How do you filter out bad trades? | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Looking for bad trade/noise filter indicator 6 replies

EAs: Spread filter VS News filter 3 replies

What is the best filter you use in your Trading Strategy? 16 replies

Do you use a volatility filter? 8 replies

You Know You've Got It Bad When: 106 replies

How do you filter out bad trades?

  • #1
  • First Post: May 30, 2011 7:47am May 30, 2011 7:47am
Hey all, I need some advice or some tips in filtering out bad trades.

I want to increase the accuracy of my systems from 60% to something higher without reducing my reward:risk ratio.

Some ideas on how to filter out false signals during choppy times would be appreciated.

I have already tried macd, macdx2 (24/52/18), ema crosses.


Now I'm thinking bollinger bands and distance between the bands to help?

Any ideas welcome!
  • #2
  • May 30, 2011 8:00am May 30, 2011 8:00am
What times are you trading?
  • #3
  • May 30, 2011 10:42pm May 30, 2011 10:42pm
Quoting nahman
Hey all, I need some advice or some tips in filtering out bad trades.

I want to increase the accuracy of my systems from 60% to something higher without reducing my reward:risk ratio.

Some ideas on how to filter out false signals during choppy times would be appreciated.

I have already tried macd, macdx2 (24/52/18), ema crosses.


Now I'm thinking bollinger bands and distance between the bands to help?

Any ideas welcome!

With all due respect this is like a train wreck happening before my eyes

It sounds like your issue is getting stopped out before you get started?

let me ask you this...

Before you place a trade what is your current risk and reward?
  • #4
  • May 30, 2011 11:14pm May 30, 2011 11:14pm
Quoting nahman
Any ideas welcome!
1. Pair selection: pair the strongest currency/ies against the weakest - this will, on balance of probability, lead to continuation of the strongest, cleanest trends.

2. Time of day: examine volatility on a session by session basis. What are the best (most volatile) times of the day to trade? To what extent does this vary from pair to pair?

3. Candles: Look for big candles (preferably also with high volume) signaling that price is starting to resume in the direction of the major trend, following a pullback/lull/congestion.

4. RR: Focus not on increasing win%, but finding low risk entries around key S/R levels that can potentially lead to high RR trades.
  • #5
  • May 31, 2011 1:56am May 31, 2011 1:56am
4 hour timeframe, tp 50 pips, sl 50 pips, 60~65% accuracy or so on some pairs
  • #6
  • May 31, 2011 2:30am May 31, 2011 2:30am
Quoting hanover
1. Pair selection: pair the strongest currency/ies against the weakest - this will, on balance of probability, lead to continuation of the strongest, cleanest trends.

2. Time of day: examine volatility on a session by session basis. What are the best (most volatile) times of the day to trade? To what extent does this vary from pair to pair?

3. Candles: Look for big candles (preferably also with high volume) signaling that price is starting to resume in the direction of the major trend, following a pullback/lull/congestion.

4. RR: Focus not on...
Cheers all for the replies!

for 1) do you have any strategies for determining weak/strong currencies ie pip range or new daily high/low across like pairs?

2) most volatile for range/volume for me from australia would be 4pm-12pm where it's cleanest

3) on 4hr doesn't always work as there can be a selloff after a big candle up, unless there;s sustained up move

4) I'll take this point into consideration, would you focus on S/R in terms of price action or volume as S/R?
  • #7
  • May 31, 2011 6:45am May 31, 2011 6:45am
Quoting nahman
4 hour timeframe, tp 50 pips, sl 50 pips, 60~65% accuracy or so on some pairs
Why would you want to change anything. If your figures are correct you have positive expectancy and it's a no brainer to keep trading without optimising any further.

Simple mathematics state you are making a 100 pips every 10 trades, so increase your trade size and make money. If you make 10 tardes a week with the above figues and trade 2 lots, you make $2000 easy dollars, 3 lots $3000 and so on.

The above is all based on sustaining the figures quoted and the sample to generate those figures is large enough that you can confidently pull the trigger everytime your setup occurs.
  • #8
  • May 31, 2011 7:30am May 31, 2011 7:30am
Quoting ciscosmithy
Why would you want to change anything. If your figures are correct you have positive expectancy and it's a no brainer to keep trading without optimising any further.

Simple mathematics state you are making a 100 pips every 10 trades, so increase your trade size and make money. If you make 10 tardes a week with the above figues and trade 2 lots, you make $2000 easy dollars, 3 lots $3000 and so on.

The above is all based on sustaining the figures quoted and the sample to generate those figures is large enough that you can confidently pull the...
yeah your living the dream right now.. your going to be just like ever other new trader who loses everything in search of the holy grail

but there is also so many flaws in your thinking I think you should understand what a "choppy" market is and what a bolinger band is
  • #9
  • May 31, 2011 10:14am May 31, 2011 10:14am
Quoting nahman
4 hour timeframe, tp 50 pips, sl 50 pips, 60~65% accuracy or so on some pairs
yeah man, if your figures are accurate, then I would not change anything and just repeat what you do over and over again. You will be a millionaire in no time.
  • #10
  • May 31, 2011 10:16am May 31, 2011 10:16am
Please tell us your sample size as well, 60% accuracy means nothing without the law of large numbers factored in
  • #11
  • May 31, 2011 10:52am May 31, 2011 10:52am
are you keeping a journal of all your trades? my suggestion would be to do so that you are able to evaluate your trades not just from the standpoint of loss or gain but also from your frame of mind when executing your trade....
in addition to your TA what was the news and the fundies,...you may not use these in your trading strategy but may want to be more aware of them..also, were you stressed when trading or fairly relaxed...this has a big impact on how you're able to read the charts and see the various set ups to a good trade...your emotional framework is the sound foundation for building on a good trading system...honestly, I wouldn't kid you about that!..
  • #12
  • May 31, 2011 11:33am May 31, 2011 11:33am
Quoting nahman
Cheers all for the replies!

for 1) do you have any strategies for determining weak/strong currencies ie pip range or new daily high/low across like pairs?

