The post suggests more, though doesn't illustrate it all (because I couldn't think of a simple way to do so). The key is that x and y can be statistically related yet still have zero correlation (because that is just one specific measure of relatedness). Interpretations of causation are not data driven. They come from theory.

The role of confounded factors came up in the examples because that is just an easy way to illustrate the points.

That's the subject of the next post on this topic, forthcoming.

I get my blog content via RSS. All that fancy formatting and layout never reaches my eyes. That plus consideration of the average half-life of a blog post makes additional effort...well...kind of pointless. (Unless the point is to make art, which is different.)

calories burned > calories consumed

This is true in the same way that being poor is due to money spent > money earned. It reduces a complex and difficult problem to a smug soundbite.

This is obviously true, but it's not as simple as it may seem.

The body has a complex energy regulation system and hunger, energy consumption via metabolism and energy storage are interlaced. It's not as simple as "energy in" = independent variable.

It has a preference to store (instead of immediately use) calories from carbohydrates (due to insulin). So a diet low in total calories but still high in carbs is going to be painful, and very frequently fails.

Read Gary Taubes' book, people.

This is fundamentally true but I usually add this caveat: All calories aren't equal.

Having a negative energy balance is going to ultimately lead to weight loss, yes. But it's important to eat good foods (whole grains, fruits, vegetables) so you 'feel better' even though you're eating less calories.

This journal article (A Role for Sweet Taste: Calorie Predictive Relations in Energy Regulation by Rats: http://www.apa.org/journals/releases/bne-feb08-swithers.pdf ) is about a comparing natural and artificial sweeteners. The point of the article is the rats fed artificial sweeteners gained more weight than the rats fed real sugar.

It's just a little evidence in support of the original author's assertion that 'Light products are for pussies.' The truth is that while that Diet Coke may have less calories in it, it can also lead to an increase in appetite (read: weight gain.)

If you only you could expand that into a book or expensive program... darn it all.

I wish I could upvote you twice!

This is some sort of radical fad of eating less and exercising?

Perhaps they will make a pill for this?

I like speed. I like the Google Toolbar. I look forward to the day when Chrome satisfies both.

I confess to knowing very little about the limitations of URLs so perhaps this is s dumb question: why not go all the way down to the shortest possible? Can they just use the minimum unique string of characters? For example http://xyz7g. What are the limits?

You need a domain name, a period, a top level domain, a slash and then an address at that domain.

The shortest top level domains are the country ones - which are two characters long. The shortest domain name on top of that would be one character.

So http://j.mp/Whatever is in fact the shortest you can get.

Note that silly services like ours ( http://tinyarro.ws ) further compress the character length by using unicode after the / so the URLs never grow beyond 3 characters (and can do 2 characters for a very long time).

Last time I checked, even though the web front end counts characters, the Twitter back end counts bytes (as does, I believe, SMS). So unicode doesn't actually save anything (for twitter).

Conveniently, it does save characters on the website at entry time as their site only checks by character count. Also their backend via website doesn't validate by bytes either.

Their API has fluctuated from checking bytes and characters over time-- I think right now it checks bytes. SMS does it by bytes.

Their API has also done both simultaneously at various times, mogrifying tweets as they transition between queues/memcache.

SMS doesn't actually use bytes natively -- it's 160 7-bit characters packed into 140 bytes. As is their way, Twitter fucks this up: they use the 20 spare characters for the "username: " prefix, but limit usernames to 15 characters -- 3 are completely wasted! Why not allow usernames to be 18 characters?

They've historically fucked up plenty of other SMS encoding details like sending & escaped as & and murdering unicode in weird ways -- always truncating the message at an arbitrary tier instead of validating/refusing it up front.

It's technically possible to put a site on a top level domain. Now that ICANN has started allowing the purchase of TLDs we will soon be going to http://google and we'll probably see a URL shortener at http://l

Now if only we could get rid of that troublesome http://

well you have to have a valid domain name, and xyz7g is not one.

twitter could have easily implemented some kind of even shorter system by using some kind of rarely-used hash tag character like this:

Check out this link : ±xyz7g

Which would be auto-parsed to http://bit.ly/xyz7g. That would save 13 characters.

(a) It is a highly stylized example to illustrate game-theoretic concepts. In a sense I am asking the reader to imagine that THE GAME is about the runner getting to second or not holding all else about the game constant. That's not really baseball, but it is a baseball-like situation.

(b) The situation described is both vague (was the probability referenced by the color commentator for this runner? For this team? For this situation?) and specific (a specific runner/pitcher pair, there are zero outs not two).

(c) You are right about the goal of baseball. I had originally written it with the goal being scoring. But then it is a vastly more complex game.

A very good article. I'm just getting into games now, and seeing lots of small simplified examples like this is exactly what I need to improve my intuitional understanding.

I'll have more on basic game theory, including a post that will explain a relatively painless and fun way to learn the basics. Stay tuned.

@brk - That makes sense. It's what I would have done anyway.

You can use the reply link right below a comment so you don't have to refer to someone's nick and the discussion will remain threaded.

I'm honored to be cited here. I heard about Y Combinator on EconTalk. I had no idea it was a place to find links to interesting posts and news.

Anyway, this post is applicable to investing and bubbles. How far from common knowledge is assumption about rationality? If it is sufficiently far then it is rational to buy/sell at a price very different from the one that would obtain with common knowledge rationality.