Long Only Dow System >85% win rate

[FetteredChinos]

thought i would update the results of this system for the past 2 years.

25 trades, 20 wins (80%) yielding 1,362 points on the Dow according to the original rules

some mega losses owing to recent turbulence, but still im pleased that its holding up relatively well.. approx 50pts per trade after spreads, which is broadly in line with prior testing..

toodle-oo

fc

 

[Trader333]

What is the number of points incurred as intra-day drawdown ?


Paul

 

[FetteredChinos]

nearly 500 during the february 07 meltdown, after which it recovered in following months...

until the july-august follow-up meltdown.

 

[crud-cutter]

Traded twice in July 2007 - is this in original spec?

thought i would update the results of this system for the past 2 years.
fc

Hey FC,
How come your results spreadsheet has two entries for July07 one on 24th showing a loss of 354.58 closing on 1st August, but then you have another on the 31st July showing a gain of 251.34 closing on the 2nd August. So in any case the 2 trades overlap.
But I thought that the original only traded once per month, rather than as many times as would fit in the trading period?

Puzzled,
Ian

 

[FetteredChinos]

yeah, i noticed that. its a vagary of the formula i used in the spreadsheet. technically i should strip out the second trade, ie the gain of 251, but i wanted to use the same workings i did in the original spreadsheet for the sake of comparison.

overall though, it shouldnt alter the results too significantly over the long haul.. feel free to adjust if you want..

 

[pertrader]

If you allow multiple trades each month the curve looks a little more stable.


59 trades, 51 wins (86%) yielding 3,137 points.


View attachment multiple down day updated results.xls

 

[aspex]

I just discovered this thread so did an exercise on the FTSE100 from 1 Jan 2002.
The result is a gain of +5634 using the multiple method.
Not too bad as the net gain is from index 5218 (1/1/02) to 6479 (24/12/07)
There were 171 trades effected with a points spread of 2, this would drop off 342 points or double that if the CFD was taken outside market close.

 

[london_lad]

the goal really is to find a once a week, set-and-forget strategy, so that i can divert more of my spare time to playing golf and snooker...plus also to reduce stress levels..

the problem with these strats, is that psychologically i need to have a high win %, in order to minimise the chances of having a losing month, as the sheer amount of time it would take to recover from that is rather disturbing..

therefore, im trying to find something quite simple, but with a >70% win rate that i can trade once a week..and be consistent over many years...

could take me a while to find this, but i have come close.. Its just the pesky bear market of 00-02 that worries me, in case we have another one of those very soon..

as for tests on other strats i have posted here... in hindsight the SIBKISS derived ones just arent worth the risk (returns have really tailed off), the swing trading FTSE ones are still going reasonably well, but again have the problem of large drawdowns, which i am not willing to tolerate.

FC

FC: what is that FTSE wing day trading strategy? and is it still working in these markets? (2009)?

 

[perspective]

Hi,

First post here. The system, as described, does give a high % win rate in the order of 85%+ for DJIA. However it looks like it's just another over-fitted system.

I optimized the 2 following parameters:
1. sell after x days
2. sell if stop loss of x% breached to the downside

...then walked forward the optimized parameters on consecutive 6 month periods of out-of-sample data for the DJIA. cmae optimization engine used (non-exhaustive) and %winners was the fitness parameter for optimization.

Results Oct 2003 - 2009:
out-of-sample %winners - 83.3, 77.8, 88.9, 87.5, 80, 66.67, 75, 83.3, 77.78, 57.14, 87.5

A couple of bad 6 month periods in there, and you can go a whole year and lose money.

The other thing is, a high %winners is fine but doesn't mean anything if your winners are smaller than your losers, which is the case with this system. The net profit is all that counts.

I can't find a way to use this profitably.

The fact that this is a 5-star thread makes me think this forum is a dud. Seriously.

 

[metroplex]

"The fact that this is a 5-star thread makes me think this forum is a dud. Seriously."

You got that right cochise...

 

[Vaco]

"The fact that this is a 5-star thread makes me think this forum is a dud. Seriously."

You got that right cochise...

if you look a little closer you will notice that it has only been rated by 1 member.

Hardly a significant representation of the membership.