Comments on: Schilling, Byrd are off the market. So what now for Phils? http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/ In-your-face baseball commentary Sat, 21 Nov 2009 16:48:13 -0600 http://wordpress.org/?v=2.8.6 hourly 1 By: Paul Moro http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53823 Paul Moro Fri, 09 Nov 2007 20:27:30 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53823 One last interjection - I think. I think it's important to remember that a lot of the studies that the good statisticians are doing have never been done before. There are bound to be mistakes. You start with a hypothesis, and you do your best to see if it holds true. Sometimes it works, sometimes it doesn't. Sometimes it looks like it works and then someone or some event goes and disproves it. Then you go back and see where you went wrong. Again, I don't think of the things that SABR guys do as fact. Ultimately, they're all theories. Some make sense to me, others don't. One last interjection – I think.

I think it’s important to remember that a lot of the studies that the good statisticians are doing have never been done before. There are bound to be mistakes. You start with a hypothesis, and you do your best to see if it holds true. Sometimes it works, sometimes it doesn’t. Sometimes it looks like it works and then someone or some event goes and disproves it. Then you go back and see where you went wrong. Again, I don’t think of the things that SABR guys do as fact. Ultimately, they’re all theories. Some make sense to me, others don’t.

REPORT COMMENT

]]> By: Sarah Green http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53822 Sarah Green Fri, 09 Nov 2007 20:11:07 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53822 Nick, it's not accurate to describe that quote as old. It was from mid-September of this year! And while I'm always interested to see things like run differentials and projected records, those things are only compelling when you compare them to what actually happens and when you try to explain *why* a certain team underperformed or overperformed. As anyone who suffered through Introduction to Literary Theory with me knows, I am a bit of a literalist. Nick, it’s not accurate to describe that quote as old. It was from mid-September of this year! And while I’m always interested to see things like run differentials and projected records, those things are only compelling when you compare them to what actually happens and when you try to explain *why* a certain team underperformed or overperformed. As anyone who suffered through Introduction to Literary Theory with me knows, I am a bit of a literalist.

REPORT COMMENT

]]> By: Nick Kapur http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53821 Nick Kapur Fri, 09 Nov 2007 18:17:48 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53821 Those are some great points in your last comment, Sarah, of which I agree 100 percent with all. Just to be fair to the BP guys however, you are taking a rather old quote from them. One of the BP guys just wrote an excellent column about how Coors Field has consistently provided the Rockies with the greatest home-field advantage in all of baseball. So Nate Silver might have changed his tune by now (like the rest of us he also probably never dreamed the Rockies would make the playoffs anytime soon). Here's that article: http://baseballprospectus.com/article.php?articleid=6898 Those are some great points in your last comment, Sarah, of which I agree 100 percent with all. Just to be fair to the BP guys however, you are taking a rather old quote from them. One of the BP guys just wrote an excellent column about how Coors Field has consistently provided the Rockies with the greatest home-field advantage in all of baseball. So Nate Silver might have changed his tune by now (like the rest of us he also probably never dreamed the Rockies would make the playoffs anytime soon).

Here’s that article:

http://baseballprospectus.com/article.php?articleid=6898

REPORT COMMENT

]]> By: Paul Moro http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53820 Paul Moro Fri, 09 Nov 2007 17:05:09 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53820 Absolutely. Which is also why even statisticians love to watch the game. Look, there's no way to predict exactly what's going to happen at any given moment. No one ever claims to be able to do this. But here are certain things that happened in 2007 that stats predicted: 1. The terrible White Sox season. All the key players overplayed their projected numbers in 2006. Therefore, it was expected they revert to their career norms, which will lead to much fewer wins. 2. The Cleveland Indians much improved record. They had a positive run differential ( 88) last year and inexplicably finished 6 games under .500. As long as each player performs at the projected level, they will be playoff contenders in 2007 without changing a thing. 3. The Mariners falling out of the race in the AL West. Their run differentials are far inferior to their actual record. 4. The Rockies resurgence. OK, so no one in their right mind would have guessed in early September that they'd be in the playoffs. But again, their on-field performance was far better than their actual record. So while it was a total surprise that they won so many games at the end, it wasn't a surprise that they improved their winning percentage considerably. On the flip side, no one can seem to pin point how the Diamondbacks won the NL West, aside from the fact that the entire division was crappy. That's a situation where a lot of stasticians picked the D-Backs to win but no one guessed correctly how they'd do it. So stats can fairly accurately predict how things can play out over 162 games. It's far more accurate than anything I can predict in my head without stats. But stats cannot predict what happens in any given game. It can't predict what happens over seven games. There are far too many variables. Again, no one using stats should even try to say that it can be done consistently. So if you ever see something like it, don't even bother reading it. It's pointless. Absolutely. Which is also why even statisticians love to watch the game. Look, there’s no way to predict exactly what’s going to happen at any given moment. No one ever claims to be able to do this. But here are certain things that happened in 2007 that stats predicted:

1. The terrible White Sox season. All the key players overplayed their projected numbers in 2006. Therefore, it was expected they revert to their career norms, which will lead to much fewer wins.

