Sunday, October 16, 2011

Friday Baseball Post

The Rangers, to their second consecutive World Series. Congratulations

The last team to lose and then win a WS in consecutive years? That would be the 1988 and 1989 A's. Before that, you got back to the 1976/1977 Yankees. The last team to lose twice in a row? The 1991 and 1992 Braves. An American League team? Well, the 1963 and 1964 Yankees, but they had just won the previous two years.

In the history of the American League, until now only the Yankees, Blue Jays, A's, Orioles, Tigers, Twins (as the Senators), and Red Sox had previously won back-to-back pennants. Oddly enough, all but the A's and Orioles did it the first two times they ever won the pennant. So the Rangers are the eight AL team to join the club.

In looking that up, I learned something I don't think I ever knew. You all know that Philadelphia won the 1905 pennant and then were clobbered by Christy Mathewson and the Giants in the famously deadball WS, right? What I did not know is that the Athletics won the same number of games as the White Sox that year, but played four fewer games; had the rules required everyone to play out the schedule, Chicago would have had two games remaining and Philadelphia 6.

Getting back to consecutive pennants, all 8 of the NL franchises that began the 20th century have turned that trick, although the Phillies hadn't done it before the current team in 2009-2010. However, none of the NL expansion teams have done it, while this is the second AL expansion team to make it.

There has never been a WS between two expansion teams.


  1. Since my hometown team got eliminated, I really don't care who won the Playoffs, and then the Series. That said, every time Fox would do a cutaway to Bush, smiling away at every Rangers run (and there were MANY), I suddenly wanted the Tigers to win, very badly. Sadly, they didn't.

    While I appreciate having a respect for the office of the President, no matter what your politics are, part of me wonders what parallel universe I've fallen into, seeing a self-admitted war criminal sitting there, enjoying life as if he didn't nearly destroy the United States of America.


  2. Interestingly, the Rangers managed a grand total of 0 postseason series wins in their first 49 years of existence; in the past 2 they now have four. Is this just random? For me, a puff piece on the Rangers clubhouse during last year's postseason may give a clue to their newfound success.

    Washington's clubhouse has a whiteboard reminding players that their goal is ~17 "productive" at-bats a game. There are several ways to get a productive Ranger at-bat, including all the ones the sabermetric guys love (hit, walk, HBP). Notably, Washington includes three additional ways - not part of the sabermetric canon - that may offer a clue about succeeding in a short series.

    You can also have a productive at-bat by reaching on an error, moving runners along, or forcing a pitcher to throw 8 pitches. I find that third one particularly interesting; if you're facing Lincecum in Game 7 of the WS, its quite possible that an AB that eats up 10% of his alloted pitches, ending in an out, is as valuable to your team as a seeing-eye single on the first pitch of the at-bat. The error one is interesting too - the sabermetric guys don't like to count errors, but tell a Brewer fan today that errors don't have the same effect as hits in a critical game.

    Further, Washington's system reflects the fact that winning a critical game (or games) is a team effort, while winning one or two more of 20 random games in July is an aggregation of individual statistical profiles. I suspect that once the sabermetric focus changes to short series, people will respect the work Washington does (not so much now, as he is a bit too hyper for the cool professorial types in the sabermetric club).

  3. "There has never been a WS between two expansion teams"

    That amazed me. I figured that in 50 years it had to have happened *once.* But I checked (, and they helpfully put an [x] by each expansion team), and sure enough, there have been about 20 WS's since 1969 that featured at least one expansion team, but there was always one of the "original 16"opposing them.

    In that case, Go Brewers ;-)

  4. That should be 2008-2009 for the Phillies.

  5. Forget the sabermetrics. Nelson Cruz is worth any boatload of sabermetricians.

    And rainfall interrupting both of Verlander's Game 1 starts had a hand it as well. The sabers are as valuable as the weathermen, basically. A wounded Tigers' lineup HAD to count on Verlander winning 2 games in that series, and resting the bullpen to ready for the Rangers' murderer's row in later games. Didn't happen.

    The Rangers have a powerful lineup. It's that simple. You take a powerful lineup and add in 2-1/3 sloppy innings of Scherzer's mechanics, and 4 walks, and then add in a Cruz, and you get yourself a WS appearance.

    Kudos to the Rangers. They are the real deal. If they get their starting pitching sorted out, they can win it all.

  6. Ugh. I can't believe I got the Phillies years wrong. Awful.


    I'm not sure where you get your ideas about a "sabermetric cannon," but as far as I'm concerned the correct sabermetric position is that errors should be discontinued; ROE is certainly something that analysis count, and consider a hitter talent. Same with the pitch count thing.

    Advancing runners on outs is another story. I think the general sense is that it's not a skill, and it certainly shouldn't be (overly?) rewarded. Put it this way: you're more likely to move runners with a SH, so if the manager wants to trade an out for moving runners, he should be giving the bunt sign.

