Damn your misconceptions!

[hobbsyoyo]

But, like, he's sooo clutch!

 

[sophie]

"Has performed better than average in the past", sure. But usually those statements are made as a predictor of future success, which, if I understand Owen's posts correctly, are not statements that can be supported by the data.

Yeah. Mostly this. This is partially because a lot of it is contextual and based on the eye-test/hindsight/confirmation bias, as I said before, and a lot more of it is because people for the most part have a very poor understanding of win expectancy and which situations in baseball are actually important. What constitutes a "clutch situation"? Going by raw win expectancy numbers, a player hitting a leadoff homerun to start the game increases that team's win expectancy by 15%, making it comparable with knocking in a run late in the game. But few to nobody would actually consider leading off a "clutch" situation.

You are right Warpus, it is possible to quantify that stuff, and some (notably the writers over at Baseball Prospectus [where Nate Silver got his start]) have tried to quantify it. Their findings was that there was little to no actual correlation between a player's "clutchiness" year to year. There were a couple (like, 3 or 4 over the last 40 years) players who showed a positive correlation between their clutch factor year to year, but that correlation's translation to actual win production for his team was very minute (I think something like 3 or 4 wins over a 20 year period), and it took so long for that positive correlation to stabilize (15-20 years; essentially the player's entire career) that it's not really useful at all as far as a predictive stat, and the contribution is so remote that it's not really worth remarking on.

 

[NetGear]

Yes, the point is none of this makes actual statistical sense, because walking a batter always increases run expectancy. There is literally no statistical reason to ever walk a player intentionally unless the player is so disproportionately better than everybody else on his team that the following player's batting average is like, 60-80 points lower than the guy you just walked.

but that's obvious. you're adding another man to the plate so obviously the run expectancy is gonna go up. teams don't care about what the scores are at the end of the game. They care if they won or not.

And 60-80 point range isn't all that rare? Even without pitcher batting, you have plenty of players batting within range of low 200s to low 300s. Add to that who's in a slump, who's hot, nursing injury, handedness and general pitcher preference for opponents they historically have easy/hard time and it makes alot of sense to walk the guy than you make it out to be imo.

 

[nc-1701]

While we're on the statistics bent.

People who over-fit data and/or don't realize how much over-fitting happens. In fact a great many models (especially connecting politics and economics) exist by massively over-fitting data and are thus totally useless as models. But people believe them because when applied to historical events they perform (near) perfectly.

 

[azzaman333]

Yeah. Mostly this. This is partially because a lot of it is contextual and based on the eye-test/hindsight/confirmation bias, as I said before, and a lot more of it is because people for the most part have a very poor understanding of win expectancy and which situations in baseball are actually important. What constitutes a "clutch situation"? Going by raw win expectancy numbers, a player hitting a leadoff homerun to start the game increases that team's win expectancy by 15%, making it comparable with knocking in a run late in the game. But few to nobody would actually consider leading off a "clutch" situation.

You are right Warpus, it is possible to quantify that stuff, and some (notably the writers over at Baseball Prospectus [where Nate Silver got his start]) have tried to quantify it. Their findings was that there was little to no actual correlation between a player's "clutchiness" year to year. There were a couple (like, 3 or 4 over the last 40 years) players who showed a positive correlation between their clutch factor year to year, but that correlation's translation to actual win production for his team was very minute (I think something like 3 or 4 wins over a 20 year period), and it took so long for that positive correlation to stabilize (15-20 years; essentially the player's entire career) that it's not really useful at all as far as a predictive stat, and the contribution is so remote that it's not really worth remarking on.

I know that there's been papers written recently about choking in tennis, which I'm pretty sure showed that there are certain players who are more likely to choke, and conversely the players who are better at being clutch. Not exactly baseball I know, but I wouldn't completely discount the thought of certain players under/overperforming in high pressure situations.

Of course, commentators and so-called experts typically have no idea about any of it and love to exaggerate these factors, so I definitely agree that the way people use those terms is irritating.

While we're on the statistics bent.

People who over-fit data and/or don't realize how much over-fitting happens. In fact a great many models (especially connecting politics and economics) exist by massively over-fitting data and are thus totally useless as models. But people believe them because when applied to historical events they perform (near) perfectly.

The dangerous ones are those who know that putting high order polynomials in is bad, but then goes and puts in way too many independent variables, half of which have correlations with each other.

 

[illram]

Yes, the point is none of this makes actual statistical sense, because walking a batter always increases run expectancy. There is literally no statistical reason to ever walk a player intentionally unless the player is so disproportionately better than everybody else on his team that the following player's batting average is like, 60-80 points lower than the guy you just walked.

