Use access key #2 to skip to page content.




October 22, 2008 – Comments (4)

There's a lot of talk about alpha.

Frankly, I could give a flying fig for alpha.


Let's define alpha.  If I say that I get positive alpha on my returns, that means I get the performance of the broad market (in CAPS, the S+P 500's returns), plus a certain percentage.  That percentage - the amount of 'outperformance' - per year is alpha.  Alpha is good; if you are a mutual or hedge fund manager who has to be in equities, it is *everything*.

I don't want alpha.  I want the better of (S+P 500 returns plus alpha) and (the risk free rate.)  If the market drops 40% (as it recently did) I am not consoled when my portfolio only drops 25% (as it recently did.)  If my portfolio is headed for a 25% loss I ought to be in T's.

"Oh, you can't time the market," say people.  "You can't pick individual stocks to outperform."  "You can't beat the pros," they whine.  Some of these same people who say this brag about their alpha.

I can understand why the pros have institutionalized inferiority complexes about their performance, but I think those are "features" that we can add to the list of things that retail investors don't have to worry about, along with things like "How am I going to outperform the market with this $20 billion mutual fund that has to be 80% in equities by charter?"

Now, that said, I'd like to know my CAPS alpha.  It is not just 'average pick score.'  It is 'average pick score' divided by (average pick length, in years).  That data's all in the db - shouldn't be hard to mine it.

4 Comments – Post Your Own

#1) On October 22, 2008 at 9:09 PM, XMFCrocoStimpy (97.58) wrote:


Alpha is everything, provided that you are willing to short the market in beta dollar adjusted amounts against your portfolio (and/or go long the market in equal dollar amounts against your shorts).  The disadvantage that a mutual fund or institutional investment manager has is that they are likely precluded from shorting, whereas the informed individual is free to follow such a path.  Effective hedge design is an artform, especially in crazy market times like this, but done properly you can consistently earn your "alpha" with lower volatility than the market.


Report this comment
#2) On October 22, 2008 at 10:00 PM, rd80 (95.80) wrote:

It's a little more complicated than average pick score divided by pick length because of compounding.

For an individual pick, it would be
((1 + pick score)^(1/nr. of years)) - 1.

Example:  For a pick score of 36 that had been open for 3 years, the alpha would be
(1.36)^(1/3) - 1 = 10.8%

To do the portfolio, you would need to repeat the calculation for each pick, then divide by the number of picks. 

If CAPS had an 'export to Excel' function this would be a piece of cake.  hint hint


Report this comment
#3) On October 22, 2008 at 10:51 PM, ikkyu2 (98.20) wrote:

rd80, you are correct of course.  Hadn't occurred to me.  Your solution is not very elegant when you consider the lumpiness of dividends, but then again dividends are not particularly elegant themselves.  I think it still works, more or less.

Report this comment
#4) On October 22, 2008 at 10:55 PM, ikkyu2 (98.20) wrote:

Stimpy, if your contention is that alpha should be calculated against a baseline of all market volatility - i.e., if the market goes up 3% on Tuesday and down 5% on Wednesday, I need to book a gain of 8% over those two days to be on track for an alpha of 0 - I'm not sure I'd buy that.

I don't think you really need to short the market, but it is nice to be out of it before it tanks.  I only got about 30% of my equity allocation out of stocks before the bottom dropped out of this one. 

Report this comment

Featured Broker Partners