Name: Email:

Headline:

Introduction:
Using RCAP as an example, Scott runs through a sabertrical primer for fans who may not be familiar with the current state of baseball statistical analysis.

Column: HTML:
Last time I talked about potential trades the Astros could make and my column got a lot of responses. Most of my columns get some response and I thank those for sending their comments my way. I usually don’t respond to every comment unless they point out an error in what I said in the previous column. Unfortunately, people on some of the fan sites have been giving me a hard time recently for what I have said. That’s certainly their right to do and I can’t sit here and criticize more well-known columnists like Richard Justice and remain thin-skinned myself. However, it occurred to me that some of their comments have some validity. I had to factor them through the prism of style of substance (one said I was “retarded” and “pompous”), but they had a point. I have the habit of throwing out terms without giving them full definition. It happens all of the time. Many of you would be surprised to hear that I’m a school teacher, but this phenomenon happens there as well. Unfortunately, this isn’t a classroom and people can’t raise their hand and ask for immediate clarification. As it happens, most of us have things that are natural to us and come as a second thought. For others, this isn’t the case. What I thought I would do is go through a sabermetrical primer and use our current trade situation at the same time. Sabermetrics is not something I invented (or pulled out of my body somewhere) it has been around for 30 years or so and can be defined simply as “the scientific study of baseball statistics.” Sabermetrics has two main purposes: to accurately rank or rate players from the past OR to evaluate team performance in a systematic way to improve the team in the future. Last time I mentioned a metric (statistic developed by sabermetricians) called RCAP. RCAP stands for runs created above average at that position. The general idea is that you compare catchers to catchers and shortstops to shortstops and factor out ballpark norms to eliminate bias. So, if a player creates 50 runs as a catcher and is said to have a 0 RCAP then it means that the average catcher would have created 50 runs if he had played in that player’s home ballpark and had the exact number of plate appearances. People on the aforementioned site ridiculed RCAP and said that “if it says that Brad Ausmus is the reason for the Astros losing then it can kiss….well….you know” Let me address that sentiment with two points. First, the metric itself doesn’t prove or say anything. The person using the metric interprets it. I will do that in this column. Secondly, no one player can be blamed for a team being more than ten games out of first. Whitey Herzog once said (when he was in Kansas City in 1969) that he was “two players short of the pennant: Babe Ruth and Cy Young.” It was obviously hyperbole, but it was also incorrect. They wouldn’t have made it even with them. So, one sub-par player can’t be solely responsible. Before we begin we must start with the concept that batting average doesn’t tell us anything. Some people will respond (and have responded) to my thoughts that Brad Ausmus and Adam Everett are the weak links of the lineup by saying, “both of them are hitting nearly .250 and they are both exceptional fielders.” We will leave the fielding part of that equation (although important) for a different day. Let’s start with one easy statement: not all .250 batting averages are created equal. I think we can all agree with that statement. Therefore, we need to find out what is missing from batting average. What we find is that two things are missing from batting average: we don’t know what happens in the plate appearances where the batter doesn’t get a hit and we don’t know how many bases the hitter gets with his hits. This is why sabermetricians focus on slugging percentage and on base percentage. Let’s take a look at Jason Kendall, Brad Ausmus, and Jason Larue since Ausmus came back to Houston in 2004. <pre> PA H BB TB AVG OBP SLG Jason Kendall 2600 689 202 879 .298 .371 .380 Brad Ausmus 1914 416 147 564 .242 .304 .328 Jason Larue 1682 359 113 618 .242 .321 .416 </pre> Now, these numbers tell us a lot. They tell us that Kendall has played almost a season’s worth of games more than Ausmus. So, not only is Kendall durable, but he also has served as his own backup catcher. This means that his teams haven’t been subject to the Raul Chavez experience (or equivalent) that we have. They also tell us that Kendall got on more often and collected more bases. Jason Larue and Brad Ausmus have identical batting averages, so they are a better comparison. As you can see, Larue has a better OBP and SLG, so batting average isn’t everything. As anyone worth their salt will tell you, these numbers can be deceiving. Of course Ausmus is going to have a lower OBP since he has a lower average and of course he’s going to have a lower SLG with that lower average. What we should do next is determine how much is hidden in the higher batting average for Kendall. We can do that using two statistics. First, we can use a metric called secondary average. This is a metric that Bill James used to determine what a player does when batting average is factored out. If you have read my columns before you know I have used this. It is calculated with the following formula (TB-H+BB+SB-CS)/AB. The second formula (metric) we will use is something called bases per plate appearance. You simply add walks, hits, and HBP and divide by plate appearances. <pre> SEC BPA Jason Kendall .174 .429 Brad