Stock System

[jdoe]

By entering the 2 + 2 trading competition is has got me to thinking about a system to use. I thought I would post my ideas and let you all poke at it.

Say I start with $15,000 no margin

Lets say I have a STOCKBOT to follow all of my commands for me.

I am going to tell my STOCKBOT: at 10:30 pick a random stock with high volatility and volume then randomly decide to take a short or long position.

I am next going to tell the STOCKBOT to stop-loss cover if the position puts me in the red by L%

I am next going to tell the STOCKBOT to run a basic TRAILING STOP PROGRAM if the position puts me in the green by P% to lock in profits.

Close out the position at the end of the day.

Questions:

What should L% be?
What should P% be?

If L < P and pure random forces are at work here wouldn’t this system be +EV?

All comments welcome.

Thanks,
Jdoe

[wildwood]

I'm thinking P should be at least 3x L to give you enough profits to overcome slippage and transaction costs. Slippage and transactions costs are very important to factor in. They sometimes will make short term profitable systems unprofitable. Gaps are not much of a factor since you're not holding overnight. Just my thoughts.

[Dan Mezick]

It looks like you may want to trade volatility.

If you want to trade in volatility, the best opportunity for that may be in the options market.

The books of Larwence McMillan may be the best books for learning more about techniques to do that.

I believe a system of randomly going long or short a random stock that has a random spike in volatility may produce very wide variance (drawdowns to core equity) and random results.

[Sniper]

[ QUOTE ]
Lets say I have a STOCKBOT to follow all of my commands for me.

[/ QUOTE ]

Unlike online poker rooms, online brokers hapilly handle automated trades, and there are several software apps specifically designed to automate system trading. Tradestation may be the most popular.

[ QUOTE ]
I am going to tell my STOCKBOT: at 10:30 pick a random stock with high volatility and volume then randomly decide to take a short or long position.

[/ QUOTE ]

If you want to trade based on a system, why would you pick one that makes random selections... thats -EV

There are many predesigned systems that have proven track records... +EV

[ QUOTE ]
If L < P and pure random forces are at work here wouldn’t this system be +EV?

[/ QUOTE ]

The stock market is not random!!!

[wildwood]

[ QUOTE ]
If you want to trade based on a system, why would you pick one that makes random selections... thats -EV


[/ QUOTE ] quote from Trade Your Way to Financial Freedom by Van K. Tharp. pg 200, "Our random entry system - consisting of random entry, a three-times-volatility trailing stop, and a simple money management system involving 1% risk-made money on 100% of the runs." (Tharp is one of the original market wizards)

[ QUOTE ]
There are many predesigned systems that have proven track records... +EV


[/ QUOTE ] Systems not designed by the user are extremely difficult to trade through the inevitable drawdowns. Schwager talks about this on his video "Winning Methods of the Market Wizards"

[ QUOTE ]
I am next going to tell the STOCKBOT to stop-loss cover if the position puts me in the red by L%

I am next going to tell the STOCKBOT to run a basic TRAILING STOP PROGRAM if the position puts me in the green by P% to lock in profits.

Close out the position at the end of the day.


[/ QUOTE ] [ QUOTE ]
The stock market is not random!!!

[/ QUOTE ] OP is not taking positions lasting longer than one trading day. I think random does apply to extremely short term time periods. OP is cutting losses short and letting profits run. A short term trend following system that might work if money management and position sizing principles were included.

[Sniper]

[ QUOTE ]
quote from Trade Your Way to Financial Freedom by Van K. Tharp. pg 200, "Our random entry system - consisting of random entry, a three-times-volatility trailing stop, and a simple money management system involving 1% risk-made money on 100% of the runs." (Tharp is one of the original market wizards)

[/ QUOTE ]

Just because a system makes +$, doesn't mean that you couldn't do better with a non-random system.

Comparison to poker, where a losing player can still be +$ after RB & Bonus, but certainly could close some leaks and be up even more.

Van Tharp is excellent reading.

[ QUOTE ]
Systems not designed by the user are extremely difficult to trade through the inevitable drawdowns. Schwager talks about this on his video "Winning Methods of the Market Wizards"

[/ QUOTE ]

I agree that you shouldn't be trading any system unless you fully understand it.

[ QUOTE ]
OP is not taking positions lasting longer than one trading day. I think random does apply to extremely short term time periods. OP is cutting losses short and letting profits run. A short term trend following system that might work if money management and position sizing principles were included.

[/ QUOTE ]

I don't agree that the market moves randomly, especially over short time periods. The market is moved by supply/demand, market makers playing the spread, news, technical patterns (because so many people play them thus impacting the supply/demand), cash inflows/outflows (bonus time, taxes, monthly allocations to retirement accounts), hype, etc.

[Dan Mezick]

I suspect the major leak with this design is the huge slippage that will be experienced when you take the wrong side and the security selected moves quickly against your stop. You are going to experience some true movers that go the wrong way and generate huge slippage on your stops. Some of these may produce potential catastrophic losses when such trades string together in sequence.



I suspect the AVGLOSS is going to be quite a bit higher than the AVGAIN because of this, so you will need a very high % correct on directional movement... something that is going to be hard using a random trade direction (long short) method.

[Dan Mezick]

This system might work if you keep the position size very small and execute on a great many trades. But, even with that approach I believe slippage and xaction costs will prove it to be -EV.

You might consider:

1. Any and all systems design books by author Perry Kaufman;

2. Google: "Trading Strategies: Using computer simulation to maximize profits and control risk"

(2) is an very valuable read for anyone looking seriously at developing and implementing mechanical systems. It has much to say about all the major issues related to risk definition and control.

Alot of what is said in this excellent read (65+ pages) is non-obvious even to semi-experienced traders.

I hope this is helpful to you.

Links:
http://www.tradelabstrategies.com/

http://www.tradelabstrategies.com/cu...release301.htm

[wildwood]

[ QUOTE ]
Just because a system makes +$, doesn't mean that you couldn't do better with a non-random system.

[/ QUOTE ]
I think this is true, but it was not my point. Random entry can be +ev as Tharp proved showing that exits and money management are more important than entries.

We simply have to agree to disagree on whether or not the market is random in the short term. I think it is.

[Sniper]

ok, I went back to the source and pulled Van Tharp's book off my shelf to review the context of your quote...

Here are some additional quotes from the same chapter:

"The first approach is to assume that reliability has some importance and to look for signals that are better than random."

"Readers who have studied my Peak Performance Course for Traders and Investors know the importance of stalking the market. Stalking amounts to waiting for exactly the right moment to enter the trade so that risk is minimized."

"I've divided this chapter into four sections. The first section has to do with random entry and with the research designed to increase one's reliability over random entry into the market."

In short, Van tharp uses the example of a random entry system as a baseline to compare other more managed systems. He is certainly not recommending that you trade a random system.