The way we analyze the game of baseball has gone under about as much change over the past 10 or 15 years as in the previous 100. Old-school scouts, and what they see with their eyes, are colliding with new-school “stat nerds,” and what they see with their calculators.
The meeting in the middle is the evolution of how we watch baseball. Somewhere between sights and stats, most likely, lies the best way to evaluate baseball, but each side of the aisle will argue that. These new metrics and advanced numbers have been around for a while but are continuing to gain steam and could make their way into the vernacular of all baseball fans — not just those who delve into the vast abyss of advanced analysis — sometime soon.
Here are four to watch.
Sabermetric fans hope this is Baseball Stat I: the stat to end all stats.
For some time, those who analyze the game have searched for the most complete, single number to evaluate player performance rather than relying on seven or eight. Most of baseball’s existence has been dominated by three: batting average, home runs and runs batted-in. Then OPS (on-base plus slugging) took over, the thinking being that if you combine how often a hitter gets on base with how much power he has, then “that” must be the best tool to evaluate players with. It’s a quality number but doesn’t value players who make their living with speed and defense (like Michael Bourn of the Indians).
WAR, or Wins Above Replacement, is a simple number with a complex (unknown) formula. It takes into account everything in baseball — hitting, power, running, fielding, throwing, ballpark factors, everything that can be measured — and then, using an enormous amount of metrics, data and history, finds what the “replacement level” player (average bench or minor-leaguer) should look like. A player’s WAR, then, is how many more wins he can give his team than if the most average player was at his spot instead. If your WAR is 3.5, then your team should expect three or four more wins than if they signed a replacement level player.
If in 10 years WAR is The Stat of baseball, Mike Trout is to thank. WAR had been around for a while but Trout’s historic season, one that was so complete it challenged a triple crown winner for the MVP, allowed it to gain real traction outside of the stat-nerd world of sabermetrics. Trout’s WAR of 10.7 (which means the Los Angeles Angels won 10 games just because of him) was one of the best seasons in history.
WAR isn’t without opposition. First, some naysayers claim you can’t truly capture baseball in any number, let alone one meant to encompass the entire game. Second, that the formula isn’t really known and, up until recently, wasn’t exact. FanGraphs and BaseballReference.com, the two entities calculating this number, finally agreed on what the “replacement level” line of production is. In years past, checking different sites meant you would find two different WARs for the same player. That seems to have been corrected. To take it back to Bourn, who doesn’t have a great OPS, his WAR since 2009 of 19.0 is third among all outfielders.
This one’s largely for the fantasy nerds. Batting Average on Balls In Play measures just that — what your average is only on the pitches that you put into play. It essentially measures how lucky a hitter is — how many bloop singles fall as hits or how many guys find the gaps as opposed to hitting line drives right at fielders, etc. BABIP has proved fairly accurate in determining which players might see a spike in batting average from year to year and which players should expect to see a regression. The league average BABIP is around .300, so if Player A had a BABIP of .385 in one season, he will be expected to regress to the norm the next year. Likewise, if Player B’s BABIP was .220, his average should increase. This also works the other way with pitchers, in that a high BABIP would be “unlucky.”
The Magic of 39
This was outlined in a story by Joe Lemire of Sports Illustrated. Over the past 22 seasons, if your team managed to bring 39 batters up to the plate (12 more than the minimum), you had a 50.1 percent chance of winning. This makes 39 the “magic” line of plate appearances between the probability of winning or losing. Following this strategy would be to follow the Moneyball line of thinking of Billy Beane, detailed in Michael Lewis’ book, not to risk outs and get as many batters to the plate as possible. Per this data, if your team faces between 27 and 38 batters, your chances of winning average out to 74.3 percent. Thirty nine or more? It drops to 31 percent.
Many numbers aim to find a true value of a player, meaning it tries to take out luck or other factors. Fielding Independent Pitching, by Tom Tango, does that. Tango came to the conclusion that a pitcher can only control three things: home runs, walks and strikeouts. Every other ball in play is up to his defense. This number cuts through all that to find how a pitcher really pitched. The formula assigns run values for home runs, walks and strikeouts to find a new number independent of a pitcher’s ERA. A good example is Cliff Lee of the Philadelphia Phillies midway through last season, when he was still somehow winless but his FIP showed he was, at the time, the 12th-best pitcher in all of baseball. Run support, fielding and dumb luck inaccurately affected his other, more well-known numbers and didn’t reflect his actual performance.