DislikedI am interested have to do some back testing, why only one pair.Ignored
Here's Tom with the weather.
- #22
- Edited 2:34am Oct 4, 2017 2:23am | Edited 2:34am
- Joined Jul 2011 | Status: Gold Again | 10,569 Posts
Hi guys. At the start of every week (Sunday 9GMT here) wait for the first 4h candle to complete (candle 1). After that look for a 4h candle (candle 2) to close above or below candle 1. Buy rules: Place a buy stop order 2 pips above candle 2 Sell tules: Place a sell stop order 2 pips below candle 2 Risk: 2-5% risk per trade Stop loss: 1.5 times the 4h ATR (14 days period)
1. How do you calculate the ATR of a 4 hour chart using 14 days?
I assume you mean the ATR of 4 Hour chart using 14 x 4 Hour periods, nothing to do with days
So why 14? there are 30 x 4 Hour bars in a week, it makes more sense to average the data of the past week right?
2. Reporting profits as pips is useless unless you include the risk to reward, or at least the profit as an R ratio
Could be a matter of opinion, maybe you scale every trade to be a set amount of dollars per pip
Or report it as percentage of account, eg. risk 1%, made 1% profit (pips could be 20 or 50, but dollar value is the same)
3. The profit target could be improved, as the stop loss is known, it should be at least 1 to 1 Risk to Reward
See this example for this week actually, by targeting last week's low, and the stop above recent high
This would be reported as a profit of 1 R, even though 50 pips may sound like a lot more money was made
Good luck, and let see how this performs going forward, surely some astute programmer can run a backtest (not my expertise)
PS: on edit, chart shows 14 period ATR of 4H x 1.5 is 43, while using 30 gives 45 so not a huge difference
It takes a lot for me to IGNORE you, but you can try.
- #23
- Oct 4, 2017 2:49am Oct 4, 2017 2:49am
- Joined Jul 2011 | Status: Gold Again | 10,569 Posts
Actually the USDCHF, as we all know being opposite of the EURUSD, gives an interesting look at Risk to Reward settings and target
With an entry of 0.9714 and a stop at the swing low of 0.9670 - this is 44 pips, close to the ATR(30) on 4H
Targeting the previous week's high, 0.9770, for 56 pips, nice Risk to Reward at 1.3
I can see where this method could work both in ranging and trending markets
Thanks for sharing @ cocolizo67
With an entry of 0.9714 and a stop at the swing low of 0.9670 - this is 44 pips, close to the ATR(30) on 4H
Targeting the previous week's high, 0.9770, for 56 pips, nice Risk to Reward at 1.3
I can see where this method could work both in ranging and trending markets
Thanks for sharing @ cocolizo67
It takes a lot for me to IGNORE you, but you can try.
I am interested have to do some back testing, why only one pair.
I am using Eurusd, because is the most traded pair and so far is showing good results. Feel free to experiment other pairs and post back results
Cheers
{quote} Very interesting, but a couple of comments 1. How do you calculate the ATR of a 4 hour chart using 14 days? I assume you mean the ATR of 4 Hour chart using 14 x 4 Hour periods, nothing to do with days So why 14? there are 30 x 4 Hour bars in a week, it makes more sense to average the data of the past week right? 2. Reporting profits as pips is useless unless you include the risk to reward, or at least the profit as an R ratio Could be a matter of opinion, maybe you scale every trade to be a set amount of dollars per pip Or report it as percentage...
Answering your questions:
1) I made a mistake, actually I meant a simple 14 period 4h ATR. I am used to work with daily charts thats why the mistake. Just use a standar 4h ATR. I edited the first post to clarify it. Thanks for pointing it out.
2) Yes. Its just a very basic backtest, only to show the potential of the system. Once we have an EA, we can make a more detailed backtest.
3) I don´t like to have fixed amount of target profits. Bigger R/R doesn´t mean more profits, because you have to consider that your win/loss ratio will go down. In the end, is a personal preference of every trader how they wanna handle the trade. In my case I like to take quick profits before the market takes it back. But if you see a good R/R reward ratio in a particular trade, by all means go for it
I just want to share some thoughts on this...
Firstly, the result for the week 4 in what the originator posted isn't accurate per the rules: before the exit bar indicated there is a bar with close also below entry point, so it qualifies as exit per the rules. This gives only about 10 pips on that trade.
Secondly, I fastly crunched my numbers on the last 5 weeks on EU ( I will indicate the week number as unique identifier of the week ):
_____SL___DD___TP___myTP___trailingSL
w36: 60___37____35___90______10
w37: 66___32____30___23______11
w38: 56____0____15____-7______9
w39: 51___29____10___58_______9
w40: 44____8____27___57_______7
My conclusions until now:
- max DD under min SL - OK
- avg R/R ratio about 0,43 => about 71% needed W/L - WARNING ( close to ERROR
)
- now W/L is 100% so bigger than needed W/L - OK, but ... to be more crunched and tested
Concerning the last two columns: I searched if a trailing stop is better in this case. I applied two rules for trailing:
- start trailing only after the first bar which closes above/below the entry point, in the favor of the trade ( for long/short );
- trail only the bars with the same color as the trade.
I took as trailing value SL/6 pips below the low/above the high for long/short ( last column ).
Results are better: even if the losing trades appeared, R/R ratio grows at about 0,79 ( still under 1
), and needed W/L is now about 56%.
Firstly, the result for the week 4 in what the originator posted isn't accurate per the rules: before the exit bar indicated there is a bar with close also below entry point, so it qualifies as exit per the rules. This gives only about 10 pips on that trade.
Secondly, I fastly crunched my numbers on the last 5 weeks on EU ( I will indicate the week number as unique identifier of the week ):
_____SL___DD___TP___myTP___trailingSL
w36: 60___37____35___90______10
w37: 66___32____30___23______11
w38: 56____0____15____-7______9
w39: 51___29____10___58_______9
w40: 44____8____27___57_______7
My conclusions until now:
- max DD under min SL - OK
- avg R/R ratio about 0,43 => about 71% needed W/L - WARNING ( close to ERROR
)- now W/L is 100% so bigger than needed W/L - OK, but ... to be more crunched and tested
Concerning the last two columns: I searched if a trailing stop is better in this case. I applied two rules for trailing:
- start trailing only after the first bar which closes above/below the entry point, in the favor of the trade ( for long/short );
- trail only the bars with the same color as the trade.
I took as trailing value SL/6 pips below the low/above the high for long/short ( last column ).
Results are better: even if the losing trades appeared, R/R ratio grows at about 0,79 ( still under 1
), and needed W/L is now about 56%.
- #37
- Oct 5, 2017 4:55am Oct 5, 2017 4:55am
- Joined Jul 2011 | Status: Gold Again | 10,569 Posts
I still do not get it this ATR thing, sorry!!!!! Abybody plz throw some light??? Stop loss: 1.5 times the 4h ATR (14 period)
I think they teach it in babypips school over at website of that same name
but here is an indicator you can play with, this one is great because you can check weekly, daily, 4H etc.
cheers
It takes a lot for me to IGNORE you, but you can try.
{quote} It's an Average True Range of whatever time frame you desire, and for the number of periods eg. 14 I think they teach it in babypips school over at website of that same name but here is an indicator you can play with, this one is great because you can check weekly, daily, 4H etc. cheers {image} {file}
