Comments on: The richness of crowds?

The richness of crowds?

allan — Tue, 26 Jul 2005 14:51:47 -0800

Marketocracy is a free, handy site where you can practice building your own stock portfolio.
MOFQX is a moderately successful mutual fund driven entirely by the top 100 performers out of some 37,000 Marketocracy members. With market-beating returns and an innovative method, some think that the fund might be a great idea--perhaps the wisdom of crowds made manifest--but others are less bullish.

By: Nelson

Nelson — Tue, 26 Jul 2005 15:00:17 -0800

The claim "market-beating returns" comes from a comparison to the S&P 500. That's not really a reasonable comparison. For an alternate view, check out the Yahoo performance data for "rank in category" or the Morningstar 1 star rating.

By: jcruelty

jcruelty — Tue, 26 Jul 2005 16:21:29 -0800

Always good reading: A Random Walk Down Wall Street, which argues against managed funds.

By: nofundy

nofundy — Tue, 26 Jul 2005 17:00:25 -0800

I'm wondering what Nassim Taleb thinks of this. I bet not much. Currently reading Fooled By Randomness and highly recommend it. Not just for market people.

By: sfenders

sfenders — Tue, 26 Jul 2005 17:30:30 -0800

A Random Walk must surely be outdated by now. I don't think anyone can really still believe it's an actual random walk. At the least, Mandelbrot has some things to say about that.

By: sfenders

sfenders — Tue, 26 Jul 2005 17:52:51 -0800

I can't remember where it comes from, but the classroom demonstration with flipping a coin comes to mind. (Wherein you get everyone to flip a coin. Take all the people who got "heads", and have them do another coin flip. Repeat until you've found the one person with the remarkable talent of flipping "heads" n times in a row. Very impressive.) So, that's obviously a danger for Marketocracy. But I do believe it could be overcome if they're clever enough to account for it. They would have to track performance very carefully for it to work, making sure it was over a long enough period and through enough trades. I've no idea if they do it right, and it would be much easier to do it wrong, but it's an interesting idea.

By: allan

allan — Tue, 26 Jul 2005 17:55:50 -0800

I'm still with Malkiel that managed funds are generally a waste: if you only have a few holdings, an index fund wins, and they are only really useful in larger portfolios for very specific reasons. MOFQX could do well if one can argue that "past earnings does not predict future performance" applies more to the long run than the short run. The top 100 Marketocracy performers will almost by definition do better than [most metrics] each month. Inside a 1 month lag period, if performance can be partly predicted by last month, then the aggregated performance might do well simply for statistical reasons. It's a neat idea, but I'm still skeptical.

By: sfenders

sfenders — Tue, 26 Jul 2005 18:03:34 -0800

Actually, maybe a better way to think of it is that they're replacing picking stocks with an even more difficult challenge of picking "funds" for their meta-fund. Very likely, actually, since that's exactly why I don't like mutual funds in the first place: I think picking mutual funds is actually much more of a random thing than picking stocks.

By: groundhog

groundhog — Tue, 26 Jul 2005 18:44:22 -0800

The expense ratio is almost two percent. No thanks.

By: milkrate

milkrate — Tue, 26 Jul 2005 20:02:19 -0800

To get an idea of the value an active money manager adds you ought to use the correct benchmark - which should represent the space that the fund invests in. The S&P 500 isn't an appropriate benchmark, since this is a mid-to-small cap fund. Over the time period they talk about, all the mid and small-cap indices outpaced the S&P 500 - any mid or small cap fund should have beaten the S&P 500. The right benchmark for the fund would be something like the S&P 400, Russell 2000, or Russell 2500.

By: jikel_morten

jikel_morten — Tue, 26 Jul 2005 20:34:56 -0800

sfenders: Sounds like the intro to Buffett's famous value investing essay "The superinvestors of Graham and Doddsville." In the essay, Buffett describes the entire country flipping coins, as you've described.

By: H. Roark

H. Roark — Tue, 26 Jul 2005 21:20:49 -0800

Random walk is a joke. The convergence of chaos theory and finance in the last decade or so has shown that there are patterns to everything, including the stock market. The thing is, the math is beyond most folks means right now and even if it was they lack the capital and will to make the arbitrage strategies work over the long run. 10 years from now, when the strategies are more fully understood and employed, the spreads will shrink and yet another money-making strategy will pass on. Until then, just keep reading "Random Walk" and I will keep doing my thing.

By: TwelveTwo

TwelveTwo — Tue, 26 Jul 2005 23:41:53 -0800

A predictable stock market is the worst thing imaginable. Simply think of the jobs lost without the unpredictability of stock. Who would seek stock management, stock advice, stock columns, stock analysts or, stock gurus? No one! The streets would become flooded with men and women unskilled in anything based in reality. Some may make it working as political statisticians, others as pollsters, but the great many would flounder. Without guidance the skill they once used so greatly would work against them, many might even mistake the connection of the passing of Noon and their repletion as causal. Thousands would starve to death. It would be catastrophic. This is why I help keep the stock market random by selecting my stocks through the use of random number generators. So when one looks on and asks what am I doing purchasing 52 shares in a company that specializes soley in the production of novelty glasses for the blind, I can proudly say, "Doing my part!" And you can too.

By: Mutant

Mutant — Wed, 27 Jul 2005 04:08:36 -0800

Dividends folks, dividends! I only purchase shares if I'm paid to hold them. No dividend, I'm not interested. I'm getting about 11.75% on my personal portfolio at present. It's generates monthly, passive income. At the end of the month my biggest problem is how to effectively deploy the capitail into additional, income generating securities. A firm has way more control over the dividend they pay to shareholders than they do their share price.

By: nofundy

nofundy — Wed, 27 Jul 2005 05:18:01 -0800

I can't remember where it comes from, but the classroom demonstration with flipping a coin comes to mind. (Wherein you get everyone to flip a coin. Take all the people who got "heads", and have them do another coin flip. Repeat until you've found the one person with the remarkable talent of flipping "heads" n times in a row. Very impressive.) This is known as survivorship bias. Works in all kinds of situations. H. Roark, ever hear of heuristics? Often times the market behavior has absolutely nothing to do with mathematics and everything to do with human emotions. But go on doing what you're doing and never mind that black swan in your path to fortune.

By: jikel_morten

jikel_morten — Wed, 27 Jul 2005 07:43:52 -0800

This is known as as survivorship bias Cool, thanks nofundy. I haven't heard that term.

By: sfenders

sfenders — Wed, 27 Jul 2005 10:30:19 -0800

Yeah, survivorship bias. That Buffet speech is indeed where I first heard of the concept, and it conveys it in a very memorable way. That the marketocracy folks don't make a big deal out of explaining how they deal with that problem leads me to suspect that they don't. Well, that and the fact that their "masters 100" fund hasn't done at all well in the past year.

By: Charles Dainoff

Charles Dainoff — Thu, 28 Jul 2005 12:05:07 -0800

My company, 1st Principles, Inc., has developed ValueTool, stock selection software driven by fundamental analysis and powered by company financial data. In essence, ValueTool offers individual investors a shortcut: it provides a "snapshot" of a company's value compared with its stock price using seven different valuation equations commonly used by investment professionals. Instead of putting in all the time and effort the professionals do, however, all the ValueTool user has to do is type in a company name or ticker symbol and hit "Enter". The software does the rest, and provides the information in a manner that's easy to read and understand. Right now, we're giving away free ValueTool memberships, so there's no risk. It's Web-based, so there's nothing to load or store. Check it out at www.valuetool.com, and give yourself an unfair advantage today.