Re-igniting Dancingphil's ZigZag strategy | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Simple Zigzag + Fibonacci Strategy 747 replies

Zigzag indicator with pip count? 9 replies

Creating an EA using a zigzag strategy 23 replies

Alert for zigzag? 1 reply

Attachments: Re-igniting Dancingphil's ZigZag strategy
Exit Attachments

Re-igniting Dancingphil's ZigZag strategy

  • #41
  • Apr 9, 2024 7:23am Apr 9, 2024 7:23am
From your stats here's where my attention goes:

Inserted Code
winners=290                     (28.5%)
winners.long=161                (55.5%)
winners.short=129               (44.5%)
winners.avgRunup=27.1
winners.avgDrawdown=-4.1
winners.avgProfit=13.7
 
losers=723                      (71.2%)
losers.long=347                 (48.0%)
losers.short=376                (52.0%)
losers.avgRunup=5.0
losers.avgDrawdown=-8.4
losers.avgProfit=-8.4

So basically when you open a trade:

  1. If it was a winner you would measure on average a drawdown of 4.1
  2. If it was a loser you would measure on average a drawdown of 8.4

But the most interesting thing: losers on average runup 5 pips. So here's come my question, could you post statistics for a strategy that:

  1. Set buy limit 5 pips above donchian upper channel / sell limit 5 pips below donchian lower channel
  2. Trailing stop 5 (4.1 is more precise but I'm accoutning for some deviation) pips (hope you see the reason behind the use a trailing stop: it's like using new donchian breaks as reference point for calculating if the new break will be a loss or win. Sorry if that does not makes sense, I can try to explain better).
  3. Obviously, never 2 trades opened at the same time.

You created a strategy for the sake of collecting statistics regarding zig-zag, this suggestion of mine is a strategy that uses those statistics and take/manage trades in logic like: "new donchian break: will it be a winner or loser? If price rise 5 pips witout decreasing more than 4 pips, then likely a winner with a potential win of 27 pips. If price rise more than 5 pips I open trade. Trade opened 10 mintues ago (example) price is consolidating but it did not went more than 5 pips on the opposite side. Now price creates a new donchian break, I'm still on the trade but it's like I'm inside a new signal/trade so, again, if trade goes 5 pips opposite direction from new donchian then most likely I'm inside a loser trade, so rise/lower the stop 5 pips below/above donchian max/min."

I know I typed two times the same thing, I'm a bit in a hurry + eniglish isn't my main language. Just wanted to make things clear

Let me know if you find this interesting enough to code it.

... on a side note, why you're still on windows 98 ?

  • #42
  • Edited 5:49am Apr 10, 2024 5:33am | Edited 5:49am
Regarding the nested ZigZag setup: It's important to trade only the first 1-2-3 trade of the inner ZigZag. The win rate of the first 1-2-3 pattern is significantly higher than the rate of consecutive patterns. In other words: if you trade a consecutive pattern you essentially trade the same signals as OptoJay did with his 1-2-3 EA, with the known results. You may have luck or you may not.

Calculate your lotsize accordingly so size of the breakout pattern doesn't matter: same PL no matter the size of the pattern.
Attached Image (click to enlarge)
Click to Enlarge Name: XAUUSD,M1.png Size: 32 KB
It takes blood, sweat and enough tears to wash away the blood.
  • #43
  • Apr 10, 2024 5:45am Apr 10, 2024 5:45am
Quoting oculus
...Let me know if you find this interesting enough to code it.
Currently I'm working on something very similar to your proposal.

My intention for the ZigZag EA was a bit different. I was aware that an "always in the market" setup can't be profitable. But I can see an interesting pattern: Whenever ZigZag produces some losers in a row, it's only a question of time until the next breakout extends exactly to the size of the formerly made losses, and maybe a little further. So I tried to find out whether waiting for 3 losers and trading the inevitable breakout with a clear target (the before lost pips) makes sense. And indeed, it can be done. However, automation is difficult. For automation in general it holds: the more simple the better/more robust.

