Loading...

Two in A Row: Astros, Cubs Prove Power of Data Analytics for Baseball

Two in A Row: Astros, Cubs Prove Power of Data Analytics for Baseball

The debate is over. Or, at least, it should be.

It ended on the night of Nov. 1, 2017, when the Houston Astros beat the Los Angeles Dodgers 5-1, claiming the team’s first ever championship.

Importantly, especially for those with an interest in data analytics for baseball, the Astros’ victory also marked the second year in a row that an analytics-driven team won the World Series. The Chicago Cubs did it the season before, ending a 108-year championship drought on the northside of Chicago.

Does this mean every Major League Baseball team will follow suit? No, but some will.

The Oakland Athletics and Tampa Bay Rays already have shown that commitment to data-driven strategies can keep a keep a team competitive even with a low payroll. Now, the Cubs and Astros have shown it can also win championships. And many would argue the 2015 champion Kansas City Royals also were a great example of “Moneyball” tactics winning a championship.

Still, some strategies used as part of an analytics-driven approach remain controversial. Some teams seem concerned about full commitment. But a lot of the criticism stems from the fact that some people (yes, even sports writers) don’t understand the tactics.

Lose To Win

ESPN believes about a third of the teams in baseball aren’t even trying to win this season, instead building for a better future. Others have suggested that baseball should just pull out of Florida after both the Rays and the Miami Marlins unloaded veteran players during the offseason.

That’s not the case, at least for the Rays. And maybe not even the Marlins, although that situation is much more complex and would require a college professor and entire semester to unravel.

The Astros represent the most extreme example of letting go of expensive players and fielding young, cheap talent. Houston lost 416 games in the 2011-2014 seasons. The payoff? The young players matured, the team got better and eventually became champions.

The “lose to win” idea gets criticized in the media, but data analytics professionals know it actually makes sense. Teams aren’t “trying to lose.” They are finding value and planning for a better future.

And yes, this includes getting higher draft picks after a season of losing. The Astros, for example, got first round picks George Springer (2011), Carlos Correa (2012) and Alex Bregman (2015) during their losing years. All three played in the World Series.

Finding Value in Analytics for Baseball

This is more about the “Moneyball” approach made famous in the 2003 book about the Oakland Athletics by Michael Lewis. To compete against teams with more money, the A’s had to find new ways of evaluating players. That meant a deep dive into analytics for baseball.

What they came up with is now commonplace. Valuing on-base percentage over batting average, for example. Or the advantage of good defense and pitchers who can induce ground balls.

The larger question teams face in terms of value is whether to keep an aging star with deteriorating numbers or promote younger players with high potential and lower salaries. It’s an issue faced every off-season and during the trade deadline, especially for teams with financial constraints.

The Rays, led by one of the most data-driven front offices in baseball, faced that question this season. They chose to trade fan-favorite Evan Longoria to the San Francisco Giants and bring up youngsters.

The media howled that the Rays were tanking. But Fangraphs, a site that focuses on analytics for baseball, noted that the Rays were no worse off than they had been before the trade. Writer Jeff Sullivan pointed out that the Rays almost always shoot for 81 wins. It’s a calculated, data-driven approach.

The results? The Rays have won at least 80 games in eight of the last 10 seasons. They’ve reached the playoffs four times during that span and won an American League pennant. It’s hard to argue with that success, although people still do.

Don’t Overpay

As part of finding value in younger players (or, in some cases, overlooked veterans), data-driven teams also refuse to overpay for stars in the game. Longoria, for example, was about to enter a year in which he could have dictated the terms of his trade by refusing to go to certain teams. Pay attention and you’ll notice small market, savvy teams (Rays, A’s, Pittsburgh Pirates, Kansas City Royals) dealing players before they reach certain points in their contract where they will have more control, more guaranteed money or both.

Add Pieces Only When Necessary

Last year, the Astros rolled mostly with young players. But when the time came to make the final push toward a championship, they spent the money to bring in free agent pitcher Justin Verlander at the very last possible minute.

This has, generally, been the approach of most data-driven teams – build around a core of young, inexpensive players and then bring it players who are needed when the time comes.

Different Evaluations

Fans love to see a player hit a home run or steal a base. But while they are focused on the drama on the field, analytics professionals are watching the game in completely different ways. Some of the statistics they now use include the following.

WAR. Wins Above Replacement is a complex formula used to judge how much value a player has above or below an average replacement player, expressed as a single number. That number indicates how many wins a player contributed to above (or below) what an average replacement player would have done.

ERA+. An earned run average is calculated by simply taking the total number of earned runs allowed multiplied by nine and divided by innings pitched. ERA+ is seen as more accurate as it adds information about individual ballpark dimensions that can help or hurt pitchers. The baseline average ERA+ number is 100. As an example, Dodgers pitcher Clayton Kershaw had an ERA+ of 180 in 2017.

Fielding Independent Pitching. The FIP offers another, better way to judge a pitcher’s effectiveness. The FIP only factors in what pitchers control – preventing home runs, walks and hit batsmen, as well as delivering strikeouts. They don’t get credit for great plays made by the people playing behind them.

OPS. Forget batting averages or even on-base percentage. The on-base plus slugging number is a more accurate assessment of a player’s ability to produce runs. On-base percentage calculates the percentage of times a batter gets on base, whether by a hit or drawing a walk. Slugging is the total number of bases reached divided by the number of at-bats. Add the two together for the OPS. An OPS of above .900 is elite.

DRS. What about fielding? The defensive runs saved number indicates how many runs a defensive player saved his team – or cost his team – through his play. An average Major League player will save his team 15 to 20 runs during a season.

Exciting? They are if you are into analytics and see the difference players with good numbers can do for a team.

But it also means fielding a team with non-household names as Ben Zobrist. But it should be noted that Zobrist, an analytics’ darling, played on the Rays team that won the pennant and both the Royals and Cubs teams that won the World Series.

These are just some of the steps used by data-driven teams. Now that every team understands the value of analytics for baseball, the competition now is focused on how teams put the data to use.

It’s not only led to the last two (maybe three) World Series champions, but also made baseball one of the most interesting sports to follow.

2018-04-16T16:29:47+00:00
';