BP Sox projection.

[voodoochile]

Originally Posted by kittle42

Weather isn't played on paper, and there is no use to predicting it until the days are finished.

This is actually a fairly good analogy because like weather there are a LOT of complex factors that go into predicting baseball outcomes.

Now like that article that "fixed" the Sox results by figuring out where the extra wins are coming from (or at least suggesting a plausible statistical fix).

Question because I honestly don't know. Since all that was discussed in that "fix" article how accurate has Pecota been on the whole at this stuff? Sox W/L was off by 5% (or so) is the rest of the league more accurate?

[amsteel]

Originally Posted by voodoochile

Question because I honestly don't know. Since all that was discussed in that "fix" article how accurate has Pecota been on the whole at this stuff? Sox W/L was off by 5% (or so) is the rest of the league more accurate?

I would also be interested in seeing how PECOTA performed in relation to RS/RA and a team's Pyth. W/L.

[dickallen15]

Originally Posted by kittle42

Based on the fact that few understand it, or really any statistical model.

I think I understand it, and as I pointed out, they were fairly inaccurate predicting the outcome of the 2012 season in February 2012. It will probably be fairly inaccurate predicting the outcome of the 2013 season in February 2013. If anyone could actually be really accurate predicting the outcome of a season a couple of months before it started, it wouldn't be available for virtually nothing. Just like it would mean nothing if the Sox were expected to run away with the division, it means nothing the formulas have them finishing in 3rd place.

Its OK for people to defend BP, I read it myself. But to try to rip people for calling a spade a spade, and that would be calling out past inaccuracies and having that referred to as "anti-intellectual" is a joke.

[dickallen15]

Originally Posted by voodoochile

This is actually a fairly good analogy because like weather there are a LOT of complex factors that go into predicting baseball outcomes.

Now like that article that "fixed" the Sox results by figuring out where the extra wins are coming from (or at least suggesting a plausible statistical fix).

Question because I honestly don't know. Since all that was discussed in that "fix" article how accurate has Pecota been on the whole at this stuff? Sox W/L was off by 5% (or so) is the rest of the league more accurate?

I think the Sox stand out because they almost always outperform the projection. The fact is, the Sox projection was off 10 games last season. There were 11 more teams that were off by at least that much. They had 2 teams right on, and I think there were 12 teams within 5 games. But some of those projections were too high instead of the Sox usual too low. I would admit people probably wouldn't be too upset with BP if they overestimated Sox wins on an almost annual basis.

[kittle42]

Originally Posted by dickallen15

I would admit people probably wouldn't be too upset with BP if they overestimated Sox wins on an almost annual basis.

Exactly, and this is what bugs me the most.

[dickallen15]

Originally Posted by kittle42

Exactly, and this is what bugs me the most.

Yes but that is part of being a fan. It's like having a little brother. You can call him an idiot, but if someone else does, it ticks you off.

[PalehosePlanet]

Here are a couple of articles about how PECOTA has had a difficult time predicting the Sox.

http://www.fangraphs.com/blogs/index...g-projections/

http://sports.yahoo.com/blogs/mlb-bi...9372--mlb.html

[asindc]

Originally Posted by dickallen15

Yes but that is part of being a fan. It's like having a little brother. You can call him an idiot, but if someone else does, it ticks you off.

Yes indeed.