The hare and the tortoise revisited

[farnamstreet]

Eugene Shahan sat down with Warren Buffett's article "the superinvestors of Graham and Doddsville" and did some figuring. He found out some interesting things about super-investing, short-term performance, and how investment management firms should be structured.

 

http://bit.ly/VxD7W

 

[ExpectedValue]

since Scribd is extremely annoying-

 

direct link to PDF document:

 

http://www.bestinver.es/pdf/articulos_value/The%20Superinvestors%20of%20Graham%20and%20Doddsville%20by%20Warren%20Buffett.pdf

[nodnub]

since Scribd is extremely annoying-

 

direct link to PDF document:

 

 

Agreed!  Thanks!

[bargainman]

The question I have, after reading more than half way through "Fooled By Randomness", is: Is Buffet's method for demonstrating the non-randomness really valid?  In the book Taleb talks about survivorship bias, and how many times people aren't looking at the full sample set.  In the article Buffet specifically only picked 7 *successful* managers.  He never talks about all the managers that went through Graham's class and used his methods that were unsuccessful.  If they were unsuccessful they never would have made it in the business and would have just dropped out.  So that leaves me wondering.  The first time I read Buffet's article I agreed with everything he said  But this time I'm not so sure.  According to Taleb you can't just look at the successful people and draw this conclusion.  I mean there are a *lot* of Value funds out there.  How many of them actually outperform?  is it statistically more than other index and growth funds?  Makes me wonder...  Thoughts?  Has anyone else out there read Taleb's book?

[maxprogram]

I think Buffett attempted to challenge the survivorship bias argument when he stated that "...these winners were all well known to me and pre-identified as superior investors, the most recent identification occurring over 15 years ago." Also, we know (from Buffett biographies, old friends, etc.) that those he selected for the speech weren't just the "best" out of the people involved with Graham -- they were for the most part friends of Buffett's before and during the time their track records were established.

 

Also, my guess is that most, if not all of those he listed continued to outperform the markets after 1983 (at least those who continued investing).

 

I like to think of luck/skill in investing like very large dartboard: the unskilled, or those without the right strategy, are throwing darts blind all over the board. A few might hit the bullseye every once in a while, but in the long-run most wont succeed. Yes, out of a pool of millions of these investors, you should get a few who have "flipped heads" 20 times in a row. But the people like Buffett and other Grahamian investors know the right area to throw the darts. On each individual throw, there is a degree of randomness and luck involved -- but the probability is higher that these investors will get a higher score, on average, after many throws. So, over time, they will be more successful than the rest.

 

I think that Taleb recognizes this in his books. A few quotes from Fooled by Randomness:

"All the best and surviving traders trade on ideas based on some observations (including past history), but they make sure that the costs of being wrong are limited (and their probability is not derived from past data)."

"Ergodicity means, roughly, that (under certain conditions) very long sample paths would end up resembling each other. Those who were unlucky in life in spite of their skills would eventually rise. The lucky fool might have benefited from some luck in life; over the longer run we would slowly converge to the state of a less-lucky idiot. Each one would revert to his long-term properties."

[scorpioncapital]

But wasn't Buffet's argument that the investors came from the value investing camp? I don't think his argument was one of picking 7 successful managers from all managers which does suffer from survivorship bias. To prove his point all he has to do is pick a few from a very specific "sub-set". Now one could argue how many value investors he needs to sample and I'm sure future academics since that article have done studies on value performance vs all active managers.

[nodnub]

The first time I read Buffet's article I agreed with everything he said  But this time I'm not so sure.  According to Taleb you can't just look at the successful people and draw this conclusion.  I mean there are a *lot* of Value funds out there.  How many of them actually outperform? 

 

Klarman might say that a lot of the value funds out there are "value pretenders".