New member - Help with my system please!

Quote from MrJohnGalt:

Before you give some suggestion...

Do some tests or make some sense...

1. Tested your suggestion with MACD, Stoch and CCI. Doesn't work.

2. Put it through parameter optimization. No significance.

3. Regardless of cyclical factors or changing markets... You'll lose money in the long run how ever you mess around with the parameter. No significance even if you make the parameter optimized on X amount of drawdown, specific interval or cyclical adaptation.

Because this is Index Futures... tested it on ES and NQ...
More...

I suggested to use divergence and TRENDLINES . For some reason you overlooked that TRENDLINES part .
Other than that you are correct on plain divergence testing. Good job at pointing that out .
 
Quote from Walther:

I suggested to use divergence and TRENDLINES . For some reason you choose to ignore that TRENDLINES part .
Other than that you are correct on plain divergence testing.
More...

Sheez... OK... lemme put some Top x Bottom slope stuff on there... I'll test it...

Slow day... :)

Also... I think the trendlines will be used in a conventional way as support and resistance, breakouts, and counter-trend setup, then using divergence and RSI / Bollinger on... 1 min. to 10 min. bars...
 
Quote from Jinkers:


The system is very simple, in the ES, on a 9 minute time frame:

RSI set at 6
Bollinger Bands at 10 and 1

Buy when RSI dips below 30 and reemerges above 35 AND price is under BB bottom line. Sell when price either touches top line of BB or breaks through it and reverses back down to touch it again. 2 pt stop.

Only trade during decent volume and range (after 8:30ish)
More...

Seems like a pretty decent strategy. The only problem I have with it is there's no trend indicator. Are you using something to help you determine if you're in an up or down trend?

If you had something to determine the trend and only took the signals with the trend, I don't see why it couldn't have some potential.

Also, have you thought about shorts? Maybe, something like RSI above 70 and dips below 65?

But, I think first you need to choose something for a trend indicator, and only take the signals with the trend.
 
Quote from MrJohnGalt:

Sheez... OK... lemme put some Top-Bottom slope stuff on there... I'll test it...

Slow day... :)
More...

You cannot really test that. pm me email address you use for a junk and I will mail you a chart so you know what I mean .

Tequilla or absint ?
 
Quote from funky:


so am i missing something? get some backtesting software (the best ones out there are pretty cheap in comparison to the amount of risk capital you'll probably be trading). it might take a few days to learn the basics and code up your system.

www.tradestation.com
www.wealth-lab.com
www.esignal.com

pick one and have a blast :D good luck!!!
More...

Add to the list Investor RT (www.linnsoft.com). Beats the others hands down for backtesting. Much faster and easier to use.
 
Quote from Walther:

You cannot really test that. pm me email address you use for a junk and I will mail you a chart so you know what I mean .

Tequilla or absint ?
More...

Heh... what do you mean I can't test that?

Of course I can... trendlines = Top x Top / Bottom x Bottom slope ( not recent 2 or thing like that... Going through multiple tops and bottom )

Then it depends on how the price bar itself reacts. Given some freedom of the value... High = TrendL ... High > TrendL and ( Close < TrendL or Close = TrendL ) ... etc.... etc....

When it comes to lines and circles, I think I've done quite some tests with it....

And don't care about sharing some chart... I've dealt with discretionary traders wanting to have their methodology tested and most fail, ending up them hating system trading... He did say "system"...

Anyways... I'll just give it a go or no go... Shouldn't be long, I have most of the line function in archive...

Also... I've got multiple timeframe stuff covered too :)
 
IMHO If this is the gist of your system then get ready for HUGE drawdowns. Especially during STRONG trending markets.

You need a few more components before trading this system.
All you have right now is only part of the total structure.

But hey what do I know?


Best regards

T-REX




Quote from Jinkers:

Hi, I've been reading these forums for months; read the best systems thread, the simple systems thread and everything else pertinent to ES trading.

I would love it if some of you with more experience would tell me if my system is going to work in the long term or am I likely to lose it all.

The system is very simple, in the ES, on a 9 minute time frame:

RSI set at 6
Bollinger Bands at 10 and 1

Buy when RSI dips below 30 and reemerges above 35 AND price is under BB bottom line. Sell when price either touches top line of BB or breaks through it and reverses back down to touch it again. 2 pt stop.

Only trade during decent volume and range (after 8:30ish)

Is this clear?

I had originally thought based on paper trading this was a decent strategy, but now its not working out for me as well as I'd hoped, is the system fundamentally flawed?

Also, does anyone have the resources to backtest it for me? I'd really appreciate it if its not a huge hassle.

Thanks!
More...
 
OK... here's what I did...

First, I tried to use the TrendL as a setup but because the indicators were lagging... (yeah I was stupid enough to not notice that)... I changed it around...

I wouldn't say all but I put in the conditions of conventional use for RSI, Divergence as setup and Bollinger and TrendL as a entry trigger.

Considering this being an Index Futures Thread, used NQ for the test. Reason not ES is that intraday moves are too choppy for the use of RSI and Divergence stuff. Also, this was backtested for 150 days... because I didn't want to take more than an hour for the tests....

Well, the result was... NO... The setup and trigger has no edge in itself, long or short... actually don't go short... you'll be chopped up more...

But... there were some good run-ups but the criterias of reversal doesn't catch it fast enough... Even if Profit target is covered efficiently, that only takes the system to little above B/E...

What I would suggest is to try to come up with some underlying market condition, a hypothesis... then try to figure the best technique to consistantly expose it, rather than just putting some textbook indicator randomly.

This doesn't really matter whether you have the trend or not...

The setup itself has no edge... even if you get a trend catching logic, you'll mind as well use other setups and entry trigger for it...
 
You must log in or register to reply here.
Back
Top