    It's worth noting, BTW, that over the season the Rangers were very good on both sides of the game, but overall their run prevention was (even) better than their run scoring -- it was masked a bit by the ballpark.

  7. I also was surprised no World Series with 2 expansion teams. Though the Orioles were the St. Louis Browns, something about changing the nickname along with the city makes it feel as if it were an expansion team. So the 69 Series was with 2 teams that basically hadn't existed in the first half of the century. In the same way that if the Nationals won the World Series next year, it would feel weird to talk about how the drought is over and they finally did it.

    (If you'll excuse my going off-topic...
    At plumline, Greg S keeps talking about how Obama has admitted things weren't better than they were four years ago. But he's been president for less than 3 and in his "no" answer specifically referred to things that happened in the past 4 years but before he was president. I'm not saying a year from now, he won't also admit we're stuck and blame the damage on things from 2008 and GOP obstruction, but I thought it was worth pointing out the nuance anyway.)

  8. Jonathan,

    Thanks for taking up my argument. I think we might be talking around each other a bit here. I'm specifically attempting to determine whether Washington's approach makes sense in trying to win one (or a few) pressure-filled games, as opposed to the Grabiner/James approach of finding the best guys for the long, undifferentiated part of the season.

    So for example, while I've never heard Washington interviewed on the topic, I suspect his inclusion of runner-advancing outs is not a reflection of the fact that he wants them, necessarily, but rather that they count toward the total of at-bats that will help the Rangers win TODAY. (Surely Washington would trade every runner-advancing out for a two-run homer if he could, no?)

    So we wouldn't necessarily expect him to order up sacrifices, as he isn't seeking that (the way a Grabiner or James-type seeks certain profiles in preferred players in the sweep of the long season). Its just that a runner-advancing out is among the good things that can happen. Might be a sacrifice, but not necessarily: consider Albert Pujols coming to the plate with a runner on second and nobody out.

    Pujols is a dead-pull hitter; the 2nd baseman is often behind the bag. In a system like Washington's, might Pujols be more motivated to go inside-out on a pitch and hit it toward right field? Could be a pretty good play, especially late in a tie game with the situation as described.

    Also, I agree that if we're doing the Grabiner/James exercise, sure we should exclude errors in deciding who we want for the undifferentiated sweep of July games. But if we have to win today, reaching on an error is one of the good things we need to happen. Surely LaRussa would say the four Brewer errors in the pivotal game 5 were foundational to that critical Cardinal win, no? If we need to win this game, reaching on errors is good, no?

    And while its hard to control an error the way you can the stats that sabermetrics embraces, most errors do seem to arise from pressure of close or unusual plays. Unless the infielder is Steve Sax, a routine grounder right at him is not gonna result in an error.

    So again - thinking strictly in the sense of "we need to win this game" - rewarding hitters for reaching on errors may help the cause, may result in the hitter running or sliding a little bit harder, which may ratchet up the pressure on the defender a critical little bit, which may make it more likely that victory-causing good things will happen.

    I may be taking this too far, of course. Don't know if I would try to publish these ideas in a refereed baseball journal, for example. But it does seem to me that there is something there, that Washington is on to something, that his postseason success is not accidental - that there is more to understanding (high-pressure, clutch) baseball then the massed data approach of Grabiner, James, et al.

  9. Again, I don't think that anyone -- or at least not mainstream sabermetric analysis -- holds the position you think they hold. I suspect I'm not writing very clearly, so I'll give it another go.

    Reaching on errors (ROE) is something that sabermetric types *do* count; in fact, my position would be that we should just get rid of errors altogether. ROE is definitely a batter skill.

    But the second order thing is that the real sabermetric position on all of this is that you can't just assume stuff like that; you have to actually study the evidence. It makes sense that ROE is a batter skill, but that doesn't really count; what counts is what you find once you study it (and, as I said, it turns out that it is a skill, so it should count when you measure how good a hitter is).

    Which gets to what you're saying. It may make sense to you or others that there are different things needed to win one high-stakes game or a handful of high-stakes games or games against very good teams or whatever, but the sabermetric position on it is that our intuitions about these things don't really count; we need to actually study the evidence. If it does turn out that there are important differences between October baseball and June baseball, then by all means they should be identified and smart teams should act on them. But we also know that sometimes our intuitions about these things turn out to be wrong. And if that's the case, well, we have to respect that, whatever we originally thought was the case.

    Now, in real life, there are going to be things that are awful difficult to determine, and actual managers and GMs will need to make choices despite that, and no good analyst would deny that. But there's nothing about sabermetrics that includes an assumption about June being like October, or there being one best kind of player or best way of doing things.

  10. Thanks again for clarifying; I definitely take your point that our intuitions can often be wrong - certainly my sense about what Washington is doing right is only that, a sense, and his recent success may just be an example of the old availability heuristic.

    It will be interesting to see whether and what type of data can be gathered to validate what is, admittedly, a view that is somewhat difficult to subject to data collection.


Note: Only a member of this blog may post a comment.

Who links to my website?