PDA

View Full Version : How far did the Sox fall? Some historical perspective . . .


Johnny Mostil
09-28-2006, 10:27 AM
Seeing another thread on the "disappointing" '06 season made me wonder: just how far did this team fall off from the standard it set last year? My answer: maybe not as far as we think.

To be sure, I thought this team could win about 95, which might--or might not--have been enough to pass the Twins (who now have 94 wins). So, yeah, I'm disappointed in the team, and I don't intend this to be a pollyanna post. And nothing below contradicts any opinion on what perhaps the team should have done instead to retool last winter.

But the reality is teams that have very good seasons almost always fall off the next year. And, from what little I've researched so far, I say the fall for the Sox hasn't been much worse than that for other championship teams.

First, consider the regular-season records of the past 25 World Series champions. I constructed a table of the past 25 World Series champions, showing their regular-season record in their championship year, their regular-season record the following year, and the difference between the two years. (I'd be glad to post this if I could figure out how--any volunteers to help this ignorant soul?)

Only four of the past 25 champions--the '87 Twins, the '96 Yankees, the '00 Yankees, and the '01 Yankees--improved their record the following year. (None of them repeated, and only one even returned to the LCS).

More to the point, if you were to rank the 25 teams by "drop-off"--i.e., by their winning percentage in the subsequent year minus their winning percentage in the championship year--you'd see the Sox rank 13th, or right in the middle. I might, if I get time, (or if there's interest), take this back to the first Series. I'm guessing I'll find more of the same--i.e., very few Series champions improve, and the Sox drop off is about in the middle.

Second, consider the teams that won at least 99 games, as the '05 Sox did. As near as I can tell (and I only eyeballed this on baseball-reference.com, so corrections would be welcome), only ten won more the next year: '08 Cubs, '11 Giants, '30 Athletics, '38 Yankees, '41 Yankees, '41 Dodgers, '53 Yankees, '89 Athletics, '97 Braves, and '01 Athletics. I suppose a better standard would be to see how teams playing at least .600 ball (the '05 Sox played .611 ball) did the following season. I'm guessing I'd still find considerable "regression to the mean."

I suppose the good news is there's nothing to keep the Sox from bouncing back next year--and that other teams are now more likely to "regress to the mean." "Regression to the mean" isn't a statistical law, of course, but it does seem to explain many things . . .

downstairs
09-28-2006, 10:49 AM
Oh, golly.

Regression to the mean cannot be used to prove correlation between two separated events that, in reality, have nothing to do with each other. (e.g. two distinct baseball seasons)
It's just a fancier way of saying "they're due."

"Regression to the mean" is Total BS when it comes to human events like baseball. That's because the mean is based on past events- but the mean can change based on future events.

The real reason they Sox sucked this year is because they sucked. Plain and simple. They were a BETTER team than in 2005 and played WORSE.

If you want to go through history, find all the teams that were the best in baseball, but choked and didn't even make the playoffs. Find all the teams that routinely go

Johnny Mostil
09-28-2006, 11:03 AM
Oh, golly.

Regression to the mean cannot be used to prove correlation between two separated events that, in reality, have nothing to do with each other. (e.g. two distinct baseball seasons)
It's just a fancier way of saying "they're due."

"Regression to the mean" is Total BS when it comes to human events like baseball. That's because the mean is based on past events- but the mean can change based on future events.

The real reason they Sox sucked this year is because they sucked. Plain and simple. They were a BETTER team than in 2005 and played WORSE.

If you want to go through history, find all the teams that were the best in baseball, but choked and didn't even make the playoffs. Find all the teams that routinely go

Well, OK, I probably shouldn't have used the term "regression to the mean," which can be BS for more than just human events. But I am curious how you can say the '06 Sox were a better team than '05 but played worse. (I may even agree with you; I'd just like to know how you justify the statement.) That they played worse--well, that's obvious. But why were they "better"?

itsnotrequired
09-28-2006, 11:18 AM
Well, OK, I probably shouldn't have used the term "regression to the mean," which can be BS for more than just human events. But I am curious how you can say the '06 Sox were a better team than '05 but played worse. (I may even agree with you; I'd just like to know how you justify the statement.) That they played worse--well, that's obvious. But why were they "better"?

"Better" is a relative term but if you want to go the stats route, the 2006 team (when compared to the 2005 team) had a higher average, more hits, more runs, more RBIs, more doubles, more HRs, higher OBP, higher SLG, etc. Even situationally, they had a higher average with RISP, higher average with RISP and 2 outs, higher average early in the game, higher average late in the game, etc. The only offensive stat that is really down is stolen bases.

