtue at all, it is in its ease. The Cubs’ new plan under Epstein started with ditching easy mediocrity for purposeful sucking: a complete rebuild, what we like to uncharitably describe as “tanking” — except that a data-driven approach to winning makes it clear that there are powerful incentives to losing 95 games instead of 85 games.
Everyone on the Cubs team that Epstein inherited with any present value was sent packing, both to bring in prospects and to make the 2012 and 2013 teams worse on the field, giving the Cubs not only higher draft picks, but a larger draft budget under a new CBA that strictly limited how much teams could spend on amateur players. Those trades brought in guys like Travis Wood and Jake Arrieta and Kyle Hendricks, while also ensuring that the Cubs would lose 197 games over a two-year span — their most ever — earning them top-five draft picks that they used on Kris Bryant and Kyle Schwarber.
The Cubs focused on drafting and developing hitters over pitchers because the data makes clear that young hitters are a much safer bet to develop: All you need to remember is that before the Cubs selected Bryant no. 2 overall in 2013, the Astros used the no. 1 pick on … Mark Appel. Along with nailing their draft picks for Bryant and Schwarber, they traded a pitcher (Andrew Cashner) for a hitter (Anthony Rizzo), and capped off the rebuilding process by trading a pair of starting pitchers (Jeff Samardzija and Jason Hammel) for a top shortstop prospect named Addison Russell.

Ben Zobrist (Getty Images)

And when it came time to flip the switch and use the Cubs’ enormous financial resources to supplement all of their cheap, young talent, the team shopped on the elite side of town, signing Jon Lester (and, yes, Jason Heyward), and also Ben Zobrist. Zobrist has long been the poster child for analytics, because a traditional view of the game sees him as a nice complementary player but not a star, when in fact his exemplary plate discipline, broad offensive skill set, and defensive versatility make him elite. From 2009 to 2012 Zobrist had the most bWAR of any hitter in baseball, a fact that will never not surprise, and he ranks sixth among all hitters in that metric over the last eight years.
A generation ago Zobrist might have played the game in quietly excellent anonymity, his talents gone relatively unappreciated. But in the last two years, the A’s traded a bundle to get him in his final contract year, the Royals traded an outstanding prospect (Sean Manaea) to rent his services at the trading deadline for three months, and the Cubs gave him a four-year, $56 million contract even though he’ll be 38 by the time it ends. And after being the Royals’ best postseason hitter on the way to a championship last year, Zobrist somehow outdid himself, driving in the winning run in Game 7 and earning World Series MVP honors.
While anti-analytics teams like the Twins and Phillies have to start over with young, analytics-minded GMs, the Cubs under Epstein built one of baseball’s best teams in a generation almost from scratch in five years. The Cubs are just the second team of the century to win a World Series and have the best regular-season record in the majors. They have so much depth that they can pull their starting catcher (Willson Contreras, who hit an RBI double) in Game 7 of the World Series in order to give Jon Lester a comfortable caddy to throw to and see that backup catcher (David Ross) hit a home run and draw a walk in his two plate appearances.
And having won a world championship, the Cubs can now reasonably set their minds to building something even greater. They were tied for the youngest average lineup age (27.5) of any world championship team since the 1969 Mets. In Game 2 of the World Series, they became the first team ever to start six hitters under the age of 25 (Bryant, Schwarber, Javier Báez, Contreras, Jorge Soler, and Russell) in a playoff game. The core of their lineup and their two best starters (Lester and Hendricks) are all under contract for at least four more years. Thanks to analytics, this long-wretched franchise is poised to form a dynasty.


