hi, o990I6mh:
thank you for your extended explanation of this system. Now much clearer. Just one more glitch:
I’ve already tried all your zip folders on this thread, but I can’t get the template to show MTF moving average indicator on my chart no matter what. Strangely other indicators are working fine. Can you take a look for me? By the way, I’m using Hotforex Mt4.
Oh, one more thing, when we are in a trade and we want to trail the stop, do you mean we just set our new stop under the lower Donchian channel next time it gets ‘flat’?
Thanks for your explanation of your system. I shall follow it with interest. I am also having a similar problem with the MTF Moving Average, it won’t load onto the chart so I have put a 60 SMA on the Daily and Weekly charts and will look for entries on the H4. I like the logic of this system. Thank you again.
Thanks for reply.
If I had the same indicators on Forex Tester 2, I would of liked to back test this as its 100% mechanical. Unfortunately I don’t so I can’t, but I used to trade a pure price action strategy very similar which I will back test in the near future as I need a profitable trend strategy as part of my trading portfolio. This will be my next project after I finish the Slingshot back testing. This takes a long time to complete for conclusive results.
Well keep up the good work as your my kind of trader (mechanical)
Frankly I have no idea why the MTF indicator won’t work for you, but I have made an effort and replaced it with this one: MTF_MovingAverage - MQL4 Code Base
I’m attaching this alternative MTF indicator here along with a new template that uses this indicator instead.
The more mechanical the better, it leaves less room for human error. The only part that isn’t 100% mechanical is the inital stop loss. Of course it’s possible to mechanize this as well but so far I find that I’m able to keep it tighter when I place it by hand.
hi, thank you for your effort. Still one problem:
When I put all your indicators of your new zip file in place and apply the template, it seems there are 2 moving averages on my trading timeframe. Is it supposed to be only 1 (as seen from your screenshot)?
Now it is becoming a little clearer to me. Since you mentioned initial stop loss, can you maybe explain your stop loss placement when your start a trade and are in a trade with a chart if you don’t mind?
On Capt Currency’s 3 Ducks thread (which I really enjoy and which is how I found this one), you agreed with the fact you need to build a strategy that fits you, not the other way around. You also mentioned that as a part-time trader you prefer (or have to) trade on the longer time frames and I am in the same situation. Since you have been trading this system for a while, could you give a brief explanation of how you incorporate your method into you daily routine? Being on west coast of US, I either have to stay up late for Eur open or get up early for NY open but then have to work rest of day. Just trying to figure out the best way to try and implement your system.
Also, you mentioned somewhere else about the detailed records you keep. I don’t need to know the specifics, but was wondering what info you found to be the most beneficial to learn from.
As you can tell these are total noob questions so tell me to take it somewhere else if need be.
Finding or building a system that fits your daily life is key. What I do is I use a separate stochastic indicator on the H4 charts. This indicator alerts me via MT4 on my android phone if stoch H4 goes above 80 or below 20. I keep a short list of the pairs where the higher timeframes are aligned. When my phone buzzes I check and if it’s one of the pairs where the higher timeframes allow trading then I connect to my home computer through TeamViewer from my work computer and place the pending order.
Then I have some scheduled periods on days I start later or finish early, I call them intraday sessions and for those periods I sit down to watch the screen for up to 6 hours or so, looking for trades on the 5min timeframe on low spread instruments.
The biggest takeaway from both my records and the backtesting I do is that letting winners run is of huge importance. It’s easier said than done though. If there was one improvement I’d like to be able to make it would be to allow profitable trades to run longer.
If I have understood this correctly it would be quite likely that there will often be several pending orders open at any time. Have I got this correct?
099016mh, not trying to get in your way, but I see you are offline and answering Dig will probably help me clarify for myself. If you get in later and see anything wrong, please correct me.