No no no no. Run expectancy is a context neutral thing. An IBB is a context dependent action where the goal is run prevention. The comment over at fangraphs explains this better than I can:

Expected Runs is only useful in a context-neutral situation (IE, in theory), or very early in the game.

If you’re rie in the 9th, with 2 outs, men on 2nd and 3rd, it doesn’t matter if walking the #5 guy to get to the #6 increases the expected runs. All that matters is what it does to the chance to get the next out. There’s functionally no difference between a walk and a grand slam in that situation.

Walking a .325/.375/.300 guy to get to a .250/.300/.600 guy can be a completely reasonable decision despite expected runs going way up.

The article basically comports with conventional baseball wisdom: the IBB is useful in certain contexts to prevent runs.

Yeah. Mostly this. This is partially because a lot of it is contextual and based on the eye-test/hindsight/confirmation bias, as I said before, and a lot more of it is because people for the most part have a very poor understanding of win expectancy and which situations in baseball are actually important. What constitutes a "clutch situation"? Going by raw win expectancy numbers, a player hitting a leadoff homerun to start the game increases that team's win expectancy by 15%, making it comparable with knocking in a run late in the game. But few to nobody would actually consider leading off a "clutch" situation.

You are right Warpus, it is possible to quantify that stuff, and some (notably the writers over at Baseball Prospectus [where Nate Silver got his start]) have tried to quantify it. Their findings was that there was little to no actual correlation between a player's "clutchiness" year to year. There were a couple (like, 3 or 4 over the last 40 years) players who showed a positive correlation between their clutch factor year to year, but that correlation's translation to actual win production for his team was very minute (I think something like 3 or 4 wins over a 20 year period), and it took so long for that positive correlation to stabilize (15-20 years; essentially the player's entire career) that it's not really useful at all as far as a predictive stat, and the contribution is so remote that it's not really worth remarking on.

OK. But just because it is not possible to statistically gauge this, or it is not useful for predicting someone's output, does not mean it does not exist.

 

[Rashiminos]

1) Think of the evolutionary costs! Our brain is a massive calorie hog. It uses, if I recall correctly, something like a third of the average person's caloric intake just to maintain. Why would we develop something at such great cost if 85% of it is never used or redundant?

a) Things that are not actively used do not require as much energy. See Exercise. b)We don't all use everything that we acquire (at once), let alone the things that were given to us.

2) Why do people with traumatic brain injuries often suffer from some sort of mental handicap? Surely the other '85%' can pick up the slack. Right!?

Sometimes it does, but we're not usually able to get broken things to work.

3) You should doubt everything you see in a fictional movie just as a matter of course.

Yes, but doubt isn't an endpoint.

4) Name one case where someone actually tapped into the 'hidden 85%' and developed super powers. Just one please.

Reiki.

5) If this were true, why is it that in this age of boner pills don't we have researchers racing to find some way to tap into the '85%'? Sure, there are pills that some research suggests can improve cognitive function in ordinary people but as far as I know you never hear of genuine attempts to tap into the useless chunks of gray matter.

Take your pick: Ignorance, lack of interest, deliberate obfuscation, etc.

6) Why is Alzheimer's a thing? See 2).

Poor diet... etc etc ... broken etc etc...

7) Even if we had an untapped well-spring of mental ability, why is it that this also means that we have innate, physics-defying abilities like telekinesis if we could only use our whole brain? Is the other '85%' made out of unobtanium or something?

Probably not. We might need to review whether physics is actually be defied as opposed to our notions about physics.

8) Why would God bless his most perfect creation with so much dead weight in untapped potential? We're made in his image, right? So does that mean God could only violate light-speed if he were using his whole head? Could he show us the secret of the hyper-cube if only he could tap into the fullness of that sweet baby Jesus noggin of his?

Putting aside notions of God for now, having potential and actualizing potential are separate things.

9) I'm sorry, but no matter how hard you try you're never going to tap into power I don't have. Sorry to burst your bubble, but even the smartest people aren't 85% smarter than the rest of us and if you believe this falsehood, you aren't one of the smart ones to begin with.

Is it your experience that all humans are exactly the same physiologically? If not, what supports your assertion that no human could be 85% "smarter" than the norm (let alone 1000% maybe)?

10)

If he can't lift a brick with his mind then no one can.

Matt Damon assumes a crouched position. Matt Damon puts a brick on top of his head. Matt Damon stands up. Matt Damon has lifted a brick with his mind.

