Use access key #2 to skip to page content.

portefeuille (98.81)




June 09, 2009 – Comments (2)

******* warning **********************

*********** this is a DRAFT *****************

*************only proceed if you are really really bored and can take my "style" **************

******************* written "in a hurry" **************************************************

Are bravobevo and I 1% of the "caps" game valuation?

I have (via my 12 players) made 3044 calls so far. bravobevo has made 1925. That makes 4969 calls for the two of us. The "formula" for assigning the "star ranks" in the "caps" game are proprietary (see this paper (pdf)).


As part of the data for the study, Motley Fool also provided the daily rankings of these stocks from “1-Star” (the worst) to “5-Stars” (the best) for this one-year period. The CAPS website states that these ranking are based on a proprietary algorithm that gives extra weight to predictions by highly-rated players and to recent picks.4 In general, each star rating corresponds to a quintile of stocks based on an underlying (unreported) CAPS cardinal ratings for those stocks. That is, “1 Star” stocks consist of those stocks at the 20th percentile or below in cardinal ranking, whereas “5 Star” stocks consist of those stocks at the 80th percentile or above in cardinal ranking based on past CAPS picks. We had no knowledge of the specific details about the proprietary system used to generate these rankings, and made no effort to identify its properties.


(from that paper, p. 9/39)

okay, I will made some "effort to identify its properties". It has been said quite a few times (I might look up the sources, but since people tend to be annoyed by my numerous links and citations anyhow I think I will not.) that the star ranking system (SRS) is not "directly" based on the ranks of the players making the corresponding calls but that these ranks do play a major role in determining the rank of the objects of those calls. I assume that the SRS is designed to produce the best results in the following sense. The higher the rank of a stock/ETF in the SRS is, the greater its "outperformance" vs. the benchmark (the S&P 500 index) should be. The highest rank is 5 stars and the lowest rank is 1 stars. Some stocks/ETFs have no rating. "underperformance" is to be counted as negative "outperformance". Again, I really do not like this very non-exact formulation, but another thing people tend to dislike is my "overuse" of "formalism", so I will just keep writing in plain words for the rest of this post. Now I could again list some link collection (of the (in)famous 1,2,3 ...,120 kind ...) but I think really everybody by now should have some level of knowledge in the area of "why the 'caps' game 'accuracy' is stupid / can be manipulated / whatever we like to call it). bigpeach will have some interesting results on how random walk produces a "caps" game accuracy of around 90%. I think that should be the sudden death to all claims that the "caps" game "accuracy" has a lot to do with the naïve meaning of accuracy. Why am I mentioning this? I mention it because I will assume that the "caps" game staff is smart enough to acknowledge the deficiency of the "accuracy" and will not at all or just to a small degree use it as input for the SRS. (This is however immaterial to the determination of the answer to the initial question: are bravobevo and I 1% of the "caps" game valuation?). So now what might this "proprietary" formula that determines the SRS look like. It is hopefully not really a formula but a set of rules that leads to an algorithm that determines the SRS. (Now that I think about it I remember someone saying that they (the "staff") are constantly trying out new "algorithms" trying to improve the SRS.)

I think I will continue later, just posting this so people in the mean time can stop me from continuing (the "staff" could end the speculation by telling "what it is (the "proprietary formula").

Just a quick summary of what my results will be. I think you can give an average "impact factor" (IF) of 6 or so to the picks of  bravobevo and I, where 1 would be the average impact and a call with IF = n would mean that it has the same "impact" on the rank assigned to the object of that call in the SRS as n calls (on the same stock/ETF) with IF = 1.

So the calculation would be something like: 3100000 / (6 * 4969) = ca. 0.00962 -> ca. 1%.

Of course this IF = 6 needs some explanation, but if you don't like it make it 2 or 3 and add some players like chk999 and goldminingxpert (and the other 3 members of his gang).

then the "headline" would be  

"bravobevo, chk999 and the gangs of goldminingxpert and portefeuille are 1% of the "caps" game valuation!"

so that's it for now ...

2 Comments – Post Your Own

#1) On June 09, 2009 at 6:05 AM, portefeuille (98.81) wrote:


3100000? see this ...

Report this comment
#2) On June 09, 2009 at 9:56 AM, portefeuille (98.81) wrote:

The "formula" for assigning the "star ranks" in the "caps" game are proprietary ...

The "formula" for assigning the "star ranks" in the "caps" game is proprietary ...

Report this comment

Featured Broker Partners