If you look at pitching, the numbers are the exact opposite. I would say they were a better team offensively and a worse team defensively and pitching.

Johnny Mostil
09-28-2006, 11:32 AM
"Better" is a relative term but if you want to go the stats route, the 2006 team (when compared to the 2005 team) had a higher average, more hits, more runs, more RBIs, more doubles, more HRs, higher OBP, higher SLG, etc. Even situationally, they had a higher average with RISP, higher average with RISP and 2 outs, higher average early in the game, higher average late in the game, etc. The only offensive stat that is really down is stolen bases.

If you look at pitching, the numbers are the exact opposite. I would say they were a better team offensively and a worse team defensively and pitching.

Fair enough. I just don't know how that translates into how many games they "should" have won. (I know the '05 team won 8 more games than their Pythagorean record indicated. As near as I can tell, the '06 team has, so far, won 2 more games than their Pythagorean record indicates. So the '05 team had a better Pythagorean record and played even further above it. But I don't have much opinion on the validity of the Pythagoran record.)

downstairs
09-28-2006, 01:39 PM
Well, OK, I probably shouldn't have used the term "regression to the mean," which can be BS for more than just human events. But I am curious how you can say the '06 Sox were a better team than '05 but played worse. (I may even agree with you; I'd just like to know how you justify the statement.) That they played worse--well, that's obvious. But why were they "better"?

Obviously not "better" from stats and results. And, sure, "on paper" is meaningless.

But, lets face it... we brought pretty much all of the World Champ team back, except added a Hall of Fame DH who had a Hall of Fame year.

We had no lasting injuries this year or last.

Put it this way... say we sweep MIN. That's 91 wins. In no way, shape, or form are we 8 games worse this year.

Johnny Mostil
09-28-2006, 02:01 PM
Obviously not "better" from stats and results. And, sure, "on paper" is meaningless.

But, lets face it... we brought pretty much all of the World Champ team back, except added a Hall of Fame DH who had a Hall of Fame year.

We had no lasting injuries this year or last.

Put it this way... say we sweep MIN. That's 91 wins. In no way, shape, or form are we 8 games worse this year.

I more or less agree with the bold above, but, alas, I do think this team was slightly worse--maybe four games worse, which still might not have been enough to get to the post-season.

Somewhat off-topic, but while a lot of discussion on these boards (or at least the discussion I've read) seems to focus on the bullpen, I'm surprised the starters haven't received more blame. Was last year freakishly good for them? Or was this year freakishly bad? What can reasonably be expected? I know, if we could answer that question, we'd all be GMs . . . but ERA last year was the lowest in the past three for all four, and Buehrle and Garcia had lower ERAs in '04 than '06.

I even think there's something to your original point that sometimes they just plain sucked (in ways that can't be measured . . .). I'm just not sure how much suckage is reasonable in any given year. (Well, none, of course, but it'll still happen . . .)

fusillirob1983
09-28-2006, 02:24 PM
Was last year freakishly good for them? Or was this year freakishly bad? What can reasonably be expected?


Based on all the starters' ERAs the past few years, I'd have to answer those questions with:

Yes
Yes
I have no idea

Iwritecode
09-28-2006, 03:16 PM
Somewhat off-topic, but while a lot of discussion on these boards (or at least the discussion I've read) seems to focus on the bullpen, I'm surprised the starters haven't received more blame. Was last year freakishly good for them? Or was this year freakishly bad? What can reasonably be expected?

Yes, the starters AND the bullpen were pretty dominant last year.

This year the starters weren't nearly as good, which was too be expected. Looking at their numbers, they weren't horribly bad (except for Buehrle). They were pretty average, maybe a little above overall.

The bullpen on the other hand was absolutely horrible this year. If they could've held onto even half the leads they blew this year we'd probaby be in first right now.

palehozenychicty
09-28-2006, 03:40 PM
The bullpen on the other hand was absolutely horrible this year. If they could've held onto even half the leads they blew this year we'd probaby be in first right now.

Therefore, the 8 to 9 game difference in record is made up. Improved middle relief, more speed/contact from either LF/SS will equal playoffs. Watch.

eastchicagosoxfan
09-28-2006, 06:28 PM
Seeing another thread on the "disappointing" '06 season made me wonder: just how far did this team fall off from the standard it set last year? My answer: maybe not as far as we think.

To be sure, I thought this team could win about 95, which might--or might not--have been enough to pass the Twins (who now have 94 wins). So, yeah, I'm disappointed in the team, and I don't intend this to be a pollyanna post. And nothing below contradicts any opinion on what perhaps the team should have done instead to retool last winter.

But the reality is teams that have very good seasons almost always fall off the next year. And, from what little I've researched so far, I say the fall for the Sox hasn't been much worse than that for other championship teams.

