Use access key #2 to skip to page content.

starbucks4ever (97.37)

The one-star revolution is coming

Recs

6

June 04, 2009 – Comments (4)

In case you think those 5-star ratings are very meaningful, think again. To be rated 5 stars, a ticker must be recommended by many top players, and to be rated 1-star, it has to be mostly popular among low-rated players. However, the difference between these categories of players is not that great. For instance, if you have a 65% accuracy and a score of -150, you get a rating of 35. But if you improve that score by only 300 points, you get a score of +150, and your rating goes up to 85. The author of these lines had his score jump from 34 to 89 in just a couple of days, which, by the way, were not marked by any extensive stock picking on my part. I did not become any smarter or stupider during these two days, and my opinions on my (currently) 178 picks are still the same. But those same bets that looked suspicios for many months because they came from a low-rated player are now looking respectable because with my new rating, I now appear to be smart. So my red and green thumbs are now assigned more weight by the Caps program, and the 1-star tickers that I "bought" in Caps are now likely to see a promotion to 5-stardom? But the fundamentals of these stocks haven't changed since last week. I have seen many players' ratings undergo a dramatic change this month, and I now expect that many tickers will gain or lose stars in the coming weeks. So if you put your trust in the "collective intelligence" of Caps players, remember that this "collective intelligence" may be less meaningful than you think.

4 Comments – Post Your Own

#1) On June 04, 2009 at 7:11 PM, portefeuille (99.59) wrote:

To be rated 5 stars, a ticker must be recommended by many top players, ...

not necessarily.

... and to be rated 1-star, it has to be mostly popular among low-rated players.

no, that is not a necessary condition.

So if you put your trust in the "collective intelligence" of Caps players, remember that this "collective intelligence" may be less meaningful than you think.

yes, I agree. but also see this ...

There is a post of mine on a different aspect of the "stars" system in the "caps" game here.

Report this comment
#2) On June 04, 2009 at 7:39 PM, UKIAHED (35.95) wrote:

portefeuille

yes, I agree. but also see this ... 

Thanks for posting the link - I missed that thread - quite interesting.

 

Report this comment
#3) On June 04, 2009 at 8:40 PM, FleaBagger (29.57) wrote:

You misunderstand the criteria needed for the respective star ratings. While not perfect, it's a lot more sophisticated than you state, and is largely successful.

Report this comment
#4) On June 04, 2009 at 11:17 PM, portefeuille (99.59) wrote:

You misunderstand the criteria needed for the respective star ratings. While not perfect, it's a lot more sophisticated than you state, and is largely successful.

... and it is (as far as I know) the only really proprietary thing about the "caps" game (see the paper (pdf) mentioned here). The authors had no detailed knowledge of the "mechanism" of the "star ratings".

Report this comment

Featured Broker Partners


Advertisement