Strategy DayOpen Straddle for DAX

Paul · Sun, 04 Nov 2018 13:17:32 +0000

Hi, This strategy is the DAX-open straddle which is posted in the library as DayOpen Straddle for DAX 3-minute timeframe. It’s based on the idea, that while trading is possible 24/7, the big institutions start their day at 9u. Here’s more info https://en.wikipedia.org/wiki/DAX That’s why I took 9u am as start and made it a trade only in the first hour. At 9’ am the strategy defines the open and buys x points above or sellshort x points below, whichever comes first. The assumption is, that once it’s opened, there is high probability that it continues at least x points in that direction. To make it work, there had to be added extra criteria, which are important. i.e. For a long position, the (current bar) high has to be lower than the lowest daily high up to 2 days back. If that’s the case, the index has room to increase. (downside is that when there’s a continues rally, you don’t get in) While it’s mentioned for the 3min bar, the focus should be on the 10- or 15-min bars. In my testing, on both timeframes, I start from 1 January 2015. I didn’t expect these kinds of results! Have a look and if you have any idea’s how to improve or find other markets please post! (pictures posted are with 0.6SL, 0.4MFE and no PT) (NOP 30 for 15min, NOP25 for 10min) Cheers, Paul

This topic has 241 replies, 39 voices, and was last updated 3 years ago by Monobrow.

Viewing 15 posts - 151 through 165 (of 242 total)

← 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 →

07/09/2019 at 5:22 PM #102291

Paul

here’s compared only ts, only bb, both

All a bit the same result, but as you see you surrendered potential profit last day, which wouldn’t happen if both were active.

also both with safety active with 0.50; lower average loss, lower drawdown

4 users thanked author for this post.

Nicolas,

AE,

GraHal,

winnie37

Paul

Participant

Topics: 40

Replies: 812

Been thanked: 660 times

07/14/2019 at 3:57 PM #102557

Thanks you Paul, I love this strategy since I saw your first version.

You said this strategy doesn’t work really good in downtrend, but in TF 1 minuto works even downtrend?

Thanks

Participant

Topics: 11

Replies: 53

Been thanked: 12 times

07/15/2019 at 2:37 PM #102617

Pablo Jimenez

Thanks for all, this strategy is awesome… Could anybody post the latest version of the code ?? Thanks you

Pablo Jimenez

Participant

Topics: 4

Replies: 9

Been thanked: 0 times

07/15/2019 at 4:44 PM #102625

Gregg

Thanks Paul for this strategy! I’ve just spend 2 hours to read all the 11 pages of this topic and trying every version on every UT. Great job! Now I have to decide which UT I will choose to begin.

Gregg

Participant

Topics: 12

Replies: 45

Been thanked: 10 times

07/15/2019 at 6:45 PM #102627

Gregg

Hi again guys,

what about overnight? I tried to add this king of code and it doesn’t seem to be more profitable, but maybe safer with some gap.

//OVERNIGHT BUY 
IF longonmarket THEN
IF time >= 210000
sellshort at market
ENDIF
ENDIF

//OR OVERWEEK BUY
IF longonmarket THEN
IF time >= 210000 and OpenDayOfWeek = 5 THEN
sellshort at market
ENDIF
ENDIF

//OVERNIGHT BUY

IF longonmarket THEN

IF time >= 210000

sellshort at market

ENDIF

//OR OVERWEEK BUY

IF longonmarket THEN

IF time >= 210000 and OpenDayOfWeek = 5 THEN

sellshort at market

ENDIF

What do you think about it?

Gregg

Participant

Topics: 12

Replies: 45

Been thanked: 10 times

07/16/2019 at 12:52 PM #102665

Paul

@AE as it is right know, long could get triggered fast in a downtrend.

@Pablo Jimenez latest version is already posted

@Gregg the strategy doesn’t hold a position in the weekend, which is preferable I think. It only holds overnight during the week.

Adding that code during the week, it closes a long position at 21h and takes a short one.

Safest is change 240000 to something lower so it closes each day all positions. Perhaps turn of trendsafetyexit.

1 user thanked author for this post.

