Free end of day system, around 1.25 to 1 winnings to losses ratio

[Davidee]

Hello,

I was just doing an article on pivot points to check if they were statistically significant.

Here's the test -

Take a reading of the trend strength as a number between 0 - 100 using the formula (100/(High-Low))*(Close-Low). 0 means close = low, 50 means the close was in the middle of the range, 100 means close = high.

Do an 8 day SMA on the trend strength reading.

Then if the trend strength SMA is above 65 go long, if it's below 35 go short.

I wanted to compare going long or short the moment the close put the reading above 65 or below 35 with waiting for the price to trace back the pivot point.

Close the entry at the closing price 24 hours later, close the pivot point order at the next day's close regardless if it is hit or not.

I found no evidence that pivot points were statistically significant, but I thought the trend following aspect alone produced enough of a tradable edge. I'll be making this into a 'proper' system soon, right now I think there is enough of an edge to be tradable.

Here's the article -
http://www.myforexdot.org.uk/PivotPointAnalysis.html
I have included the results back tested on 10 currency pairs from 2006 to date.

I'd be really interested to hear what people would suggest to improve the system.

 

[danbradster]

I tried this code in probacktest on a daily chart for 2 years of the ASX200, it made a loss.

total = 0
//n1 is auto-filled from 2 to 10 to test all possibilities
for x = 0 to n1-1 do
	total = total + (100/(High[x]-Low[x]))*(Close[x]-Low[x])
next
trendstrength = total / n1
if trendstrength > 65 then
	if NOT LONGONMARKET then
		BUY 2 SHARES AT MARKET RealTime
	endif
elsif trendstrength < 35 then
	if NOT SHORTONMARKET then
		SELLSHORT 2 SHARES AT MARKET RealTime
	endif
else
	SELL AT MARKET RealTime
	EXITSHORT AT MARKET RealTime
endif

So it longs at 65, shorts at 35 and closes positions if the value is between the two.

 

[bluewater18]

Quote (from the web link): "The ‘major’ currency pairs faired the best, which should have been no surprise to me as these pairs tend to trend the best and this system/experiment was aimed at only trading in strongly trending markets."

Hi Davidee

How do you define "trend the best" and "strongly trending markets"?

Bluewater18