Oct

21

Oversold, from Paolo Pezzutti

October 21, 2007 |

Suppose you bought any Friday where the stochastic indicator was oversold at the close. What is the percentage of winning trades, placing a sell limit order of c+x points for Monday? I checked in the past 10 years all the situations. If the order is not filled, you exit at Monday's close.

 3 points 96%
 5 points 86%
 7 points 80%
 9 points 70%
11 points 65%
15 points 63%

Larry Williams explains:

the problem is  such an approach has massive equity drawdowns and small average profits per trade. The losses, when they come, are much bigger than the gains. Accuracy alone does not make for a good system or trader. Risk/reward trumps accuracy every time. Eventually large losses devour strings of wining trades.
 
To evaluate such an approach, look at the equity curve; not just the numbers.

Jim Sogi adds:

The equity curve Larry talks about is a thing of beauty. We all know what happened after 1987 as well. The survivors prospered. If you want to argue sample, only time will tell. History unfolds in mysterious ways and you can never know the future. If you always look at 1987, you'll never trade. One way to avoid annihilation in addition to money management is to stay nimble in addition to having deep pockets. Wall Street has deeper pockets than you.

Phil McDonnell writes:

Phil McDonnellAs an augmentation, the following discussion of the features of a normally distribued random walk with absorbing upside barriers should prove helpful.  Naturally as traders this simply means using the theoretical distribution with an upside profit target.

Using a profit target will:
1.  Double the probability of being at or above that target at the end of a fixed period of time.
2.  Have no impact on your expected gain or loss.
3.  Reduce your variance and standard deviation
4.  Result in larger losses than gains

This result derives from the fact that the normal distribution is symmetric and self-similar.  Thus it obeys a property called the Reflection Principle. Each price path has an equal and opposite mirror image.  Each price point reached has a distribution of points past it and an equal and opposite distribution of points which were 'reflected back'.   Elementery proofs for the analogous case of stops, using nothing more than high school algebra, are given in my book Optimal Portfolio Modeling.

It should be emphasized that this is the theoretical model.  To the extent that one can find empirical evidence that the market does not conform to this, there may be something tradeable.  But just because you can manipulate your distribution to double the probability of a winning trade does not mean that the average winnings will be any better My Motto: You need an edge — never let your money leave home without it.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search