Random Pick System - Can you beat it ? -Experiment.

How important is the "system" that you use? Or is it ALL about being emotionless and sticking to your rules ?

I am going to test a random picking system and see if it really matters or whether it is more about the let profits run type of stuff.

So here is my very simple plan....

- I will be trading using A Virtual Trading game at
investopedia

- I will randomly select stocks from the S&P 500. I have an alphabetical list. I will use this tool (http://www.random.org/nform.html) to generate a number from 1 to 500.

- Each trade will equal approx. 10% of my total current portfolio, if I don't have at least 9% in cash to trade then I don't trade. (Can only indicate number of shares so this may be a bit off.)
- I always hold 10 shares. However, since I only open a trade when I have time (Unfortunately I have a 9to5 job) I could have from 0 to 10 holdings on any given day if stops are triggered.

--Entry Conditions---
- Random generation of share symbol.
- If S&P 500 on previous day closed above its 50 Day MA - I go Long.
- if S&P 500 closed below its 50 DAY MA - I go short.
- All trades are market and will be open until filled.

Change - Now implementing Coin toss long/short decision.

--Exit Conditions--
-STOP LOSS -
- Set at -5% of the purchase price (.05% of total portfolio)
- Hopefully I remember to set auto stop, otherwise it will be manual and I may miss a stop.

-TRAILING STOP-
Decided with my limited time a formulatic approach for Trailing stop will be too difficult with the online trading game I am using.
-Trailing stops will be set regularly as trades move into profit, with a less systematic approach.

-- Other Exit conditions --
- If I get bored of seeing the same stock without it moving very much, I can exit the trade.
- I can exit a trade at any point when in a profitable position.
- I cannot exit a trade before my STOP LOSS is hit if I am in a losing position

Any comments?
This was a bit quick, so I may have to edit as I reread. But essentially you get the idea.

S&P Above MA - Long

First trade. Random number 364 = DE
Buy 125 @ 80.52 (waiting to fill but must dash off for now)

Date 09/07/2006 12:29
Trade Type Buy Market
Symbol DE
Quantity 125
Price $80.77
Fee $19.99
Total Cash Value $10,116.24

I read somewhere, that a monkey threw darts to pick 10 stocks and a portfolio manager picked 10 stocks in the s&p 500 and the monkey beat a portfolio manager.

It will be interesting to follow this, thanks for detailing your idea. Usually, as long as a method has a positive expectancy and uses reasonable money management, with the trader disciplined enough to follow the plan day in day out, profitability is highly likely...and it's arguable that the simplest methods are the best over the long term.

Are your quotes from a s/bet company so they reflect costs/spread? Anyhow, you've hit the nail on the head in identifying money/trade management as the single most important element of trading (IMO). Good luck - should be an interesting thread.

systematic BUY and SELL

by buying and selling based on a single rule, like buy below or sell above a 200 day moving average you can became a lonesome winner on every instrument.

Is it that what you mean ?

All quotes are what I get filled from the simulator that I am playing, which is supposed to reflect the true price.

Will post total costs of trade, to show true profits.
selected wrong trade earlier as i hadn't sorted spreadsheet. Also i think there is an outstanding trade that could mess me up, but I can't find it in my account.... some hiccups to start with !!

Markets are closed. The following trades will be filled when open.

FPL - LONG
BUY 225 @ Market (Will fill details when trade goes through)

IR - LONG
BUY 260 @ Market (Will fill details when trade goes through)

NEM - LONG
BUY 200 @ Market (Will fill details when trade goes through)

That is all for today. Can't put too many trades into pipeline as I risk not getting my stops in place.

I'm rather confused as to what you're testing? The system isn't random so are you testing a system?

laptop1 said:

I read somewhere, that a monkey threw darts to pick 10 stocks and a portfolio manager picked 10 stocks in the s&p 500 and the monkey beat a portfolio manager.

Click to expand...

Yes and the bunny girls from playboy beat the fund managers because they chose stocks relating to products and services they liked.

I too threw darts into the FT once and had extraordinary luck in picking an oil company that surged up. But is this trading? You could argue that the iChing is random but many people live their lives by it.

Stephen Hawking has just proposed that the Universe may have eleven dimensions of space and time so where does that leave us????

Tuffty said:

I'm rather confused as to what you're testing? The system isn't random so are you testing a system?

Click to expand...

The selection of which stocks to buy is totally random.

Obviously there has to be some controlled elements of the whole system, but I am testing how important stock selection is. People spend a lot of money for black-box indicators and developing their own indicators, just seeing whether that matters, or whether you can select random stocks and as long as you follow the other basic rules, do well.

It seems to me like picking random historical days from the same instrument and applying a simple, non-random system to it. Which is no different to taking each day as it comes on the same instrument. I can't see what randomly switching instruments is trying to show.

If you want to see what trading a partly-random system with strict money management and other rules is like, why not try that? There are some threads here about that - search for 'coin toss'.

The selection of stock is random.

Perhaps to make it more random I should decide on long/short by tossing a coin.

trades filled and stops set at 5%

Date Symbol Quantity Price Fee Total Cash Value
09/08/2006 10:06 NEM 200 $48.75 $19.99 $9,769.99
09/08/2006 10:05 IR 260 $38.20 $19.99 $9,951.99
09/08/2006 10:02 FPL 225 $44.20 $19.99 $9,964.99
09/07/2006 12:29 DE 125 $80.77 $19.99 $10,116.24

Total Transactions: 4

OK, so you're testing a stop loss system with random entry. You may find the more volatile stocks get stopped out unless some volatility measure is included in the stop loss formula.

9/8/2006 12:54 PM 16483416 Buy Market JPM 220 $45.00 $19.99 $9,919.99 $99,865.16
9/8/2006 12:52 PM 16483366 Buy Market WLP 125 $76.44 $19.99 $9,574.99 $99,885.15

9/11/2006 11:17 AM 16512182 Sell Stop NEM 200 $46.23 $29.95 $9,216.05

Total Losses.
-$553.94

9/13/2006 12:08 PM 16562239 Buy Market APL 220 $43.95 $19.99 $9,688.99

Portfolio Update.
$100,201.87

Changed 2 stop Loss to lock in Small profit.

Hi JDR,

I have just logged on and caught your thread - an interesting one and one which I have discussed at some length on this board in the past.

A guy called Van Tharp did numerous tests on this and proved that it IS possible to profit from random entry. I have done my own testing and the results show me that it definitely can be done, but getting most people to believe that is nigh on impossible! Van Tharp used a coin toss to enter the market long or short, a 3xATR trailing stop and a 1% position size.

The resulting strike rate of winners was apparently 38% which compares to most trend-following systems.

It just goes to show how much we all like to over-complicate this trading game!


Thanks

Damian

Hi Damian,
Thanks for your post.
I figured someone must have done such a study at some time. Although my risk and exit rules are not that complex, I hope to prove (to myself mainly) that these are the most important elements.

The reason why Van Tharp did this random entry test was to prove that your exit strategy and money management are far more important than your entry technique, and I agree with him.

Most people only focus on entry.

In fact, most people selling so-called trading systems are actually just selling an entry technique. They never mention things like exits, position sizing, stop loss techniques and money management.

The reason so many traders fail in this business is because they over-focus on entry, when in fact it is their exit strategy and money management strategy that will effect their trading results the most.

You CAN make money with a random entry as long as the other parts of your system are sound and remain constant throughout your trading.


Damian