2) most volatile for range/volume for me from australia would be 4pm-12pm where it's cleanest

3) on 4hr doesn't always work as there can be a selloff after a big candle up, unless there;s sustained up move

4) I'll take this point into consideration, would you focus on S/R in terms of price action or volume as S/R?
Nahman,

One problem with these kinds of threads is that people tend to post what works for them according to the way that they trade, which might be completely different to the person asking the question. Here I am guilty of doing this.

1) I use a strength meter, and look to pair the strongest currency against the weakest in the daily/weekly TFs (at the moment CHF and NZD are strong; USD, JPY and CAD are weak). Then when I get suitable candles on the daily chart (usually an outside bar or engulfing body on the D1), I zoom down to the M15 charts and use S/R to find low risk entries. Thus I have a SL of anything between 5-20 pips. Then I wait for a big daily candle to exit, which occasionally means wins of (say) 50-300 pips. So I don't need a high win rate to come out well ahead.

As a general principle, a muti-time frame analysis can work well, i.e. if you're trading intraday, filter your trades by trading only in the direction of higher TF trends. You are less likely to be stopped out if you're swimming with the tide.

2) The London session is arguably the best time to catch big moves, although the Asian session often gives some great low risk setups, where you get a string of small candles forming multiple (double, triple, etc) tops/bottoms.

3) I was referring mainly to a M15 chart, where big, high volume candles occasionally signpost the beginning of a move (e.g. breakout from congestion) or the end of a move (exhaustion).

4) I use a number of different SR types and look for confluences. This is complex, and very discretionary. It would take me many hours to post different examples. One day, when I have time, I'll start a thread about it.

Sorry, I guess I haven't been much help, given that you wanted to specifically increase your win%, and you were asking which indicators to use.

David
  • #13
  • Edited 2:28pm May 31, 2011 1:35pm | Edited 2:28pm
[quote=hanover;4656322]Nahman,

One problem with these kinds of threads is that people tend to post what works for them according to the way that they trade, which might be completely different to the person asking the question. Here I am guilty of doing this.

thanks Hanover, You are correct...
I was posting about what works for me....
I'm just getting back into trading from having had to take time off for a few years...
the last I had been trading, I had a pretty good accuracy rate...
the reason for this is that I learned to trust my knowledge and skill..
In my opinion, Nahman is doing very well but could use a little more encouragement to focus on the good trades and why they worked....
it may not be indicators that improve Nahman's game...just sayin'...
  • #14
  • Jun 2, 2011 9:06am Jun 2, 2011 9:06am
Trade less and trade at the extremes of price will both increase accuracy.
Experience is the best filter, learn from your errors.

Maybe you'll benefit from trading 2 lots, allowing 1 to run. 4hr chart gives up several hundred pips if you follow the flow.
  • #15
  • Jun 3, 2011 3:38am Jun 3, 2011 3:38am
Quoting newyear498
Please tell us your sample size as well, 60% accuracy means nothing without the law of large numbers factored in
7 year time frame 2004-current, EUR/USD, AUD/USD, USD/CHF, USD/CAD, GBP/USD and to be tested still: NZD/USD, EUR/CHF


537 winners 357 losers.

I still want to optimise to get at least 70% as it's safer to bet larger on something higher accuracy like this

there's a trade roughly every 2.8 trading days with those statistics..
  • #16
  • Jun 3, 2011 3:40am Jun 3, 2011 3:40am
Quoting insight
are you keeping a journal of all your trades? my suggestion would be to do so that you are able to evaluate your trades not just from the standpoint of loss or gain but also from your frame of mind when executing your trade....
in addition to your TA what was the news and the fundies,...you may not use these in your trading strategy but may want to be more aware of them..also, were you stressed when trading or fairly relaxed...this has a big impact on how you're able to read the charts and see the various set ups to a good trade...your emotional...
pure price action with only one moving average - nothing else, the simpler it is the more it works
  • #17
  • Jun 3, 2011 3:44am Jun 3, 2011 3:44am
Quoting jpadventure
Trade less and trade at the extremes of price will both increase accuracy.
Experience is the best filter, learn from your errors.

Maybe you'll benefit from trading 2 lots, allowing 1 to run. 4hr chart gives up several hundred pips if you follow the flow.
Hmm this possibly means (which I had a hunch on) trading only at the possible start/middle of the trend and ignore anything until trend reversal, I've noticed that this system does sometimes take losing trades near bottoms and tops of trends
  • #18
  • Jun 3, 2011 7:14am Jun 3, 2011 7:14am
Quoting insight
are you keeping a journal of all your trades? my suggestion would be to do so that you are able to evaluate your trades not just from the standpoint of loss or gain but also from your frame of mind when executing your trade....
in addition to your TA what was the news and the fundies,...you may not use these in your trading strategy but may want to be more aware of them..also, were you stressed when trading or fairly relaxed...this has a big impact on how you're able to read the charts and see the various set ups to a good trade...your emotional...
I think this might be good to evaluate what is a key factor in what is happening in losing trades, can be helpful definitely
  • #19
  • Jun 3, 2011 3:48pm Jun 3, 2011 3:48pm
Maybe you can also look into making a filter that prevents you from taking trades during major announcements.
The nail that sticks out gets hammered back in
  • #20
  • Jun 3, 2011 5:46pm Jun 3, 2011 5:46pm
Simple method for medium to long term trading systems, filter out trading against the trend. How to do this? Compare the price the market opened at 100 days ago to the price today. This works better than moving averages.
1 trader viewing now