February 19, 2004
Cubs vs. Astros
The Baseball Savant has a lengthy post on how the Maddux signing impacts the NL Central race. He also writes me with this:
I read the other day your "musing" about Maddux and the notion that you need both pitching and offense to win.
I couldn't agree more. I ran the "Expected Performance Values" for the Houston starting 8 and pitching staff and did the same for Chicago. I then figured win shares for each player based on those projections. The Cubs did indeed have a better pitching staff by about 10 win shares, but the Houston offense had a better offense by about 23 win shares giving Houston a +13 net win shares over Chicago. This equates to about 4-5 game lead over Chicago.
What I think is amazing is that last year, according to Pythagorean, Houston was a 7 game better team than Chicago even though they finished 1 game out of the Central and lost.
Now, do you think that if Houston does beat out Chicago for the NL Central, will everyone jump on the pitching bandwagon and say that the additions of Clemens and Pettitte really put Houston over the top seeing how that they were the 2 big additions to Houston and completely ignore the fact that if Houston does beat out Chicago it will probably have more to do with the fact that Kent will have been healthy all season, Berkman should have a true Berkman-like season, Ensburg will get to start everyday, and Oswalt will be healthy? What do you think?
Lots of people will jump on the pitching bandwagon, because the pitching moves were made the headlines over the season. I do wonder however, if they will jump off the "you need a great closer" bandwagon, since Wagner has gone to Philadelphia.
But the Savant points out something else very interesting, and that's how much luck plays into a season. In this case, luck was an eight game swing in the standings. And that should worry Chicago fans, because luck has a way of evening out over time (I believe it's called regression to the mean). Neither of these teams stood pat, however, and both are improved. It should be a great race in the NL Central.
Was it all luck? Or did the managerial stylings of Jimy Williams have a little to do with it?
Does that then imply that Dusty Baker is as good a manager as everyone wants me to believe? I know I can't handle the heat like some of the brothers, but...
I don't know if it means anything, but I looked at the numbers for various managers and how their teams performed vs. Pythagorean Projections. Jimy Williams came out behind in almost every season he's had, while Dusty followed the normal progressions. Joe Torre, on the other hand, has had the Yankees about 3-4 games ahead every single season over the last 6 or 7. Just FYI
I don't understand why it so hard to understand. In order to win you need to score more than you allow. If you have pitchers that allow a lot of runs your offense needs to work harder, if you have stingy pitchers your offense doesn't need to be as productive. And vice versa, just having great pitching doens't mean you don't have to score runs. And scoring a lot of runs doens't mean you don't have to have competent pitchers.
Jeez how can anyone build a team only foucsed on one side of the equation????
Careful... Regression to the mean is true in the long run. It has absolutely nothing to say about the Cubs in one season. The Cubs are equally likely to finish higher than their run differential merits this season as they were last season. People lose a lot of money gambling thinking this way ("I'm due!"). The seasons are independent of one another (mostly, and in a mathemetical sense) and the outcome of one season has no effect on the outcome of another.
Chicago fans *should* be worried that their team may not be as strong as their record would indicate. They should *not* be worried because last season's luck is unlikely to be repeated.
If you crunch the numbers, yes the 'stros have a the protencial of out doing the Cubs this year.
But luck has nothing to do with it. That's like saying Bartman lost game 6 for the Cubs. It just isn't true.
As for luck and the Cubs, 95 years, we're due!
I was doing some random checking of the various numbers regarding the Win Share totals of the Cubs and Astros last year, and I came across something strange. The Astros scored about 80 more runs than the cubs, and allowed 6 fewer (although more earned runs). This is why there is a 10 win swing in the Pythagorean Standings. However, James's WS totals list the Cubs and Astros as being almost identical in batting, pitching and fielding (i think all the numbers were within 3-4 WS of each other.) So I started thinking "Something isn't right here. James came up with both formulas. One is telling us that the Astros were 9 games better than the Cubs, and the other is saying they were basically equal." I knew that chance might come into play, but that still didn't explain things. I've read in other places criticisms that have said that James doesn't take into account Pythagorean standings, and that this is an oversight. Then I came across this page on Baseball prospectus
What I discovered is that, the basic Pythagorean standings have been adapted. They have been expanded to include more information, and have been taken to the second and third order. In the first (standard) order, the Stros have a 9 game edge. By the second order, it's down to 2 games. And by the third order, it's at 1.3 games.
When I read this, what it's telling me is that, in fact, the teams were much closer than the run totals would have you believe. Basically, the Astros should have finished a game ahead, and instead they were a game behind. This jives with the Win Share team totals, and makes me believe that no adjustment is needed. The totals given to the players on both teams last year were accurate, at least from the team level (there are still imperfections in the system, no doubt, but they are smaller, mostly with the fielding and rating of relievers)
Of course, what this will directly mean is anyone's guess. But if you believe, which I think I now do, that the Third Order Pythagorean standings are much better than the First Order, then a wild "luck swing" for the Astros should be expected. Both teams were about equal, and finished that way. On to the new season!