 

[sophie]

No no no no. Run expectancy is a context neutral thing. An IBB is a context dependent action where the goal is run prevention. The comment over at fangraphs explains this better than I can:

Presumably it comes down to who you believe then, because James Click of BaseballProspectus says basically the exact opposite:

First, let's look at a simple example. Rangers first baseman Mark Teixeira leads off an inning, giving the Rangers a run expectation of .5379 runs if they were a league-average offense. However, when we input Teixeira's averages, the run expectancy increases to .5556, meaning if the Rangers followed Teixeira with an average lineup, they could expect to score .5556 runs in the inning. If, instead, the opposing team chooses to walk Teixeira and the average lineup follows, the run expectancy icnreases to .9259. But if the next batter makes an out after Teixeira reaches first, the run expectancy falls back to .5496, lower than the .5556 with Teixeira leading off. Thus, if the man behind Teixeira is guaranteed to make an out, the opposing team reduces the expected runs the Rangers would score by walking Teixeira - the Rangers failed to protect him. Of course, no major league player is an automatic out, but we can use this method to find situations in which protection fails. IOn this case, note that the difference between the two batters would have to be very close to Teixeira's actual line on the season to force the opposition to pitch to him, because the run expectancy with him leading off the inning (.5556) is almost exactly the same as when he's on first and the next batter makes an out (.5496). Even if the batter behind Teixeira is hitting just .100, the opposition is still better off pitching to Teixeira.

[...]

These situations (where it's worth it to walk a batter intentionally) are few and far between. With men on second and third and two out, the second batter has to be on the order of 60 points worse in AVG, 95 points worse in OBP, and over 175 points worse in SLG to make it advantageous for the opposing team to walk the initial batter. Except for players hitting in front of the pitcher, such disparities are seldom found between two batters in consecutive spots in the batting order. It can be very difficult for a team to find someone to protect a player like Bonds. In 2004, standouts such as Tejada, Chipper Jones, and Ichiro would have fallen short of the performance needed to justify pitching to Bonds in this situation.

Out of all twenty-four baserunner-out situations, only half have thresholds that appeared between any two regular batters in 2004. For these twelve, for an intentional walk to have a positive run expectancy for the pitcher, the differences between the performance of the two batters can be found in Table 1-3.4. For example, with a runner on third and two outs, the pitching team would decrease its opponents' run expectation by walking any batter who was more than .028/.051/.085 better than the following batter. This is by far the most common situation in which an intentional walk is a viable strategy. Other situations, such as first and third with one out, have such an enormous threshold that they occur only when one of the league's worst batters is hitting behind Barry Bonds.

[...]

Additionally it's very rare that a situation arises in which run expectation drops after the pitching team walks the batter at the plate. Therefore, if the pitching team does walk a batter because it would rather pitch to the following man, it is almost always a mistake by opening the door for a big inning. The situation changes late in close games as the importance of a single run begins to trump that of many runs, but even in those situations, the difference between the two batters would have to be extreme.

The boys over at bp seem to have no problem using run expectancy situationally. I mean they have it for every out and base combination, I don't see why you would use it in a general sense if that's the case. I don't have Tango's The Book so I honestly have no idea what he has to say on the matter, but the BP guys really know their stuff.

*EDIT* I just jumped over to tangotiger.com and judging from the preview of his book, he also uses run expectancy to evaluate intentional walks.

 

[Zelig]

While we're on the statistics bent.

People who over-fit data and/or don't realize how much over-fitting happens. In fact a great many models (especially connecting politics and economics) exist by massively over-fitting data and are thus totally useless as models. But people believe them because when applied to historical events they perform (near) perfectly.

Yeah, buddy of mine did his entire master's thesis on how genetic models are being over-fitted and you can get equal predictive value by throwing out most of the variables.

OK. But just because it is not possible to statistically gauge this, or it is not useful for predicting someone's output, does not mean it does not exist.

Well, if it's not possible to statistically guage it, you're not going to be able to observe with a layman's opinion of "this player is a clutch".

 

[Mise]

Well you might observe it but there would be no way of knowing whether you were right or not. No scientifically rigorous way of knowing anyway. I mean if your theory of justification is sufficiently permissive then I'm sure you can say that you know something even without firm statistical evidence. I know that my mum loves me, for instance.

 

[hobbsyoyo]

Scientific theories 'are just theories' is another misconception in USA#1!

 

[Kaitzilla]

People think their Gmail emails and other cloud documents are private.

http://www.consumerwatchdog.org/new...y-when-sending-messages-gmail-people-who-care

Google’s brief said: “Just as a sender of a letter to a business colleague cannot be surprised that the recipient’s assistant opens the letter, people who use web-based email today cannot be surprised if their emails are processed by the recipient’s [email provider] in the course of delivery. Indeed, ‘a person has no legitimate expectation of privacy in information he voluntarily turns over to third parties.’” (Motion to dismiss, Page 19)

Straight from the horse's mouth!

