AL East - Red Sox
Central - Indians
West - Athletics
WC - White Sox
NL East - Mets
Central - Cardinals
West - Padres
WC - Phillies
DS Indians over Red Sox
Athletics over White Sox
Cardinals over Phillies
Mets over Padres
LS Indians over Athletics
Mets over Cardinals
WS Indians over Mets
I wanted to throw the Rangers in as winners of the West or WC, but just couldn't bring myself to do it. Their third order pythag was 87-75 last season, and with the moves they've made I definately think they'll be better this season. I think they'll just miss the playoffs but give Oakland a huge scare.
Tavish wrote:Anyone who is picking the Rangers to make the playoffs want to give some reasoning behind it? Millwood, Eaton, Padilla, Loe, Dominguez as a rotation doesn't really seem like much of an improvement over Rogers, Young, Park, Astacio, Dreese. The staff should be better but I just don't see it as a big enough of an improvement to close the gap.
Maybe I'm just skeptical after being burned by picking them to win the Wild Card last season, but this is a team that finished 16 games back of the Divison Winning Angels and Wild Card Red Sox last season and I'm curious if people are picking them expecting dramatic improvement or a serious falling off from the other contenders.
The Rangers were 8 games worse than their third order pythag. That might be inefficiency due to their home park and offensively skewed team, or they might have been really unlucky. The Rangers don't have a very scary rotation, but getting rid of some of the stiffs at the back end will be some significant addition by subtraction. Wilkerson and Kinsler will be better than the 4th outfielder and Soriano combo the Rangers used last season in center and at second. Young is still a defensive liability at short, but does hit well for a middle infielder. Dellucci and Matthews are underrated players when healhty.
I think they'll be in the 92-93 win range. Probably not good enough for the playoffs, but close.
Ender wrote:No I understand it doesn't take playtime into effect and sure its better than things like rototimes which are even worse, but ZIPS is really off on a lot of players because it doesn't understand why people have off seasons, its really bad at projections for rookies and for whatever reason its really bad at predicting pitching. Sorry, didn't mean to single out your post, been seeing a lot of people talking up ZIPS on another board and it was making me crabby, PECOTA and Forecaster are just so much better at rate stats.
I just put no value at all in what it thinks a whole team will do on the year.
I don't think that's a fair comparison. For one thing, other than Pecota, I don't think there's any good evidence that another system outperforms ZiPS. Certainly in the BP analyses, there's no real difference between BBHQ and ZiPS.
The other thing is that Pecota and ZiPS are really doing two different things. One projects the full distribution, while the other is a point estimate.
Tavish wrote:Anyone who is picking the Rangers to make the playoffs want to give some reasoning behind it? Millwood, Eaton, Padilla, Loe, Dominguez as a rotation doesn't really seem like much of an improvement over Rogers, Young, Park, Astacio, Dreese. The staff should be better but I just don't see it as a big enough of an improvement to close the gap.
Maybe I'm just skeptical after being burned by picking them to win the Wild Card last season, but this is a team that finished 16 games back of the Divison Winning Angels and Wild Card Red Sox last season and I'm curious if people are picking them expecting dramatic improvement or a serious falling off from the other contenders.
The Rangers were 8 games worse than their third order pythag. That might be inefficiency due to their home park and offensively skewed team, or they might have been really unlucky. The Rangers don't have a very scary rotation, but getting rid of some of the stiffs at the back end will be some significant addition by subtraction. Wilkerson and Kinsler will be better than the 4th outfielder and Soriano combo the Rangers used last season in center and at second. Young is still a defensive liability at short, but does hit well for a middle infielder. Dellucci and Matthews are underrated players when healhty.
I think they'll be in the 92-93 win range. Probably not good enough for the playoffs, but close.
Third order pythags are pretty bad for projection though. How good a team would likely finish against league-average pitching in a league average park doesn't change the fact that they play an unbalanced schedule against 3 above average pitching teams and in a far from league average park. 2nd and 3rd order pythags are more for speculation and not nearly as accurate (historically) as the base formula or Pythagenport.
Even by the base forumla's the Rangers underperformed by 2 games (it has them at .500). The wild card team (would have been the A's) is at 92 wins. So they are looking at about 11 games behind. I've got them down at right around 85 wins, enough to finish third again but still aways back in the Wild Card chase.
Tavish wrote:Third order pythags are pretty bad for projection though. How good a team would likely finish against league-average pitching in a league average park doesn't change the fact that they play an unbalanced schedule against 3 above average pitching teams and in a far from league average park. 2nd and 3rd order pythags are more for speculation and not nearly as accurate (historically) as the base formula or Pythagenport.
I don't quite understand why one would have less faith in 2nd order Pythagenports as opposed to 1st to provide a "true" win-loss; if attempting to peel away the luck involved in producing wins from runs seems worthwhile and possible, why not go one step further and peel away the luck involved in producing the runs themselves? (I agree that 3rd order, though perhaps best at assessing the "true" quality of a team in isolation, is less effective in giving an accurate win-loss because schedules are not in fact balanced.) Are you referring to a particular study?
Wozzyck wrote:I don't quite understand why one would have less faith in 2nd order Pythagenports as opposed to 1st to provide a "true" win-loss; if attempting to peel away the luck involved in producing wins from runs seems worthwhile and possible, why not go one step further and peel away the luck involved in producing the runs themselves?
Because 2nd order Pythags are not as historically accurate as base pythag or better yet pythagenport. In all honesty it really doesn't matter to me what a player's true talent level in a neutral park, against league average pitching unless I'm projecting players who are changing parks/leagues.
Teams don't play inside that bubble, the Rangers' offense may have been better last if they didn't have to face the A's, M's, or Angel's staff a disproportionate number of times and they may have won more games. But they did face those staffs more often and will do the same next season. The pitching may have allowed less runs if they didn't have to pitch in the AL Coors, but they will pitch there next season so it doesn't make much sense to take that out of the equation.
Wozzyck wrote:I don't quite understand why one would have less faith in 2nd order Pythagenports as opposed to 1st to provide a "true" win-loss; if attempting to peel away the luck involved in producing wins from runs seems worthwhile and possible, why not go one step further and peel away the luck involved in producing the runs themselves?
Because 2nd order Pythags are not as historically accurate as base pythag or better yet pythagenport. In all honesty it really doesn't matter to me what a player's true talent level in a neutral park, against league average pitching unless I'm projecting players who are changing parks/leagues.
Teams don't play inside that bubble, the Rangers' offense may have been better last if they didn't have to face the A's, M's, or Angel's staff a disproportionate number of times and they may have won more games. But they did face those staffs more often and will do the same next season. The pitching may have allowed less runs if they didn't have to pitch in the AL Coors, but they will pitch there next season so it doesn't make much sense to take that out of the equation.
Historically accurate, in what sense? As compared to actual won-loss records? Do you have a reference for this? And I agreed that 3rd order Pythagenports have dubious value in providing "true" won-loss records. I was curious what your qualms with the 2nd orders were (perhaps on principle), which don't neutralize for parks or opponents.
Cafe Home
Fantasy Football
Fantasy Basketball
Fantasy Hockey
by looptid » Thu Feb 02, 2006 4:24 pm 
