Helloooo! 
I’ve been trying to work on updating my system but I don’t really want to change it completely.
It’s basically a golden cross system but I’ve been thinking of adding at least 1 more indicator to improve accuracy and help me spot better entries. 
Is there a particular indicator that works best with moving averages for this?
3 Likes
You might want to try an indicator that isn’t based on or derived from moving averages (there are plenty!), so that you’re adding something new and different?
Maybe something based on support and resistance like a Donchian channel? Since more “apparent breakouts” are fakeouts than the real thing, that could be a good way to try?
How fast/slow are your moving averages?
3 Likes
I’m doing the classic 200 and 50. It’s very basic, so I’ve been thinking of how I can make it work better for me. 
This is my first time hearing about Donchian channel. I’ll look it up! Thank you sooo much for the suggestion Truncated! 
1 Like
Good luck and congratulations on your wedding, which I just read about!
Let me know if you have questions.
MA-50 above/below MA-200 is great for judging trend direction.
You might try a much faster D-channel for the entries?
Example: when MA-50 is above MA-200 and both are rising, look only for long entries. When MA-50 is below MA-200 and both are falling, look only for short entries. Long entry is when a bar opens below the lower Donchian line and closes above it (resuming the uptrend after a retracement). Short entry is when a bar opens above the upper Donchian line and closes below it (resuming the downtrend after a retracement). Might work, with a setting of maybe 10 or 12 for the Donchian???
You could possibly even try it using the Donchian midline (maybe setting of 15???) and not bother with the Donchian upper/lower lines at all??
Only ideas, don’t know If it works for forex!
2 Likes
To be honest, I would not have thought another indicator was necessary. When I see a consistent trend, based on swing highs and lows being repeatedly exceeded, and with the two MA’s in the right sequence, any weakness in the trend is a signal for me to enter - but not live, I look to set an entry order at a level beyond the weakness, between the pull-back and the trend as it were.
4 Likes
Hi, maybe KAMA? Kaufman Adaptive Moving Average
Unfortunately, no. Derivatives of historical price cannot predict future price changes any better than the expected value of the trade. By adding filters, you’ll lower the sample size but not refine the quality of the sample.
I suggest before moving further, backtesting the strategy itself 100x manually to prove it’s efficiency. Note every loss and win, and at the end of the test, sum it up and compare the loss to win ratio. Once you attain this statistical data, you’ll then be able to ascertain if it’s even worth trading this strategy or not.
You can always use more data
1 Like
Lol, 100 times is not “statistical data”: 5,000 times might be, under some circumstances.
1 Like
Hmm, lets view the definition:
statistic
- A numerical piece of information.
- A calculated numerical value (such as the sample mean) that characterizes some aspect of a sample set of data, and that is often meant to estimate the true value of a corresponding parameter (such as the population mean) in an underlying population.
Right: So when you backtest a strategy, you’re calculating the numerical data. (Wins/Losses) to estimate the true value of 100 test trades e.g (60 wins vs 40 loss) etc. Sounds about right to me, but if you wanna be anal, be my guess. The amount or quantity tested doesn’t constitute whether or not the actual practice is defined as statistical data.
1 Like
If the strategy is easily quantifiable, we can run backtests with and without an additional filter to deliver more metrics.
1 Like
I don’t think you need another indicator… but if you really want to add one you could add RSI or ADX (if you trade trends, which I guess since MAs work during trends) so you could validate your golden cross if there’s a Trend (ADX) or determine the strenght of your golden cross (RSI). if you are someone who trades tops and bottos then might Stochastik would be the better accessory
1 Like
For me, I think you shouldn’t have a need looking for an extra indicator, but if you really want another, then maybe MACD or RSI would be a nice combo.
1 Like
Those are both based on MAs. As a couple of people rightly explained above, if you want to add an indicator to Mas, add one that isn’t.
Otherwise it isn’t a “filter”: if your MAs signal an entry that loses, your extra MA-derived indicator probably will, too. It’s very unlikely to be helpful.
Entries don’t matter very much, anyway: trade management is far, far, far more important, in the long run.
3 Likes
Since MAs and most other indicators are based on price and time, perhaps consider a volume indicator.
Even though a given volume for a given broker is not necessarily indicative of total market volume, relative volume (from one price bar to the next) from your broker should be sufficient to use.
It could be as simple as a MA of volume and you’re looking for a jump in volume that x% above the volume MA to confirm strength in the direction indicated by the 50/200 price MAs.
They don’t have to be based on time (mine aren’t, because I don’t use timed bars). It obviously depends how your charts are constructed.
Volume indicators are wonderful, but not for trading spot forex or CFD’s, of course, because those products aren’t in a centralized market, and there’s therefore no way of knowing what the “volume” is.
Counterparties pretending to be “brokers” can only give you their own volume because (just like you, and everyone else) they obviously have no other figures available to give you. You surely wouldn’t want to trade from that “information”? At least, I hope you wouldn’t!
Right. I said most, not all. I suppose you’re using range bars or Renko bars? Is that not price-based?
Correct. Hence, my comment about relative volume to the broker. If you use a large, well-established broker (Oanda, for example), then the volume provided has a high degree of confidence of representing the market.
You have been snarky/critical of several of the posts in the thread. Do you have anything worthwhile to contribute to the OP’s question?
2 Likes
You actually said “MAs and most other indicators are based on price and time,” which really isn’t quite right, if you’ll excuse my mentioning it.
Even MAs don’t have to be based on time, as I pointed out.
I used them on and off for over a decade without that ever being exactly my experience, though I did check, but I found that futures volumes (available very-nearly-free from CME if you’re a non-professional) serve that purpose very well.
Sorry if I came across that way: I was just anxious to make it clear to @ria_rose that not everything stated in the thread is accurate, and I was perhaps inadequately attentive to the feelings of other members in doing so.
I actually felt that correcting some misinformation was, in itself, worthwhile. I’m guessing you disagree - which is your right, of course.
Actually neither, but let’s not make the thread about you and me? @ria_rose is asking for help, here.
Try MFI Divergences. I played around with it enough to observe some pretty interesting correlations. No back testing on my part as of yet but traders I’ve spoken with attest to it. DYOR as always.