Tuesday, October 03, 2006

Quote of the Week

"Many questions are unanswerable.  Many answers are questionable."  -- from a fortune cookie
Wow, the above quote is so true.  I have dug a little deeper into everything I have worked on the past several years.  Calling into question my beliefs and attitudes towards the market.  I was so wrong.  It all started from a seed that Eric Crittenden planted into my head.  "Sounds like an exercise in curve-fitting", he said, in reference to one of my system ideas.

Then a friend introduced me to the concept of focusing on what you don't like to do and casting it aside in order to free yourself for the things you do like to do.  The butterfly began to flutter...
"It has been said something as small as the flutter of a butterfly's wing can ultimately cause a typhoon halfway around the world." - Chaos Theory
Next, I watched the recent show on Sabermetrics where they discussed Bill James and many of the very cool things brought out in Moneyball.  Flutter, flutter.

Finally, my recent foray into the hazards and pennywinks of developing a trading platform has brought out a very interesting focus to my trading.  What would I like in a platform?  What am I really trying to test?  How is a certain test helpful to my bottom-line?

And all that has helped me to understand what I've been missing.  I've been focusing on the wrong thing!  So much of my time was spent on my next trade.  Kinda like in Sabermetrics where they found too much focus was on RBI's or Homeruns.  Bill James found Outs was where the focus should be. 

And I think in trading...the focus should be on the only fixed rule that I know exists:  If you don't use margin...your losses are limited to 100%.  But, as long you don't cap your profits in any major way...your gains are infinite.  With that in mind, where should your focus be?  And what kind of formulas and tools can we use to measure this new focus?  For example, the smoothness the Sharpe Ratio tries to show becomes something of a throw-away...a tool/formula used to measure the wrong focus in your trading.  Don't understand?  Maybe this will help.  Or maybe not:
http://www.fooledbyrandomness.com/0603_coverstory.pdf

Later Trades,

MT

2 comments:

jontait said...

I'm not really with you on this one. Taleb is dressing up an efficient markets argument in a lot of obscure multi-part words. Basically what he is saying is that the distribution of future returns could be anything, and since we don't know what it will be, then we can't profit from models of it except by chance.

Chaos theory is where I disagree with Taleb. Looking at all the current weather in the United States, can we not say something meaningful about what tomorrow's weather is likely to be in the midwest? Yes, we can, but the further out in time we try to go, the less accurate we can be with our forecast, so we model accordingly. This is good, useful curve fitting if you bet it like a forecast and not a divination.

Michael Taylor said...

Hey Jon!

Loved your interview over at StockTickr.

I see your point...and believed in it 100% up until just recently. Then kinda hit an aha! moment of clarity that allowed me to really see what Eric C., Taleb, and even Acrary were trying to say.

I'll write up a post later about my thoughts...but think of it this way. How many times in your trading life have you done really well? Gains slowly or maybe quickly building up to a great performance record? Only to have one or a few trades tear it down? That's what I'm referring to.

Not saying that you cannot find an edge or that the market is always random. Just saying that random bits do occur and when they do...they can knock your great edge down to nothing real quick before anything you have built can stop it. So, instead of trying to stop it...just open up your system and tests to those possibilities. Periods of edge mixed in with periods of random behavior.

Thinking in this way not only will keep yourself from riding the curve-fit too close but actually open you up to what random behavior can do to your returns. Just like they can knock them down...they can also mark them up.

More later...

MT