If you are doing sensitive business stuff on shared spreadsheets or through email, for heavens sake stop!

Google made the statement that people can’t expect privacy when sending a message to a Gmail address in a response to a class action complaint filed in multi-district litigation. The suit says Google violates federal and state wiretap laws when the company reads emails to determine what ads to serve based on the message’s content. The class action complaint was filed under seal because it details many of Google’s business practices about the way it handles email.

 

[Zelig]

Well, a blanket ban doing sensitive business stuff through either shared spreadsheets or email isn't really reasonable - what you really need is someone in your IT department who properly understands the available options and technologies and their ramifications and can advise appropriately.

Your second quote isn't really relevant to Google Apps for business anyway, they don't serve ads to business customers.

 

[Kaitzilla]

Well, a blanket ban doing sensitive business stuff through either shared spreadsheets or email isn't really reasonable - what you really need is someone in your IT department who properly understands the available options and technologies and their ramifications and can advise appropriately.

Your second quote isn't really relevant to Google Apps for business anyway, they don't serve ads to business customers.

Blanket ban sounds good to me.

A company that has no morals and evidently is breaking wiretapping laws?

Won't screw with business app data?



With enough profit on the table, I have no doubt they will take that next step.

Encrypt your email at least. We know they are reading that. And sharing it with the NSA.



From their lawyers once again:

Indeed, ‘a person has no legitimate expectation of privacy in information he voluntarily turns over to third parties.’”

But, but Kai!
Why would they kill the goose that lays the golden egg by doing this?
Why do things that would lose people's trust and cost them billions?

http://www.foxbusiness.com/government/2013/08/09/nsa-leaks-slam-cloud-computing-industry/

U.S. technology companies warn they could lose between $21.5 billion to $35 billion in global cloud computing contracts over the next three years due to negative fallout from the U.S. National Security Agency (NSA) spying programs on Internet users, including emails.

Too late.

"Don't be evil" is the informal corporate motto (or slogan) of Google.[1] It was first suggested either by Google employee Paul Buchheit[2] at a meeting about corporate values in early 2000,[3] or according to another account by Google engineer Amit Patel in 1999.[4] Buchheit, the creator of Gmail, said he "wanted something that, once you put it in there, would be hard to take out", adding that the slogan was "also a bit of a jab at a lot of the other companies, especially our competitors, who at the time, in our opinion, were kind of exploiting the users to some extent."[2] While the official corporate philosophy of Google[5] does not contain the words "Don't be evil", they were included in the prospectus (aka "S-1") of Google's 2004 IPO (a letter from Google's founders, later called the "'Don't Be Evil' manifesto"): "Don’t be evil. We believe strongly that in the long term, we will be better served — as shareholders and in all other ways — by a company that does good things for the world even if we forgo some short term gains."[6] The motto is sometimes incorrectly stated as Do no evil.[3][7]

While many companies have ethical codes that govern their conduct in contract, Google claims to have made "Don't Be Evil" a central pillar of their identity[8] as part of their self-proclaimed core values.[9] The words: "Don't be evil" form part of the sixth point in these Core Values, and in full states: "Do the right thing: don't be evil. Honesty and Integrity in all we do. Our business practices are beyond reproach. We make money by doing good things."[5][8]

Don't Be Evil
2004-2013
RIP

 

[Zelig]

I'm not really sure what your point is or how it related to anything in my post other than repeating your previous post.

 

[AdamCrock]

Actually we only use 15 % of the brain because:

1) In case of men the other 85% is constantly preocupied with "trying to understand women"
2) In case of women 85% is constantly preocupied with "controling men"

 

[Dumanios]

A few: Lindbergh was first to fly across the Atlantic ocean. (John Alcock and Arthur Brown were the first)

Confederate Lost Cause as a whole.

Video Games cause violence (I know, no one on CFC actually believes this.)

Theory = Guess. (Guess: "to form an opinion of from little or no evidence." vs. Theory: "the analysis of a set of facts in their relation to one another.")

And Oxygen = Air. (Oxygen is about 20 percent of Air. Too much Oxygen leads to Oxygen poisoning.)

Also, here's a Cracked piece with more misconceptions.

 

[Rashiminos]

Actually we only use 15 % of the brain because:

1) In case of men the other 85% is constantly preocupied with "trying to understand women"
2) In case of women 85% is constantly preocupied with "controling men"

Another one: Sexism is funny.

 

[Antilogic]

Another one: Sexism is funny.

That's not a misconception.

 

[Eukaryote]

Another one: Sexism is funny.

I do believe sexist jokes originated for satiric purposes. That is, I think people thought sexism was so stupid, that it should be made fun of.

Of course nowadays, they appear to have degraded into a form of cyberbullying.