In Part I, I made it clear as to why I do not hold the RBI statistic in high esteem. Here in part II, I will offer up three statistics that I do hold in higher esteem. (All three have been compiled and calculated by Fangraphs.) I will then use those metrics to explain why I chose Mike Trout over Miguel Cabrera as the best offensive player of 2012.
This stat is based on the core statistic called “wOBA” (weighted on-base average), so I will begin by explaining how wOBA works, and why it is superior to other hitting measurements like Batting Average (BA), On-Base % (OBP) and Slugging % (SLG).
First, I will point the most apparent deficiencies of the classic statistics listed above.
|What it does:||Primary Weakness:|
|Batting Average||Measures the rate of getting hits in non-walk plate appearances.||It gives equal value to every hit, be it single or home run.|
|On-Base %||Measures the rate of getting on base per plate appearance, including hits and walks.||Just like Batting Average, it gives equal credit for a walk, single, double, triple and home run.|
|Slugging %||Measures the total bases per non-walk plate appearance.||While it does give different credit for each type of hit (a double is worth twice as much as a single), the ratio is not accurate to real game play.|
What wOBA does is to give credit for each type of hit based on its actual value in terms of scoring runs in a season. In other words, based on actual on-field results, it gauges the likelihood of producing runs due to the occurrence of each batting event. What we see from wOBA calculations is that, while a home run is of course the most valuable thing you can do as a hitter, nonetheless it is not exactly twice as valuable as a double (as SLG would say it is). Think of it in terms of real baseball situations: with men on second and third, a hitter cranks out a home run. Three runs score. Now take that same situation: men on second and third, but the batter hits a double. More often than not, two runs would score. So not only is the run scoring differential between the homer and the double 3-2, but the runner who hit the double is now in scoring position. So the home run was indeed more valuable than the double, but not twice as valuable. wOBA has done this with every event at the plate, from getting hit by a pitch to walking to getting a single, double, triple or homer.
While wOBA and wRC+ (and Batting Average, OBP and SLG) do not take into account the context of a player’s at-bat, this “RE24″ statistic does exactly that. In fact, it is a fair approximation of “clutch” performance. The “24″ refers to the 24 difference situations in which a batter may find himself within a single inning: nobody on and nobody out, nobody on and one out, 2 outs and men on second and third, etc. Again, calculations (based on actual game results) have been done to determine the likelihood of runs scoring based on hitters’ various performances in each situation. In other words, with nobody out and a man on third, a batter will get credit for knocking the runner in, but since that achievement is a relatively common one, it is given only a moderate amount of credit. On the flip side, hitting a home run with two outs and nobody on gives the batter quite a lot of credit, because in that situation, scoring a run is generally very difficult to do. Players are likewise penalized for failing in situations that commonly lead to run production, and are penalized less for failing in less than ideal run scoring situations. Therefore a player who constantly hits home run or gets hits with men in scoring position will have a high RE24 score. So in its essence, this is not all that different than classic stat keeping. (Keep in mind that the difference is this: since #3 and #4 hitters come to the plate most often with runners on base, they have the most opportunity to accumulate RBI and RE24 points, but they also run the risk of getting negative RE24 scores, like if they strikeout or ground into a double play with a man on 3rd. There is no such deduction for failure in the RBI stat.)
Win Probability Added is the best “clutch” stat we have so far. RE24 (above) does a great job measuring a player’s ability to come through in scoring situations, but it does not measure the effect on any specific game. In other words, a three run home run in always a great thing; WPA will give a player more credit for hitting one in the 9th inning of a close game versus in the 6th inning of a blowout. Just check out the name of the stat: Win Probability Added (though it also Subtracts when players fail). A three run homer in the 9th inning of a close game vastly improves the team’s probability of winning, whereas the 6th inning three-run homer that makes the score 15-5 has a much smaller probability of factoring in the outcome.
The New, Improved Hitter’s Slash-Line: wRC+/RE24/WPA
As I explained above, wRC+ is a more thorough indicator of a batter’s production than Batting Average. In fact, the classic Batting Average and Home Run parts of the slash line are absorbed into wRC+, because it accounts for every type of hit. And you can think of RE24 as being a better version of a player’s RBI total; WPA as a better version of his RBI in the clutch.
The Best of the New
If you have seen my points leading up to til now, you will also understand why I favor Mike Trout as the best offensive player of 2012. Take a look at this:
***2012 MLB Top 5 wRC+***
1. Mike Trout (166)
Miguel Cabrera (166)
3. Buster Posey (162)
Ryan Braun (162)
5. Andrew McCutchen (158)
***2012 MLB Top 5 RE24***
1. Edwin Encarnacion (55.84)
2. Mike Trout (54.27)
3. Buster Posey (52.34)
4. Chase Headley (51.43)
5. Prince Fielder (50.59)
***2012 MLB Top 5 WPA***
1. Mike Trout (5.32)
2. Buster Posey (5.01)
3. Prince Fielder (4.93)
4. Andrew McCutchen (4.85)
5. Miguel Cabrera (4.82)
I would like to point out something that I think is essential to understanding the difference between WPA/RE24 and RBI. Take a look at WPA: Trout is four places above Cabrera. However, Cabrera actually accumulated more positive WPA points than Trout; the catch is that, since he hit 3rd in a powerful lineup, and had so many opportunities in key situations, he was also prone to fail more often than Trout did. In fact, his failures counteracted his successes to the point of knocking him below Trout. Of course, don’t get me wrong – Miguel Cabrera still finished 5th in all of baseball in WPA, and was tied for first in wRC+. He had an amazing year. Trout, however, was better.
I won’t even bother arguing about the other aspect of offense in baseball: base running. Mike Trout blew away the entire MLB in stolen bases and in Fangraphs’ “Baserunning” score, while Miguel Cabrera is notoriously slow, grounding into double plays and racking up fewer stolen bases in his 10 years in the big leagues than Trout had this year alone.
Wrapping it Up
I hope that any reader who actually read through this two-part post is able to see where I coming from by using these new statistics. My intention – and the intention of the people who invented these stats – is not to direct attention away from the importance of scoring runs and winning games; rather, it is to assess the accuracy of the old stats in judging how those runs are scored, and to replace inaccurate stats with better ones. If we are to venerate the best ballplayers, don’t we want to be aware when we are under- or overestimating their contributions?
Finally, look at the big picture when considering the new sabermetrics. If these “crazy computer” numbers are really that off-the-wall, then how do you explain Fangraphs’ list of the best hitters of all time (according to wRC+)?
1. Babe Ruth
2. Ted Williams
3. Lou Gehrig
4. Rogers Hornsby
5. Barry Bonds (yes, I know he cheated)
Those names sound a whole lot like the ones I always heard about growing up. The numbers are different, but the best still rise to the top.