2. The Cleveland Indians much improved record. They had a positive run differential ( 88) last year and inexplicably finished 6 games under .500. As long as each player performs at the projected level, they will be playoff contenders in 2007 without changing a thing.

3. The Mariners falling out of the race in the AL West. Their run differentials are far inferior to their actual record.

4. The Rockies resurgence. OK, so no one in their right mind would have guessed in early September that they’d be in the playoffs. But again, their on-field performance was far better than their actual record. So while it was a total surprise that they won so many games at the end, it wasn’t a surprise that they improved their winning percentage considerably.

On the flip side, no one can seem to pin point how the Diamondbacks won the NL West, aside from the fact that the entire division was crappy. That’s a situation where a lot of stasticians picked the D-Backs to win but no one guessed correctly how they’d do it.

So stats can fairly accurately predict how things can play out over 162 games. It’s far more accurate than anything I can predict in my head without stats. But stats cannot predict what happens in any given game. It can’t predict what happens over seven games. There are far too many variables. Again, no one using stats should even try to say that it can be done consistently. So if you ever see something like it, don’t even bother reading it. It’s pointless.

REPORT COMMENT

]]> By: Jojo Fireball http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53819 Jojo Fireball Fri, 09 Nov 2007 16:43:48 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53819 That's why they play the games folks... That’s why they play the games folks…

REPORT COMMENT

]]> By: Sarah Green http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53818 Sarah Green Fri, 09 Nov 2007 16:29:49 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53818 Well, yes, and it's also annoying to generalize. We cannot do without *some* generalizations, of course, but rather than dismissing home field advantage in the playoffs, as Silver does above, you'd have to look at the specific park factors and try to draw some conclusions about them. High altitude, it's fair to say, is a problem for visiting pitchers in Coors Field, just as thin air is a problem for their breaking pitches. The dimensions of Fenway's left field wall allows a player like Mike Lowell to hit a lot of doubles and a player like Manny Ramirez to pass as a decent defensive outfielder. And, to loop back to my previous comments about intangibles, I think it's reasonable to say that a visiting rookie pitching in Yankee Stadium for the first time might have some serious nerves to shake off (more than, say, if he were pitching in the Trop) due to unmeasurable-but-potentially-impactful factors like scary New York fans and the weight of history bearing down. Then there's factors like prevailing winds (always an issue with construction in and around the park) or humidity (always an issue with a knuckleballer) or midges (always an issue when global warming causes a freak insect-hatching in Cleveland in October). So I get annoyed when people say that over time, whatsyfutsy only has a 1 or 2 percent impact on the game, when on any given night, it could be a decisive factor. After all, isn't the "any given night" unknown the reason we watch? Well, yes, and it’s also annoying to generalize. We cannot do without *some* generalizations, of course, but rather than dismissing home field advantage in the playoffs, as Silver does above, you’d have to look at the specific park factors and try to draw some conclusions about them. High altitude, it’s fair to say, is a problem for visiting pitchers in Coors Field, just as thin air is a problem for their breaking pitches. The dimensions of Fenway’s left field wall allows a player like Mike Lowell to hit a lot of doubles and a player like Manny Ramirez to pass as a decent defensive outfielder. And, to loop back to my previous comments about intangibles, I think it’s reasonable to say that a visiting rookie pitching in Yankee Stadium for the first time might have some serious nerves to shake off (more than, say, if he were pitching in the Trop) due to unmeasurable-but-potentially-impactful factors like scary New York fans and the weight of history bearing down. Then there’s factors like prevailing winds (always an issue with construction in and around the park) or humidity (always an issue with a knuckleballer) or midges (always an issue when global warming causes a freak insect-hatching in Cleveland in October). So I get annoyed when people say that over time, whatsyfutsy only has a 1 or 2 percent impact on the game, when on any given night, it could be a decisive factor. After all, isn’t the “any given night” unknown the reason we watch?

