Simple Strategy : TF 10S / Bollinger / RSI

Viewing 15 posts - 1 through 15 (of 50 total)

1 2 3 4 →

Author

Posts
04/27/2020 at 3:03 PM #128443 quote
Scooby
Participant

Senior
Hi all,

I would like to apply a new stratégie which provide good signals in intraday / Scalping with a TF10s.

I found a very simple stratégie due to a french trader who apply it in manual. Then I tried in automatic.

Stratégie :
- Buy when :
  - A last 5 Close of signal (heikin ashi) is up to the upper band of bollinger
  - Close < Open (heikin ashi)
  - a last 5 RSI (14) is higher than 70
  - a small volacity before the entry calculated with the difference between the two band of bollinger
- Sell (exactly the same parameters in reverse)
I leave the positions when :
- Cross the bollinger band (down one when I sell, up one when I buy)
- close > open or open > close (depend of buy or sell)
I know that it’s a very simple stratégie, but it gives for the previous some day some good results for EUR/USD – NASDAQ – DJI

Do you think that’s it’s because we are currently in a specific situation ?

Fred23 and swedshare thanked this post

Pic2.png V1-EURUSD-10S-Boll-RSI.itf
04/28/2020 at 8:37 PM #128727 quote

Scooby
Participant

Senior

I don’t know if I did’nt respect the rules, or if nobody think that it’s a good idea 😀

I adapted the method for DJI market in real and I got a pretty good result today (obviously, I did a 0.2 contract for 1 euros by point).

04.28.20.png

04/28/2020 at 9:25 PM #128731 quote

nonetheless
Participant

Master

Salut @Scooby, here’s a true story: a couple of weeks I wrote 20 micro-trend algos all running on <1m TFs — 4 secs, 7secs, 12 secs etc.

All perfectly sensible, logical strategies with an intricate combination of funky indicators that on paper had no reason not to work. Most had 100% win rate in back test, none less than 95%. Put them on forward testing (demo!) thinking for sure I had a cash cow. Blew up my account in a day with massive losses.

Moral to the story? I don’t go near micro-trend algos any more. If I can’t do a back test of at least a year — at 200k that means 2m TF — it aint got no credibility. At all.

Having said that … very best of luck!

PS, sorry to be a wet blanket! maybe you’ve got a winner, who knows?

swedshare and Bel thanked this post

04/29/2020 at 9:37 AM #128785 quote

GraHal
Participant

Master

Scooby wrote:

if nobody think that it’s a good idea

I think it’s a good idea, so keep em coming!

I like low TF strategies, quite happy with a quick in and out … or if the strategy lands on the crest of a wave I’ll ride it out to the shore! 🙂

Does better on DJI 10 sec TF … today anyway! 🙂

Takes almost zero Long trades on DJI anyway … seems like it needs to?

04/29/2020 at 11:07 PM #128896 quote

Scooby
Participant

Senior

Yes DJI gave me good results in real today 🙂

I was with stupids lots with my probuild (0.2). I incrised it to 1 for the next days.

I did a script for the NASDAQ (same base) and it worked today aswell.

Do you know an indicator like bollinger range to catch a probable return that we can compare with this strategy ?

I’m working on the russell, a very stressed market. Maybe it can be fine with this method.

If you have an idea(s) to improve it.. 🙂

bertrandpinoy thanked this post

04/30/2020 at 9:33 AM #128951 quote

GraHal
Participant

Master

Scooby wrote:

Do you know an indicator like bollinger range

Donchian Channel may do what you require?

04/30/2020 at 10:17 AM #128964 quote

Scooby
Participant

Senior

I’m trying to adapt some others minor and exotics forex with the first version this morning.

This afternoon I gonna try the donchian for the entry trades thx !

How do you valide your 10 sec strategies ? there is only 1/2 weeks of data… difficult to be sure that it will be winning for the next few month right ?

Nevermind, yesterday was a great day with the 4 strategies with this method. I will do a message with the results day by day or something like that if someone wants to follow it.

bertrandpinoy thanked this post

04/30/2020 at 10:47 AM #128971 quote

GraHal
Participant

Master

Scooby wrote:

I will do a message with the results day by day or something like that if someone wants to follow it.

Yeah that be good with a screen shot of the performance results overview and the equity curve if not too much trouble?

With so little history the only way to validate is to Demo Forward Test for a month or so?

swedshare thanked this post

05/01/2020 at 1:13 PM #129202 quote

Scooby
Participant

Senior

Then results from yesterday on EUR/USD and DJI

bertrandpinoy, Fred23, GraHal and Paul thanked this post

Resultats-05.01.20.png DJI-Résultats-05.01.20.png

05/01/2020 at 3:53 PM #129241 quote

swedshare
Participant

Senior

It’s an interesting strategy Scooby. As “nonetheless” mentioned, the timeframe is not ideal for building a long term strategy. If you can adjust it to timeframe 1min, 5 min or 15 min it would be easier to see if it’s a winner in the long run.

I added a few pieces to your code: MoneyM, riskM, tighter stoploss and target profit plus a calender to prevent trades on days with higher spreads. See attached file if there’s something you want to use.

Scooby thanked this post

EURUSD-10s-Boll-RSI-v2.itf

05/01/2020 at 3:54 PM #129243 quote

Scooby
Participant

Senior

I put a trailing stop today on the DJI. The results obviously are better… ^^

I saw today a good trade, but without trailing stop the trade is bad because it wait to find the SL or a return on the bollinger band.

Before and after attached, and the result of the current day

GraHal and swedshare thanked this post

V1-05.01.20.png V2-05.01.20.png

05/01/2020 at 4:39 PM #129265 quote

Scooby
Participant

Senior

Thanks @swedshare !

I updated the algo for DJI with your work on EUR/USD

The backtest on 1M bars does exist ? Someone can try it ? 🙂

DJI-10S-Boll-RSI-V2.itf

05/04/2020 at 9:27 PM #129924 quote

Scooby
Participant

Senior

Results for DJI and EUR/USD.

The DJI is only on 2 days here, because I did a modification 2 days ago and restart it. Last week, it was fine aswell.

Hope it will continue 🙂

swedshare thanked this post

DJI-V2-05.04.20.png EUR-USD-Resultats-05.04.20.png

05/05/2020 at 8:13 PM #130131 quote

tecknozic
Participant

Average

Hello, did you continue your strategy in order to backtest it ? Did you get good results ?

Thank you.

05/06/2020 at 6:25 PM #130296 quote

Scooby
Participant

Senior

Hi @technozic, yes I’m still running the EUR/USD and DJI in real.

The last two days I found a difference between my backtest and the real.

I checked that this was exactly the same candle of entry and exit, that the same, but the result is different.

This is because the trading on 10 secondes is not precise ? (pictures)

Backtest.png Real.png
Author

Posts