To be perfectly clear, “analytics” doesn’t mean “numbers.” It means cutting through the bullshit. It means having a reason for every decision you make, and that reason being something other than “because that’s the way it’s always been done.” It doesn’t mean eliminating Conventional Wisdom; it means questioning it. It means getting as much data as you can, but “data” is just a fancy word for information. The Cubs don’t focus on stats at the exclusion of other forms of information — they famously picked Hendricks as the pitching prospect they wanted from the Rangers four years ago due to his makeup.
The Cubs don’t turn their noses up at any information that can grab them an edge. Nor will they turn their noses up at any tactics that might do the same, even if that may rub some people the wrong way. They place winning as a higher priority than propriety. Two years ago Epstein — and Jed Hoyer, who is, after all, the team’s GM and should not be overlooked as Epstein’s right-hand man both here and in Boston — made the awkward and slightly unseemly decision to fire Rick Renteria, their hand-picked new manager who had done nothing wrong in his one year at the helm of the franchise, for the sole reason that Joe Maddon, then as now considered one of the best in the game at his craft, had unexpectedly become available. Renteria will get his opportunity for crosstown revenge as the new manager of White Sox, but that decision, like so many other difficult ones, directly led to the celebration to end all celebrations.
The Cubs didn’t let reports that Dexter Fowler had signed a three-year deal with the Orioles back in February keep them from stealing him away before his contract with Baltimore was official — a decision that both sides undoubtedly are happy with today. And most controversially, this summer the Cubs traded for Aroldis Chapman, the hardest-throwing pitcher in baseball history, despite his arrest and suspension for a domestic violence allegation. Though charges were dropped, Chapman’s arrival still made it uncomfortable to root for the team in the abstract, even though adding Chapman filled the team’s one big weakness, the lack of a shutdown closer. (And it’s easier to be happy for the Cubs knowing that Chapman blew the save in Game 7 and was not on the mound for the final out.)
That the Cubs were able to overcome Chapman’s failures to win, becoming the first road team ever to win Game 7 in extra innings, hastened the end of this analytics war, but even if they had lost, it would only have postponed the inevitable. Because the team that pushed the Cubs the closest to the brink, the Cleveland Indians, are themselves one of the most analytically savvy teams in the sport. Keith Woolner, their principal data scientist, baseball analytics (yes, that’s a real title), was my longtime colleague at Baseball Prospectus, and I can say that with the possible exception of Nate Silver, no one I worked with there better combined quantitative brilliance with the ability to clearly communicate complex mathematical processes in layman’s terms.
Terry Francona managed as closely to the platonic ideal of a sabermetric manager in this postseason as anyone ever has, and got a roster missing two of its three best starting pitchers within sight of a championship by doing things like using Andrew Miller for more innings than any reliever had ever thrown in one postseason, frequently batting a slow DH (Carlos Santana) leadoff because Santana draws so many walks, and starting ace Corey Kluber three times in the World Series. Both teams in this matchup were firm believers in the value of analytics; it’s just that only one could win. Among championship teams, the difference is no longer between teams that value analytics and teams that don’t. It’s between teams that are open about how they use analytics (the 2016 Cubs, the 2013 Red Sox) and the teams that keep their strategies as much of a secret as possible (the 2015 Royals, the 2014 Giants).


Baseball has been solved, and the solution is simple: There is no solution. It’s when you think you’ve got the game figured out that it bites you in the ass. There is always more information to be had, and more information is always useful. The battle was never between the quants and the gut-instinct types, it was between the curious and the incurious. The curious have won. Like Japanese soldiers hiding in the mountains of the Philippines for 30 years after World War II, there will still be pockets of resistance for some time in the form of small-town columnists desperate to serve up clickbait with an anti-analytics screed. But make no mistake: The war is over.
So now it’s time to take the revolution to other fronts. Paul DePodesta — famed in baseball circles for being Billy Beane’s assistant GM during the Moneyball movement so many years ago — has opened a front in the NFL, taking over the Cleveland Browns, and let’s be real: The Browns winning a Super Bowl would be as remarkable and life-affirming as anything Epstein has done.
But the battle to come is the only battle that really matters: the war against truthiness, against the fact-free, tell-the-people-what-they-want-to-hear zones that permeate the political world. Silver laid a beachhead by revolutionizing the coverage of polling and politics with FiveThirtyEight, but it’s time to push inland. Daniel Patrick Moynihan told us we were free to have our own opinions but not our own facts, but for 30 years lobbyists and pundits and talk-show radio hosts have done their best to prove that no, if you repeat a lie often enough, it becomes close enough to a fact as to be indistinguishable from the real thing.
Like the quote above, which has been attributed to Gandhi for decades despite the absence of proof that he ever actually said it.
In 1989, Francis Fukuyama published an essay in The National Interest entitled “The End of History?” and political theorists have argued ever since over his thesis: that Western-style liberal democracy will inevitably be recognized as the most effective form of government throughout the world, and that all nations will eventually bend themselves to this form of government.
The Cubs’ championship bookending the Red Sox championship 12 years ago suggests we are approaching the end of an era in baseball: a recognition that a front office that melds analytics and scouting information, that sees no contradiction or controversy in using data of all types to inform its decisions, is the inevitable harmonic perfection of what a major league front office should look like, and that every organization in baseball is heading in that direction. It doesn’t mean the game will be less interesting; on the contrary, the competitive urge to find an edge over rivals will spur more innovation (defensive shifts! pitch framing!) over time, introduce previously discarded strategies (use your best reliever to put out a crisis in the middle innings!), and create a game of strategizing cat-and-mouse off the field that’s nearly as compelling as the game on the field.
If you’re a Cubs fan, it’s time to party like you’ve never partied before. But if you’re a baseball fan, and a fan of smart front offices doing smart things and pushing the envelope and trying new strategies in a never-ending quest to secure a competitive advantage, you should be rejoicing, too. A deeper understanding of the game of baseball not only makes for a better front office, but for a better fan experience. Cubs fans, and Red Sox fans before them, may owe Epstein a particular debt of gratitude, but we’re all indebted to him for helping make the game of baseball more rewarding to follow in 2016 than it’s ever been before.
Maybe they’ll put that on his Hall of Fame plaque, too.

0 comments:

Post a Comment