Last year, the debate between sabermetricians and traditional baseball analysts took center stage when Miguel Cabrera and Mike Trout both had MVP-worthy campaigns, but made their cases in different stat columns. Today, in 2013, the debate is no less pertinent. I had a chance to chat with former MLB outfielder Gabe Kapler about the rift between the old and new camps, and which of these emerging stats he sees as useful.
Kapler’s extensive knowledge of the game provides XN readers with an informed and leveled commentary on the growing role of advanced stats in baseball. His insight also sheds some light on the growing debate between the old and new camp, and how some analysts on either side could benefit from being a bit less staunch in their opinions on which stats are more important when doing baseball’s other dirty work, punching numbers.
Over the course of twelve seasons in the Majors, Kapler played with six teams, most notably the 2004 Boston Red Sox — the team that ended the 86-year World Series Championship drought in Boston. I didn’t get into whether or not sabermetrics played a role in lifting the Bambino’s curse, but we did cover a lot of other, more relevant topics.
TL: The debate between baseball traditionalists and sabermetricians was ignited last year with Miguel Cabrera and Mike Trout both deserving of the award, but each side citing different reasons and statistics to prove their points. Where did you stand on that issue?
GK: Mike Trout was the better baseball player last year; Miguel Cabrera was undoubtedly the better hitter. This debate extends far beyond the stats, both sabermetrics and traditional. If you simply examine the issue on the surface, I believe Trout was and is more valuable to the Angels than Cabrera to the Tigers. The guy that significantly impacts all aspects of the game for me gets the edge. Obviously it’s our duty as students of the game to not get caught up in the Triple Crown nonsense as two of the three statistics, BA and RBI, are fatally flawed.
But this debate isn’t ultimately about stats, it’s about the ripple effect that is not yet and never will be quantifiable. For example, how much better a base-runner was Trout than Cabrera and therefore, how did that put pressure on opposing pitchers, thereby benefiting hitters like Albert Pujols who had the good fortune of hitting behind Trout in the lineup? How did the calm presence of Miguel Cabrera at the plate influence the young Tigers hitters that looked to him for guidance? As long as there are critical peripheral questions, it will be difficult to find just the right formula. That said, I encourage all of us to search and get closer to the right mix of spices.
TL: Are there other sabermetrics you could see as useful to determining a player’s potential output? If so, which ones and how should we be using them?
GK:I’m just scratching the surface as it relates to digesting sabermetrics. In a way, the majority of the baseball industry is in the same boat. What I’ve recently learned to lean on, with the help of some extraordinarily bright individuals, is wOBA on the offensive side, FIP on the pitching side and DRS when evaluating defensive players.
Figuring out how to apply this information is of paramount importance. I suggest we first make a hypothesis based on what our eyes tell us in conjunction with some basic, traditional stats. I’ll use Mike Trout as an example, who I believe to be baseball’s best player. To back up my claim, I start with the stat that I value most, wOBA works fine in this instance. If I’m able to confirm his dominance of this vertical, I move on. I’ll bring in defensive metrics, WAR, etc. to confirm my belief. At some point when I’ve aggregated the data, I’ll trust myself and be able to make my claim. But it’s absolutely imperative to not lean on one metric or another, as the end all, be all. In real estate it’s “location, location, location” in baseball it’s “sample size, sample size, sample size.”
TL: I’m sure you’re familiar with the outstanding databases at Baseball-Reference.com. They have begun adding some “clutch” stats. How do you feel about this? Is “clutch” something that can be measured?
GK: The issue I have with “clutch” statistics and the analysis of them is twofold; the small sample sizes and the definition of clutch itself are at the root of my beef. What feels like clutch to one may not feel so to another. Should clutch be based on the inning, the competition, the amount of television coverage, and the market? Some players may feel pressure around attaining their next contract; others may apply weight to winning a ballgame. What we are essentially trying to figure out is how a player performs under pressure, which is highly variable. I think we will get a better understanding, but we will fall short of true predictability. I’d like to think that if all else is equal, which it never is, we can use a metric like this to help us make a decision.
TL: You entered the Bigs in ’98. Between then and now, have managers begun to take advanced stats into account when making decisions in-game?
GK: Of course. Most progressive folks around baseball are at a very minimum taking advanced statistics into account if not relying heavily on them. I don’t remember a whisper about anything other than a traditional stat line in the clubhouse in 98, 99. I can currently make a strong case that only the most interested and intelligent players discuss advanced metrics now, in 2013. But front offices have no choice but to weed out close-minded individuals. An argument based mainly on what the eyes see and not backed up with data is less and less useful.
TL: BABIP – What do you think about the usefulness of this metric when it comes to projecting whether a player is due for a comeback year, or a rebound year?
GK: I think BABIP is pretty useful if we have- again-a large enough sample size. We need to be able to look at a players career BABIP and have a substantial track record in order to make a meaningful prediction on what’s coming. I think very generally speaking, guys that have high exit speeds and in some cases, lots of topspin on ground balls and good backspin on fly balls have better BABIP. It makes sense logically to take BABIP into consideration along with a myriad of other factors such as age and decline for older players, skill advancement for younger players and so on.
TL: Is it smoke and mirrors, or is there real value in determining batting average on balls in play?
GK: It’s definitely not smoke and mirrors. It allows us to get some assistance in our estimation on how hard balls are being hit as we perfect exit speed velocity calculations and measurements. It also allows us to laugh when traditionalists say “He puts the ball in play”. So what if he does put the ball in play if it rarely leads to getting on base? I’d rather have a guy see 7 pitches in an AB and K than weakly roll over the first pitch of an at bat and kill the momentum of a lineup.
TL: Back to WAR – Is this something you see sticking around, or do you think this newfound infatuation with advanced stats is something that will eventually take a back seat to traditional stats?
GK: There is no question that WAR will stick around. The question is how quickly we make alterations to the formula so that the metric is dependable. I don’t think we can use WAR as anything but a compliment in our evaluation arsenal at the current time. In the future, based on the speed of technology and our data acquisition, I can see us nearly perfecting the WAR metric.
TL: There’s a new television show on MLB Network where a sabermetrics junkie faces off against a traditional analyst. The show is based on their back-and-forth. There’s a real debate between these two camps. Talk about “eye-test” and “feel” vs. sabermetrics and advanced stats. Are there some aspects of the game that can’t be measured in numbers, and conversely, are there aspects of the game that could be better explored by punching numbers?
GK: There are absolutely factors that are immeasurable. Off the top of my head, it’s nearly impossible to quantify how a player’s management of relationships, positive or negatively impacts his teammates and staff. How much time does staff spend on management of this player that could otherwise be used to analyze game situations or the stats that we are discussing? A progressive baseball mind will unequivocally take scouting and analytics and combine the two. The more scientific our sport becomes, the more data is king. And data comes from everywhere including but not limited to scouting, player development, performance, the sabermetrics community, clubhouse reports on makeup, etc. What disappoints me most about the baseball world is the close mindedness that exists. Old school cats rolling their eyes at numbers and math, and sabermetricians totally discounting the eyes and ears of years of hands on baseball experience is flat out irresponsible.
TL: Any final thoughts?
GK: We have so much to learn. Our game is at a fork in the road. We can either bury our heads in the sand and fight the change or we can embrace it and get on board, knowing that baseball will be deeper and more interesting to analyze. Between the lines our men will still get dirty, I promise.
Follow Gabe on Twitter @gabekapler
