rluzinski

I'd think he would time the suspension to make sure he played against the division leaders.

Brewers19 and I will be monitoring it as often as time permits. Our goal is to make at least some chat time available to more people. Ahh, but you underestimate me!

Moneyball made everyone think that sabermetrics were in some competition to eliminate scouting. Pure fiction. EDIT: spellcheck sucks http://forum.brewerfan.net/images/smilies/smile.gif

LINK

The Giants have 4 guys who are .290+, they might not string hits together to score runs, but their hitters aren't terrible. That's not a very good way to try and gauge the strength of their offense. Heck, one of those guys has 22 AB. When you have a .718 team OPS and Bonds is included, you don't have much going for you.

Well, if anyone wants to sell their used lawnmower, let me know. http://forum.brewerfan.net/images/smilies/smile.gif

Miller almost got thrown out at 2B on a shot off the wall?

That Inning

Grand friggin' slam.

I think Weeks will be completely wasted being surrounded by Mench, Miller and Vargas. I suspect Yost has decided that he has to have a consistent lineup again, however. I doubt we'll see any changes to the order anytime soon.

I don't understand why it has to simply be a case of "unluckiness". What if there is an actual non-random cause for his performance. Of course that's a possibility. As I said in a previous post, Yost and his staff are in a position to best analysis whether another factor might be playing a part. If they don't find anything, the mostly likely cause is that the sample does not reflect Bush's true talent level. Whether it's because of bad luck, a temporary breakdown in mechanics, ext..., I don't really care. Provided that the bad stats that he's accumulated have a tendency to regress over time is what's most important to me. Of course, as the sample increases, the "will regress" conclusion becomes less and less likely and the possibility of another cause becomes more likely. If you take enough small samples, you'd expect to eventually find one that looks just like Bush's. I showed in a different thread that extreme BABIP and LOB% in one year for a pitcher almost always regresses back to something more reasonable the following year. There are no guarantees, though. I would like to see Russ' evaluation of Bush out of the stretch to see what those numbers tell us. I provided his no on/runner on splits earlier in the post but as I mentioned, I'm not really sure how to interpret them or even if the difference is statistically significant. If he has less accuracy, he compensates by pitching less fine (not at the corners) , since his walk rate goes down. If he was simply more wild but didn't change his approach, you'd expect more mistake pitches AND more walks, wouldn't you? What I need is pitch location data, which isn't publically available.

Where this whole issue breaks down is the assertion that any of the sabermetric theories/hypothesis are correct. I'll be a believer when I see the proof. Proof or evidence? I'm sure you've seen plenty of evidence for many of the theories thrown around here. Whether you are personally swayed by the evidence is a matter of opinion. Obviously, some theories have more compelling evidence than others. Labeling a theory "correct" isn't really how the scientific method works, anyway, is it? Well established theories have been discarded in every field, after all. I appreciate that as a less developed/structured field, you are more skeptical of sabermetric findings, as you should be. A theory advanced in the sabermetric field couldn't have possibly been held up to the same rigorous scrutiny as some other theories in other fields have been. Still, to suggest the possibility that none of the sabermetric theories advanced are "correct" seems somewhat ignorant to me. There are some very straight forward, common sense theories that have been critically reviewed by many smart individuals and withstood the scrutiny. Heck, much of the work done is simply applying well known statistical tools from other fields of science. Knowing you come from a scientific background, I am somewhat puzzled by how you view this stuff. Science is simply the "best guess" on how our surroundings work. Most likely no theory is completely correct nor any model perfectly accurate. Is your criticism that scientists don't typically recognize the limitations of their models and/or overstate their findings? If that's the case, we are in agreement.

My Atari 2600 could not be killed either. Of course, to compare the old systems to the new consoles, which are basically high powered PCs, is unfair.

I'm sure he's bent on trying to stick it up everybody's butt on the so called fan "support" he's received the past 2 years. Yeh, he kind of chided fans in his post game interview, for jumping on and off the bandwagon and judging pitchers one two or three starts. He said to give them 30 starts and judge them afterwards. Frankly, it has to be a pain in the rear to deal with the casual fans and their fickle opinions. He was also critical of his general performance, pointing out that he can get a batter to 0-2 and 1-2 but can't seem to finish them off right now.

Don't worry. The Brewers will lose a 3 in a row somewhere in there and you can safely predict a Brewer collapse again. http://forum.brewerfan.net/images/smilies/smile.gif

Great start to the home stand. A nice 9 or 10 game lead in the NL Central would sure be nice at the all star break.

ike many of you mentioned there seems to be a lot of examples of problems on the internet but everyone I know has never had a problem with there consoles. That's because even if the defect rate is very high, it's probably still "only" like 10-15%. Your friend should have waited for Christmas, when they might have a redesigned 360 available.

