Nasdaq Daily Breakout

[FetteredChinos]

Afternoon all,

been trawling through the posts back on the Contemplating Trading Strategies Board on the Motley Fool.

rediscovered the JT Daily Breakout over on there, originated by our esteemed colleague Mr JonnyT

basically, the simple rules are go long when we break yesterday's high, go short when we break yesterday's low.

close all trades at the daily close.


using data from livecharts/dukascopy, i have been testing this on various indices.

they all work reasonably well, but recent performance has been a bit stodgy.

APART from the Nasdaq-100.

on the data i have, testing it from 1990 to present date yields a frankly astonishing 12133 Big points, or 121,335 points when spreadbetting....

this equates to approx 35 per trade. LONG TERM AVERAGE

with a win rate of 57%.

only problem is the drawdown, which at the height of the bull market hit 784 big points (7840 tradeable points).

since then, for the past 4 years, the max drawdown has been in the region of 200 big points.

much more bearable.


i've only been testing this on OHLC daily data, but frankly, even with a few "funnies" in the data, it looks pretty healthy.

could someone please test this on intra-day futures data, just to check that the entries are indeed feasible and dont get whipsawed to death?

ta

FC

 

[FetteredChinos]

Right, here we go.

some modifications to make the whole thing a lot simpler..

da roolz:-

Go long at the close every day. set a stop and reverse up at the close level minus HALF THE DAILY RANGE. currently this is about 20-21 big points (200-210 tradable points)

close all positions at the close of play.

this has yielded 126,344 points since 1985.

didnt appear to suffer too much during the 2000-2002 bear market either.

returns since jan 04 are approximately 10,048 points.

this is mental.

also, dealing costs are reduced, as we roll each long trade over into the next day if we arent triggered into a short.

i have chosen a 0.5 of the daily range, so as to try to ensure that we dont get stopped out by overnight movement. pretty much any value works, but with a correspondingly wonkier equity curve..





the nasdaq-100 doesnt currently fake moves of 10 points regularly out of hours does it?

it does appear that the Nasdaq does only move in one direction per day, unlike the dow.


interesting..


fc

 

[FetteredChinos]

works on the Dow to a degree...

 

[FetteredChinos]

and the FTSE, with even larger %'s of ADR, and therefore more likely to be accurate.


hmm again..

 

[FetteredChinos]

and rather encouragingly, over the last 20 year on the nikkei it has also been profitable..

see the attached equity curve.

this is despite the nikkei trending DOWN for the last 10 years..

this might have to warrant further investigation...

 

[FetteredChinos]

and IBM (chosen at random) since 1962....

 

[FetteredChinos]

dont all rush at once..

 

[mofo]

I'm listening me old mate, even if noone else is.

 

[FetteredChinos]

cheers pal, was getting worried for a moment

 

[gizmo]

Looks very interesting FC.

May I ask why you have chosen 22 days to determine the average range?

Also any chance you can post the FTSE & NIkkei spreadsheets please.......

Gizmo

 

[FetteredChinos]

erm cos that was where my finger stopped!!!

changing the range period makes bugger all difference over a length of time.


spreadsheets to follow..

fc

 

[FetteredChinos]

and the Ftse....

bear in mind that the futures do move out of hours..

BUT... i would be surprised if say the futures moved down half the average range, and then spiked back up before the cash market opened on a regular basis.

 

[mofo]

FC, don't suppose you've had a look at the DAX ?

 

[ddunne82]

FC,

What kind of drawdowns have you got on each of the markes so far? Also, are you looking at using futures or spread betting the cash?

David

 

[ddunne82]

Meant to put this is the last post - i dont get the rules?

This makes sense:

1) You go long at the close of each day
2) Stop/reverse level (ie short) at 0.5 daily range - closing price

What about closing your position? Do you move your stop/reverse at the end of each day? So if you hit your reverse, you are only going to be in that for that one day as you're gonna go long at the end of it?

Thanks

David

 

[FetteredChinos]

yup correct. you are only short from when triggered to the end of that day, and then long from that close.

eg yesterday on the FTSE, we were initially long from 5000.

dropped through half the average range at approx 4980 so long was stopped out for -20

when then we were short from 4980 as well to bag +42.

total profit for the day, +22

currently long from 4938, stop and reverse at 4915 to take account of spreads etc.


yes, i would move the stop/reverse each day. so it is kind of a trailing stop for long positions, but also a trap for the sudden sell-off.

max drawdown on the FTSE has been 798 points. which is not too bad when being long through a bear market.

will have to fwd test this to see if it is fully tradeable as it stands, as it looks to be a solid little earner at the mo, for frankly bugger-all effort.

as such, i am naturally sceptical.

will test Dax amongst others. from other backtests i have done, the extended hours that the Dax has makes breakout trading (which is what this is a variant of) more profitable than other indices.

Eurostoxx is even better than the dax, from what i recall, so i may have a crack at that one as well. (something like 80% of daily breakouts ended in a profit)

menthol.

FC

 

[FetteredChinos]

hmm, Dax isnt too pretty in comparison with the others from running a quick test.


seemed to suffer more than most during the bear market.

fc

 

[FetteredChinos]

Eurostoxx on the other hand loves it.

more points than FTSE and with less drawdown (457 points!)

 

[FetteredChinos]

am gonna try running it with the rules reversed.

ie going short everyday.

see if that makes too much of a difference.

at least that way you arent charged the interest on overnight positions.

 

[mofo]

Bugger. I prefer trading the dax to the ftse / ndx - never seemed to have much luck with those for some reason , whereas dax has always been a decent earner