REPORT COMMENT

]]> By: Paul Moro http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53817 Paul Moro Fri, 09 Nov 2007 15:58:50 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53817 Sarah, I think you're pointing out what's frustrating sometimes about the way statisticians are perceived. There are some guys out there who want you to think that they're stats oriented but in fact have no idea what the hell they're doing. Far too often, they look at a data set that's so much smaller than they should and claim that their findings prove something. It doesn't. All it shows is that they don't understand the importance of sample sizes. And this is the problem with things like statistically demonstrating the home-field advantage in a specific playoff series. It's not possible. Especially for a team like the Rockies. They've made the playoffs twice in their franchise history. You're not going to be able to find anything of significance by looking at these numbers. The sample size is far too small. Which is why statisticians sometimes take a broader view by incorporating situations that may seem like they shouldn't apply. But in order to find anything of value, sometimes, it has to be done. And that's basically the big difference between a good statistical study and a poor one. The good ones understand things like sample size and gather suffieicient information without compromising the initial intent of the study by bringing in completely irrelevant data to beef up the amount of information from which to work. The bad ones claim they've found something when in fact they haven't proven jack. They look at things like 50 ABs and try to find trends. It's a pointless exercise. Sarah, I think you’re pointing out what’s frustrating sometimes about the way statisticians are perceived. There are some guys out there who want you to think that they’re stats oriented but in fact have no idea what the hell they’re doing. Far too often, they look at a data set that’s so much smaller than they should and claim that their findings prove something. It doesn’t. All it shows is that they don’t understand the importance of sample sizes.

And this is the problem with things like statistically demonstrating the home-field advantage in a specific playoff series. It’s not possible. Especially for a team like the Rockies. They’ve made the playoffs twice in their franchise history. You’re not going to be able to find anything of significance by looking at these numbers. The sample size is far too small. Which is why statisticians sometimes take a broader view by incorporating situations that may seem like they shouldn’t apply. But in order to find anything of value, sometimes, it has to be done. And that’s basically the big difference between a good statistical study and a poor one. The good ones understand things like sample size and gather suffieicient information without compromising the initial intent of the study by bringing in completely irrelevant data to beef up the amount of information from which to work. The bad ones claim they’ve found something when in fact they haven’t proven jack. They look at things like 50 ABs and try to find trends. It’s a pointless exercise.

REPORT COMMENT

]]> By: Sarah Green http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53816 Sarah Green Fri, 09 Nov 2007 13:15:32 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53816 The issue here is the trend versus the individual case. In general, those numbers are accurate. But in the individual case, it depends. Think of this World Series: can you think of two ballparks with more homefield advantage for their teams than Coors Field and Fenway Park? Part of the thing that annoys me about statisticians is that they are always (by their very nature) looking for trends, but then sometimes (as above) apply those trend findings to specific cases where it may not actually apply quite the same way. The issue here is the trend versus the individual case. In general, those numbers are accurate. But in the individual case, it depends. Think of this World Series: can you think of two ballparks with more homefield advantage for their teams than Coors Field and Fenway Park? Part of the thing that annoys me about statisticians is that they are always (by their very nature) looking for trends, but then sometimes (as above) apply those trend findings to specific cases where it may not actually apply quite the same way.

REPORT COMMENT

]]> By: Nick Kapur http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53787 Nick Kapur Fri, 09 Nov 2007 09:39:34 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53787 Oh, I totally thought you meant home-field advantage in the regular season, over 162 games. Nobody has refuted that. But as for home-field advantage in the playoffs, I dunno. But if the numbers say it's that small, are you saying that those numbers are wrong? Oh, I totally thought you meant home-field advantage in the regular season, over 162 games. Nobody has refuted that.

But as for home-field advantage in the playoffs, I dunno. But if the numbers say it’s that small, are you saying that those numbers are wrong?

REPORT COMMENT

]]> By: Sarah Green http://umpbump.com/press/2007/11/06/schillings-off-the-market-so-what-now-for-phils/comment-page-1/#comment-53815 Sarah Green Thu, 08 Nov 2007 14:54:56 +0000 http://umpbump.com/press/schillings-off-the-market-so-what-now-for-phils/#comment-53815 Nick, the guys on BP talk about the negligibility of home field "advantage" literally all the time. It's in that book I borrowed from you. Here's a quote from Nate Silver during this year's pennant race: "To be honest, I’ve been pretty disengaged about the fight for home field advantage in the AL East. According to our research in Baseball Between the Numbers, the home team wins a whopping 51.5% of the time in a 5-game series, and 51.3% of the time in a 7-game series, assuming that it is of equal quality to the road team; that’s what the Red Sox and Yankees are fighting for." Nick, the guys on BP talk about the negligibility of home field “advantage” literally all the time. It’s in that book I borrowed from you. Here’s a quote from Nate Silver during this year’s pennant race:

“To be honest, I’ve been pretty disengaged about the fight for home field advantage in the AL East. According to our research in Baseball Between the Numbers, the home team wins a whopping 51.5% of the time in a 5-game series, and 51.3% of the time in a 7-game series, assuming that it is of equal quality to the road team; that’s what the Red Sox and Yankees are fighting for.”

REPORT COMMENT

]]>