Gregg

Paul

Participant

Topics: 40

Replies: 812

Been thanked: 660 times

07/19/2019 at 8:30 AM #102880

winnie37

I’m running live Dayopen straddle v4.2 with this settings// Money Management// Money Management

// Money Management
CONSECNOCYCLE = 1
CONSECSICYCLE = 0
CUMULASICYCLE = 0
traditional   = 0
percentage   =  0

// Money Management

CONSECNOCYCLE = 1

CONSECSICYCLE = 0

CUMULASICYCLE = 0

traditional = 0

percentage = 0

A curious almost flat evolution since march 2019. And no trades for 3 days now….Paul, you can confirm?

1 user thanked author for this post.

Gregg

winnie37

Participant

Topics: 27

Replies: 165

Been thanked: 16 times

07/20/2019 at 6:45 AM #103000

Paul

Yeah the strategy didn’t produce any trades and that is good because otherwise you would’ve lost.

The reason is the main criteria which uses the Mid-Level Day indicator.

To avoid certain markets, lines have to be different from each-other and now they are not so that’s the reason no trades are produced.

c1=(mid1day) //green
c3=(mid3day) //blue
c5=(mid5day) //red

c1=(mid1day) //green

c3=(mid3day) //blue

c5=(mid5day) //red

c10=abs(c1-c3) > 0  and abs(c3-c5) > 0

1	c10=abs(c1-c3) > 0 and abs(c3-c5) > 0

I included the indicator below. If that is put on a price-chart and you see the same values for green and blue or blue and red it means no trades that day.

1 user thanked author for this post.

winnie37

Paul

Participant

Topics: 40

Replies: 812

Been thanked: 660 times

07/21/2019 at 4:17 AM #103041

dar

Hi All, Is there a new version of this code available? Im running my Test on version 3.

dar

Participant

Topics: 0

Replies: 3

Been thanked: 1 time

07/21/2019 at 12:38 PM #103060

Paul

Yes there’s a new version, still working on a few question marks though.

i.e. mid level day indicator is based on 5 previous days. To make sure that there are no trades in the first 5 days of a backtest, which works on every timeframe and doesn’t influence live trading.

4 users thanked author for this post.

dar,

winnie37,

AE,

Stefanb

Paul

Participant

Topics: 40

Replies: 812

Been thanked: 660 times

07/21/2019 at 5:23 PM #103074

Stefanb

Looks really good Paul.

I like the low DD, only 240€.

During which time is it optimized?

Stefanb

Participant

Topics: 15

Replies: 74

Been thanked: 22 times

07/21/2019 at 6:57 PM #103077

Hi Paul,

When did you estimate could you please send us this new version? It looks really good!

Thanks

Participant

Topics: 11

Replies: 53

Been thanked: 12 times

07/22/2019 at 9:16 AM #103092

dar

Thanks Paul for the reply. Thats looks really good. Can’t wait to test it.

dar

Participant

Topics: 0

Replies: 3

Been thanked: 1 time

07/22/2019 at 11:29 AM #103097

Gregg

Hi @paul,

I compare the strategy with this Mid Level Day indicator and you’re perfectly right, we understand well why there is no trades with these 3 lines! I’m testing some different values for this parameter but I’m impatient to see yours 🙂

1 user thanked author for this post.

winnie37

Gregg

Participant

Topics: 12

Replies: 45

Been thanked: 10 times

07/22/2019 at 1:42 PM #103102

Paul

some info folks.

new version has slightly altered entry. Finally fixed the problem 100% although that doesn’t has great positive impact on results.

Problem was condition for buy could be true and at the same time short could be true.

It mattered if the code had the buy first, or the short first and that influenced results.

The challenge was to make it independed if short or long was first and that the equity-curve was at-least equal or better.

I’ve tested if every criteria is needed. So I compared the main criteria with the extra criteria (latest and first version) and check that with optimised nopl/nops.

As I found out, you can remove the extra criteria completely or just disable it. Run with nopl2 and nops6. Also removed safetyexit, ordertype and lunch.

Version will be published a bit later as I ran into a small issue live.