Dig, I didn’t get anything from the initial instructions that would suggest leaving multiple orders on the table. Here is my mental picture. When prices are trading above our higher timeframe indicators, we are looking for buys. On a downturn, which we hope is a retracement, we watch for stochastics to go into oversold, i.e., below 20. When that happens, we will note that the lower donchian channel is close to our current price, but the upper channel is some distance away. That upper donchian channel, plus a few pips, becomes our long entry point. If the market turns back up, we will observe prices moving ever closer to the upper donc channel line and, perhaps, the donc line moving closer to price. As they converge, we don’t add orders, we simply move our one existing order just above the nearest point of the upper donc line.
There is a good example this morning on the gpb/jpy h1 chart about 19:00 gmt. Our higher tf indicators are above price, so we are looking for buys. In a 4-hour period, prices made about a 100 pip down move that would, indeed, prove to be a retracement as prices fell from 169.79 to 168.81. When price hits that low, we also see stochs fall below 20. At this point, the lower donc is at 168.81, right with price, while the upper donc is 169.79. Theoretically, that would be our initial long entry price, “but wait, there’s more!” Prices start moving up and the upper donc starts moving down. As they converge, we move our long entry stop from 169.79 to 169.69 to 169.53 to 169.29. That last number occurred at 3:00 and would create a long entry about 169.32 or so. The 4:00 bar hits a high of 169.45, which would take us into the trade.
From there, it is all about trade management, but prices continued to rise to a high of 170.25 on the 8:00 bar. As I write this, it is 13:00 gmt, 7am central time, and the pair is trading around 170. We may still be in the trade, looking for a resumption of the uptrend. Or, we may have taken a chunk of that 90 pip move. In any case, this one is fun to trade and I appreciate the author taking time to share it with us.
Dig, I did miss something on that trade I was describing. When I first started watching the trade develop, prices were above the higher tf ma’s. However, by the time we would have entered, prices had moved below those ma’s. According to 09, the trade is made “…while making sure that the potential entry point is located above both higher timeframe SMAs.” Though it would have been a successful trade, it would not have been in keeping with the system rules. My bad.
Just to be sure I don’t confuse anyone, long trades are made only when price is above the ma’s. Shorts are entered only when price is below the ma’s.
I’m not intending to be critical of o99, and I thank him for taking the time to write up his experiences.
I’m curious if anyone has had a similar experience to me using the TrendLover system in backtest. I’ve tested it on a couple of pairs on the H1 timeframe, ie my higher tfs were H4 and D1.
I initially couldn’t get the MTF average to work so I simply added a SMA240 and a SMA1440 to my H1 chart, to approximate the SMA60 on H4 and D1.
My results were quite frustrating because I missed many, what appeared to be, potentially viable entries due to exclusion by one or other of the entry criteria. In other words, the opposing direction of the Stochastics rule and the MAs rule made it hard for any setup to satisfy both rules for an entry. By the time a setup had pulled back far enough to satisfy the Stoch criterion, it had crossed (or was about to cross) one of the higher-tf MAs which then ruled out an entry. On the rare occasions when a setup did qualify for entry, the trend was near exhaustion and the position usually got stopped out at the Initial Stop-Loss. Only about 25% of entries were profitable. Even though some profitable entries had a good run, they could not compensate for the large number of losses, so my expectancy was negative overall.
This reminds me of the experience I had with testing the 3 Ducks. By the time I had satisfied the higher timeframe criteria, the lower tf trend had turned and was heading in the opposite direction.
Does anyone have any consistent, positive experience with other timeframes?
Hi, The equivalent of a SMA60 on a daily TF is: 1440 x 60 (1440 minutes in one day times 60 periods), and for the 60SMA on the 4H chart you’d need 240 x 60. So the MAs you’d need to put on the 1H chart is not 240 and 1440 but rather 14400 and 86400.
The numbers you’ve been using correspond to a 1 period SMA of both 4H and Daily. I believe that is the explanation for you finding trouble with profitability of the backtesting.
hi, o990I6mh:
I also had difficulty using the MTF moving Average indicator so I’ll have to plot them manually. But can you explain to me again in detail how you go about plotting higher timeframe moving averages on lower timeframes?
From the explanation above it seems that you are counting minutes? But why? What if I want to plot a 60SMA of monthly chart on the weekly? What is the calculation process?
THX.
Shuo