Ausmus .174 .350 Jason Larue .248 .450 </pre> What this tells us is that in this particular case, all of Ausmus’s deficiencies in comparison with Kendall come because of his batting average. So, is Kendall worth ten million dollars? Well, his BPA is better than positional average over the last four seasons is OBP is considerably better. In fact his career on base percentage is fourth in big league history among catchers with 5000 or more plate appearances. Two of the three above him are among the top ten players at the position (Mike Piazza and Mickey Cochrane). However, Kendall doesn’t have anymore power than Ausmus after you remove the bias of batting average. So, the long answer to the simple question is that Kendall isn’t worth ten million dollars, but he is at least above average at his position and he has one exceptional ability: the ability to get on base. If that is used in the right way (near the top of the order) then the lack of power can be lived with. Yet, Larue looks like the better pick there overall. As most people know, OBP and SLG is often combined into OPS. I’m not a huge fan of OPS because it favors batting average (both OBP and SLG use batting average in their formulas) but it is a decent barometer we can use without getting into complicated numbers. In order to give these two a frame of reference I will compare them to every regular catcher over the last four seasons (minimum 1500 plate appearances). <pre> OBP SLG OPS Ivan Rodriguez .364 .514 .878 Jorge Posada .384 .485 .869 Javier Lopez .347 .505 .852 Jason Varitek .359 .464 .823 Mike Lieberthal .350 .440 .790 Paul Lo Duca .343 .430 .773 A.J. Pierzynski .335 .438 .773 Charles Johnson .324 .432 .756 Jason Kendall .371 .380 .751 Ramon Hernandez .325 .420 .745 Damian Miller .332 .407 .739 Jason LaRue .321 .416 .737 Michael Barrett .313 .420 .733 Benito Santiago .312 .416 .728 Bengie Molina .298 .380 .678 Mike Matheny .301 .333 .634 Brad Ausmus .304 .328 .632 </pre> This is when we go back to basic statistics class. The mean is the average of all of the numbers. The means in this group are (OBP= .334, SLG= .428, OPS= .762). So, Kendall is basically an average offensive catching according to these metrics over the past four years. Ausmus has been a terrible offensive catcher over the last four years. If we go according to the mode we find that Kendall is the mode. Now, I haven’t brought defense into this equation on purpose because that would only obscure what we’re talking about here. There is no way in getting around this folks, Brad Ausmus is a terrible offensive player. Since, shortstop has been mentioned too I will go through the same with Adam Everett. Since Everett hasn’t been up as long as Ausmus we will go since 2002 when he first saw regular time. <pre> PA H BB TB AVG OBP SLG Adam Everett 1005 221 57 315 .257 .316 .367 Rafael Furcal 2059 526 161 773 .282 .340 .415 Placido Polanco 1704 450 95 663 .292 .342 .430 </pre> Many people consider Polanco to be a mediocre player that probably shouldn’t even be a regular. I’m not quite sure what that is based on, but what we see here is that he is the best of the three players. I was also called crazy for listing Furcal. Of course, if you read the last column carefully, I never said which trade I would make, but which players might be available. You could do a lot worse than either of those two. Like with the catchers, let’s take a look at secondary average and bases per plate appearance. <pre> SEC BPA Adam Everett .200 .426 Rafael Furcal .250 .480 Placido Polanco .211 .453 </pre> What we see here is that Furcal is more valuable once batting average is removed from the equation. However, what is most important is that the Astros would get an offensive upgrade with either Polanco or Furcal based on every metric we have looked at so far. People can say that RCAP is a bunch of bull, they can say that OBP, SLG, SEC, and BPA are a bunch of bull too. The problem is that when all of them say the same thing it is hard to call them all bull. <pre> OBP SLG OPS Alex Rodriguez .394 .611 1.005 Nomar Garciaparra .352 .516 .868 Miguel Tejada .350 .505 .855 Derek Jeter .371 .447 .818 Carlos Guillen .355 .450 .805 Edgar Renteria .362 .440 .802 Jose Valentin .301 .468 .769 Rafael Furcal .340 .415 .755 Placido Polanco .342 .406 .748 Alex Cintron .327 .419 .746 Angel Berroa .322 .415 .737 Orlando Cabrera .325 .409 .734 Omar Vizquel .343 .391 .734 Jimmy Rollins .325 .408 .733 Julio Lugo .333 .399 .732 Barry Larkin .328 .387 .715 Alex Gonzalez .295 .414 .709 Rich Aurilia .313 .396 .709 Jack Wilson .316 .386 .702 Alex Gonzalez .292 .409 .701 David Eckstein .344 .352 .696 Adam Everett .316 .367 .683 Cristian Guzman .303 .379 .682 Royce Clayton .314 .368 .682 Deivi Cruz .292 .388 .680 Cesar Izturis .294 .338 .632 </pre> The mean OPS among the group is .747 with .7335 being the mode. In other words, both Furcal and Polanco are better than the average regular shortstop over the past three seasons. Meanwhile, you can’t call Everett terrible, but he is significantly below average offensively. So, catcher and shortstop are two positions that can easily be upgraded. I’m not going to say yet which catcher or which shortstop we should target. A lot of that depends on age, contract status, and what it will cost us. Like I said at the beginning, I don’t care to respond to every negative comment anyone ever gives me, but hopefully a couple of things have been cleared up. It has often been said that people mock what they don’t understand. It is my responsibility to help people understand. Then, they can make up their own mind. Scott Barzilla is the author of “Checks and Imbalances” and “The State of Baseball Management.”

PASSWORD: