added 6/21/2005 by Scott Barzilla
As many of you know by now, I like to play with numbers. I’m not a mathematician, so my glee doesn’t last very long, but sometimes it does last long enough to come up with something. Before I move on to this new theory of mine and how it applies to the Astros, I need to go into some history. In the 19th century, Harry Chadwick developed baseball’s first statistics as a way for teams and fans to keep track of individual player performance. It wasn’t enough just to keep score.
Since then, baseball has been more numbers driven than any other sport. Everyone knows numbers like “56”, “755”, and “511”. I would dare fans of other sports to produce as many memorable numbers as the ones we have in baseball. Unfortunately, this works against meaningful analysis sometimes. Baseball fans trust traditional statistics (or metrics as we call them in the sabermetrical community). The number one metric on our “most wanted” list is batting average. Batting average is easily the most misleading offensive metric, but it is so popular because it is so easy to understand.
Famous baseball statisticians (or sabermetricians) like Bill James have been fighting this bias for years. He’s developed several new metrics personally to combat this bias. One of these is called secondary average. It basically calculates everything that batting average doesn’t (TB-H+BB+SB-CS)/AB. However, in his last historical abstact he admitted that run production actually fits somewhere around the midpoint between batting average and secondary average. Therefore, I decided to combine the two.
What this means is that we can have a metric that “looks like batting average” but means so much more than batting average. In fact, it adds in batting average, on base percentage, slugging percentage, and base-running into one number that looks like batting average. I’ve used the term “adjusted” because you can divide the number by the league average and multiply it by any number (I’ve chosen .262 because it is the major league average for batting average since 1871). However, the numbers I’m going to show you have not been altered because they aren’t being compared to anything. Here are the Astros numbers after the Kansas City Royals series.
AB BA SEC ARV Craig Biggio 242 .264 .264 .264 Willie Taveras 236 .288 .195 .242 Carl Everett 224 .237 .201 .219 Morgan Ensberg 221 .271 .416 .344 Jason Lane 206 .243 .316 .280 Mike Lamb 146 .212 .240 .226 Brad Ausmus 140 .214 .150 .182 Lance Berkman 133 .241 .271 .256 Chris Burke 105 .181 .124 .153
This shows us something closer to what we are really watching on a daily basis. In the past several columns I’ve talked about up the middle offense versus corner offense and the science of building a batting order. To understand why the Astros have had trouble scoring runs we need to look carefully at the two factors separately and then together.
AB H BB SB CS TB AVG SEC C Brad Ausmus 140 30 16 1 0 36 .214 .150 2B Craig Biggio 242 64 14 6 1 109 .264 .264 SS Adam Everett 224 53 13 4 5 86 .237 .201 CF Willy Taveras 236 68 12 17 3 88 .288 .195 Total 842 215 55 28 9 319 .255 .211
These numbers are an interesting case study for us. To the untrained eye, it would appear as if the middles are doing okay. After all, three of them have more than 200 at bats and they are hitting better than .250 as a group. That’s nothing to write home about, but when you’ve scored one runs or less twenty-two times already you have to call that decent enough. Excluding hit batsmen, their collective on base percentage of .301. That obviously isn’t very good at all. That contributes heavily to the group’s collective .233 ARV. Let’s see how the corners are doing.
AB H BB SB CS TB AVG SEC 1B Lance Berkman 133 32 21 1 0 46 .241 .271 3B Morgan Ensberg 221 60 34 5 6 119 .271 .416 LF Chris Burke 105 19 6 2 3 27 .181 .124 RF Jason Lane 206 50 13 5 2 99 .243 .316 Total 665 161 74 13 10 291 .242 .311
I think the three of these guys embody exactly what I’m talking about. Lance Berkman, Morgan Ensberg, and Jason Lane are hitting a collective .254 (less than the middles) but they have a collective secondary average of .345. That’s a lot of hidden value right there. As a group, they have a collective ARV of .277. So, you see that we have a significant gap between the corners and the middles. That gap is widened when we remove Chris Burke from the situation.
Of course, this shouldn’t be a revelation to anyone. We talked at length in the last series about how we needed to beef up the middle of the diamond. The real key is to look at how Phil Garner is shaping up the lineup. Doing that is more difficult than it looks. In fact, it’s lot like catching that chicken from Rocky I. Garner has been shifting around the lineup trying to find a combination. The lineup shifting cannot have a positive effect on the individuals, but short of a trade, we need to find the best way to use nine guys.
The ninth guy is Mike Lamb. He has 146 at bats at first base and in the outfield (he’s also started a few games at third). The question is whether the Astros are better off with Mike Lamb at first and Berkman in left or Berkman in at first and Chris Burke in left field. As usual, there is more to consider than the numbers, but we’ll address all of that after we look at the current lineup.
AVG SEC ARV CF Willy Taveras .288 .195 .242 LF Chris Burke .181 .124 .153 2B Craig Biggio .264 .264 .264 1B Lance Berkman .241 .271 .256 3B Morgan Ensberg .271 .416 .344 RF Jason Lane .243 .316 .280 SS Adam Everett .237 .201 .219 C Brad Ausmus .214 .150 .182
Chris Burke sticks out like a sore thumb. The key is whether he should be replaced by Mike Lamb and his .226 ARV or not. There are three different levels to look at this at. The first level is the present. If the Astros want to win they need to have as effective an offense as possible. Mike Lamb would be the answer for goal. Yet, with the club eleven games under .500 (after the Royals series) they might need to think about the future.
The future has two different possibilities. First, Chris Burke could stick around and play regularly in 2006. If that is the case he needs to play everyday regardless of what he is doing. On the other hand, if the Astros plan to let Craig Biggio chase 3000 hits then the club might end up trading him. Would it be better to show teams how he has dominated AAA this season or how he has struggled at the big league level? These are the questions the Astros need to answer. If Burke stays, this is what the lineup should probably look like.
ARV CF Willy Taveras .242 2B Craig Biggio .264 3B Morgan Ensberg .344 1B Lance Berkman .256 RF Jason Lane .280 SS Adam Everett .219 C Brad Ausmus .182 LF Chris Burke .153
Is the team struggling because Burke is hitting second? I hope everyone is beyond that thought. No, replacing him with Lamb is not going to propel the team into contention either. At best, it might improve the lineup two or three games between now and the end of the season. That is of course assuming that both players continue on their present course. Even if the club moves to this lineup they still have a gaping hole in their lineup from the sixth hitter on. When you have three holes in your lineup (Lamb would be one too), you are going to have some games where it seems like you’re running in quicksand.
The point is not for the Astros to change their lineup necessarily because lineup changes don’t recover fifteen games in the standings. The point is to realize what you really do have (Ensberg, Berkman, and Lane) and what you don’t have (Everett, Ausmus, and Burke). You have to continually keep your eye on these guys throughout the season to see if they turn it around. After all, who would have thought three weeks ago that Jason Lane would have appeared on the positive side?
Scott Barzilla is the author of “Checks and Imbalances” and “The State of Baseball Management.”