The coin flip analogy is used quite a bit, but how often it comes heads and tails is determined by any number factors; It's not an analogy so much of an example I use to try and illustrate how binomial variance(coin flip stuff) adds a layer of uncertainty to all the stats we look at in baseball (which is directly dependent on the sample size). People sometimes take that to mean I think baseball can be reduced to a coin flip, which couldn't be farther from the truth. I only begin with that uncertainty, since it's easily quantified and very large, when the sample size is small. There are many other uncertainties, many of which, the manager and coaching staff are in a prime position to try and gauge. Dave Bush has had prior success in the majors and because we expect a certain level of skill persistence, the default assumption is that he should continue to have success. There are factors that can throw a monkey wrench into that assumption of course. Injuries, change in mechanics, change in mental approach and age, to name a few. But if Yost and his staff aren't aware of any of those currently being a factor, why shouldn't they assume the most obvious and likely scenario? Bush might have simply flipped 9 heads, when 5 were expected and considering the underlying data, that sure looks to be the over riding cause of his misfortunes. If the sub-par innings mount, Yost and Melvin will have to re-evaluate the situation, since it will become less and less likely that bad luck is the cause. Furthermore, if a better option presents itself that trumps the uncertainty of recent bad performance, that could justify a move as well. Since they have to consider many different factors, there is no magic number, no definitive situation.

No problems here either (since Jan of 06'). It does sound like there are many people with issues out there, however. I'm surprised you can't return your console to the store you bough it at. Are you sure what they told you is in their official return policy?

It doesn't matter what you did yesterday, last week, last month, last year, it only matters what you're going to do right now. All the manager has to go on is what his gut tells him, and what he sees in the body language of the player. A manager needs to consider everything. Prior performance should be #1 on that list but there's obviously many other factors that should go into his decisions. A good manager will use all those factors to form his "gut" feeling. This is why I call the defense from the field in football, because I can look into the young man's eyes before I put him in the game. Comparing major league baseball to high school football is a pretty big stretch, to say the least. Professional baseball is such a controlled environment, relatively speaking, it lends itself very well to statistical analysis. There are years worth of data to look at. High school football is an entirely different beast. The performance of a football player is so intertwined with his teammates, it's very difficult to separate the two. Measuring the strength of the opponent is difficult to quantify. Quantifying individual player performance is also difficult. I high school coach has no choice but to rely heavily on his and his coaching staff's subjective assessment of a player's skills. Geno's point, which was never really rebuttled, is how long does a manager ride an underperforming pony? First, how do you rebut a question? There isn't one answer anyway. I explained in painful detail the limitations to small samples and in the data in general. A manager needs to put all that data and personal observations into a big bucket and make some hard decisions sometimes.

Great. Did you total those yourself? I just took a simple (non-weighted) average from espn for each year. Any idea why your numbers for Bush don't match B-R.com? My sample didn't include this year's data.

Jacob, I showed the NL average from 04-06 with runners on and empty a few posts back in this thread. I'll check his last game on MLB.tv to see for sure. Thanks. Then we can be looking at the correct sample at least.

The piston has the notch, the cylinder has only a small scratch. I assume I could just hone it out if necessary. As for the parts, I was a click away from getting them delivered to my house, so I'm not worried about that, either. I just don't want to waste my time and money fixing something that has a small chance of success. Considering it was $180 news and the parts are almost a third of that, I'm leaning towards giving it up. Just seems like a waste, though.

To me, non stretch is bases loaded or empty. What would be the benefit of pitching from the stretch with a runner at third or runners at second and third? I guess I haven't paid enough attention to Bush (or any pitcher, really) to know when they do it.

The coin doesn't get tired after being flipped 100 times. There's no stress on the coin's elbow. The coin can't be going through a divorce, a steroids investigation, nor can the coin be worried about possibly being outed as gay (unless it's one of those flamboyant ones from South America). What you're describing is the uncertainty in the player's actual talent level at that very moment. That's another layer of uncertainty that is added to the binomial uncertainty (what I'm calling "coin flip" uncertainty. I focus on the binomial uncertainty because it's easily quantified and a huge component of the uncertainty over 200 AB or 50 IP. Anything that has a probability of success, or failure, like flipping heads, rolling snake eyes or even hitting a baseball, can't escape the uncertainty I'm describing. If it didn't apply to baseball, every .300 hitter would get exactly 1 hit in every 3 AB and every 4.5 ERA pitcher would give up 1 earned run every 2 innings. It has to apply. The combination of variance in true talent (what your post is describing) and statistical variance (what I'm describing) combine to form the observed variance in player productions. They work in tandem to create hitting slumps and no hitters. Given a large enough sample, we can diminish the binomial uncertainty component enough so that we can make a very good guess at the actual, average talent of the player. I'm sure someone on the internet has explained this much better than my lame attempt above.