First, consider the regular-season records of the past 25 World Series champions. I constructed a table of the past 25 World Series champions, showing their regular-season record in their championship year, their regular-season record the following year, and the difference between the two years. (I'd be glad to post this if I could figure out how--any volunteers to help this ignorant soul?)

Only four of the past 25 champions--the '87 Twins, the '96 Yankees, the '00 Yankees, and the '01 Yankees--improved their record the following year. (None of them repeated, and only one even returned to the LCS).

More to the point, if you were to rank the 25 teams by "drop-off"--i.e., by their winning percentage in the subsequent year minus their winning percentage in the championship year--you'd see the Sox rank 13th, or right in the middle. I might, if I get time, (or if there's interest), take this back to the first Series. I'm guessing I'll find more of the same--i.e., very few Series champions improve, and the Sox drop off is about in the middle.

Second, consider the teams that won at least 99 games, as the '05 Sox did. As near as I can tell (and I only eyeballed this on baseball-reference.com, so corrections would be welcome), only ten won more the next year: '08 Cubs, '11 Giants, '30 Athletics, '38 Yankees, '41 Yankees, '41 Dodgers, '53 Yankees, '89 Athletics, '97 Braves, and '01 Athletics. I suppose a better standard would be to see how teams playing at least .600 ball (the '05 Sox played .611 ball) did the following season. I'm guessing I'd still find considerable "regression to the mean."

I suppose the good news is there's nothing to keep the Sox from bouncing back next year--and that other teams are now more likely to "regress to the mean." "Regression to the mean" isn't a statistical law, of course, but it does seem to explain many things . . .

I think your observations indicate that it requires a special set of circumstances to repeat. Why? Because winning it the first time takes some luck, along with timely hitting ( ..." It's not what you hit, but when you hit it " ), and most importantly, defense and pitching. In 2005, the Sox won it without offensive career years from anyone, except perhapd Pods. Tell me in April that Dye, Konerko, Thome and Crede would have the seasons they had, and I'm betting the farm on a repeat. Look at the Cardinals. Over 100 wins two consecutive seasons, and no World Series title to boast of. 2005 was really special. I'm rambling off to nowhere now. The 2006 Sox didn't win, I'm moving on to 2007.

wassagstdu
09-29-2006, 07:55 AM
They were a BETTER team than in 2005 and played WORSE.

Classic.

A. Cavatica
09-29-2006, 09:59 AM
It's been asserted by statheads that the '05 Sox won more games than they should have, based on runs scored and runs allowed. That record in one-run games was insane, after all. And Pods and Blum homering in the World Series? Off the charts.

Luck is a difficult thing to study because so much of baseball is mental. If a team believes it has luck on their side, it might just play better, "making its own luck". Conversely, if a team is playing poorly and doesn't get any lucky breaks for a while, they can start believing they're snakebit, and play worse.

It seems to me that '06 was a complicated interaction between a number of players having bad years, a manager having a bad year, and the lucky breaks going the other way than they tended to in '05 -- plus a Detroit team that didn't blink first and a Minnesota team that got on a roll. The Sox formula just evaporated.

ondafarm
09-29-2006, 10:17 AM
A couple of things here from someone who works with statistics every day.

Regression to the mean is very much the appropriate phrase. The 2005 and 2006 performance of the same team, like the White Sox, is very dependent, because you have mostly the same players and the same coaching staff. And the four teams that we played the most games against, the other four AL Central teams, are also very similar groups from 2005.

Starting and relief pitching are also very dependent. If the starters have a good year then the relief pitchers have to pitch far fewer innings and they are fresher for those they do need to pitch and perform better. If your starting pitching performs well, then you relief pitching likely will.

soxinem1
09-29-2006, 12:55 PM
It's been asserted by statheads that the '05 Sox won more games than they should have, based on runs scored and runs allowed. That record in one-run games was insane, after all. And Pods and Blum homering in the World Series? Off the charts.

Luck is a difficult thing to study because so much of baseball is mental. If a team believes it has luck on their side, it might just play better, "making its own luck". Conversely, if a team is playing poorly and doesn't get any lucky breaks for a while, they can start believing they're snakebit, and play worse.

It seems to me that '06 was a complicated interaction between a number of players having bad years, a manager having a bad year, and the lucky breaks going the other way than they tended to in '05 -- plus a Detroit team that didn't blink first and a Minnesota team that got on a roll. The Sox formula just evaporated.



http://www.paxtonphoto.com/People/Yogi%20Berra%20at%20Home.jpg
"I agree. Baseball is 90% mental -- the other half is physical."