Quoting oculus
... on a side note, why you're still on windows 98 ?
I'm on Window 7 (local) and Windows Server 2008 R2 (remote VM) and it will stay like that as long as possible. Even this post I'm writing from a virtual machine on Win7
It takes blood, sweat and enough tears to wash away the blood.
  • #44
  • Apr 10, 2024 9:18am Apr 10, 2024 9:18am
Quoting emmzett
{quote} My intention for the ZigZag EA was a bit different. I was aware that an "always in the market" setup can't be profitable. But I can see an interesting pattern: Whenever ZigZag produces some losers in a row, it's only a question of time until the next breakout extends exactly to the size of the formerly made losses, and maybe a little further. So I tried to find out whether waiting for 3 losers and trading the inevitable breakout with a clear target (the before lost pips) makes sense. And indeed, it can be done. However, automation is difficult....
I've tried the logic of excluding signal based on ea performance but in the long run everything does not work. The point is that: if you take random samples from a pseudo-normal distribution, you still get a pseudo-normal distribution; i.e. just from a purely theoretical standpoint taking trades after 2-3 losses wont make a unprofitable strategy into a profitable one because the underlying signal is unprofitable.

Let me know what you think. Have a great day

Regards
  • #45
  • Edited 12:56pm Apr 10, 2024 10:33am | Edited 12:56pm
Hi guys,
we can probably use also sentiment. Because most people are always wrong. Check myfxbook sentiment for experiment. When we are low (I mean oversold) on chart and people start selling you can go buy, vice versa. However if we are overbought, and people start selling, we can go with them. It can help a lot probably. In this particular example price will start falling and people will change positions to buy. We'll however stick to selling.

It can be used as filter. Probably can be automated or added into the EA but sentiment on myfxbook is paid, so. The one on the web is updated once per hour. Might miss a lot of moves. And you'll miss it. Trust me. Once tried paid vs free and it was like Volkswagen vs Mercedes.

Check red arrow. They teach us we should go against majority all the time. But it's nonsense. Majority is sometimes right with their opinion. And this happens when trend is changing.
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_2024-04-10-17-24-06-937_com.android.chrome-edit.jpg Size: 146 KB


When this starts happening we will check what majority is doing
When they start changing opinions, we'll stick with positions we opened before.
We will enjoy easy profit.

One can probably build easily equity with this. I wonder how is it possible no one noticed this before. Why I am even writing it...

With Enjoy. Hope it's helpful for someone.

P.S I will not explain how I mean we are oversold or overbought. That would be too easy. Picture should tell you more then enough. No spoon-feeding here. Who will want learn it, will do it. I have trust in you, friends. Check my ideas if you want to know how I make resistance or support. And as always, don't forget to look at "50%". It can be literally seen on a chart. (Market Profile can help understand this)

Also little help needed. Whoever works with TradingView, and can write me an easy indicator (work for 5 minutes) let me know please in PM. I have MT4 code and can't do it on phone. Sadly I lost my PC and I'm unable to do anything to convert it on a phone because TradingView just can't code on a damn phone..

One more thing. EUR/USD looks by looking at sentiment like easy buy. Just look if sell positions starts going up. Then it's literally confirmed. Even better. Just checked 15m. Buy positions appears. So I opened too, just to prove my point.
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_2024-04-10-17-59-55-859_com.tradingview.tradingviewapp.jpg Size: 244 KB


Trade before also by sentiment =
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_2024-04-10-15-53-34-592_com.tradingview.tradingviewapp.jpg Size: 221 KB


How is it going...
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_2024-04-10-19-54-50-211_com.android.chrome-edit.jpg Size: 140 KB

Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot_2024-04-10-19-55-18-348_com.tradingview.tradingviewapp.jpg Size: 231 KB
  • #46
  • Nov 26, 2024 7:28pm Nov 26, 2024 7:28pm
does anyone know of any dancing phil students that are contactable?
  • #47
  • Sep 14, 2025 3:09pm Sep 14, 2025 3:09pm
Quoting anon007
does anyone know of any dancing phil students that are contactable?
Searching into Phil's messages I found one, but he's not replying

I would add that I never quitted reverse engineering his strategy, but I am stuck at hedging (far from the 1 bad case every 9)
  • #48
  • Sep 14, 2025 4:32pm Sep 14, 2025 4:32pm
Quoting engienmix
{quote} Searching into Phil's messages I found one, but he's not replying I would add that I never quitted reverse engineering his strategy, but I am stuck at hedging (far from the 1 bad case every 9)

Imho, the key to successful ZigZag trading is automation. While this is not Phil's original approach, it's in the nature of ZigZag that it must fluctuate between profit and loss. There is no way it can escape from that rule. Escaping would mean a market either becomes permanently flat (not possible because immediately exploited) or a market starts swinging wild permanently (again, not possible because immediately exploited).

Trending markets are better for ZigZag which makes crypto a good choice for it. Look at the screenshot of a ZigZag(30xM5). Whenever there's a green number ZigZag is in plus. As you can see it takes not more than a day for ZigZag to become profitable. There is no way it can change that behavior. It may take a bit longer but it must return to profit.

The real challenge is that it's hard to impossible to trade such a setup manually. Manual trading with retail expectations (higher than average profits) is always at risk of sudden drawdowns/crashes. Very hard to overcome. Having performance goals and risk management in a single head is against human nature. There is a reason banks strictly separate these two concerns. For retail: automation is the answer.
Attached Image (click to enlarge)
Click to Enlarge Name: BTCUSD,M5.png Size: 50 KB
It takes blood, sweat and enough tears to wash away the blood.
  • #49
  • Sep 14, 2025 5:28pm Sep 14, 2025 5:28pm
Quoting emmzett
{quote} Imho, the key to successful ZigZag trading is automation. While this is not Phil's original approach, it's in the nature of ZigZag that it must fluctuate between profit and loss. There is no way it can escape from that rule.
It's pure mean reversion, measuring how much positive/negative variance you have in your equity performance, right?
  • #50
  • Sep 14, 2025 11:05pm Sep 14, 2025 11:05pm
Quoting engienmix
{quote} It's pure mean reversion, measuring how much positive/negative variance you have in your equity performance, right?
Correct. I call it ZigZag balance. Here an example of ZigZag(18xM5) balance in S&P500 over a few months of 2022 (scale in index points, the chosen period doesn't mean anything). It will take at most 3-4 days to come back to the mean/balance. Most important: it cannot run away.
Attached Image (click to enlarge)
Click to Enlarge Name: us500,zigzag(18xm5).png Size: 39 KB
It takes blood, sweat and enough tears to wash away the blood.
  • #51
  • Edited 5:35am Sep 15, 2025 3:01am | Edited 5:35am
Quoting emmzett
{quote} Correct. I call it ZigZag balance. Here an example of ZigZag(18xM5) balance in S&P500 over a few months of 2022 (scale in index points, the chosen period doesn't mean anything). It will take at most 3-4 days to come back to the mean/balance. Most important: it cannot run away. {image}
Have you tested this behavior for a long time?
In my tests, I have found that this behavior always balances out over the long term, in rare cases so long that no trader could sustain it.

But I think you went deeper in momentum hunting, with the 2nd indicator (on your previous chart, the 40/45 if I remember correctly), you understand when the trend is about to take off, but not its direction. On indices, you might consider trading with straddle options.
  • #52
  • Sep 15, 2025 11:53pm Sep 15, 2025 11:53pm
Quoting engienmix
{quote} Have you tested this behavior for a long time? In my tests, I have found that this behavior always balances out over the long term, in rare cases so long that no trader could sustain it. But I think you went deeper in momentum hunting, with the 2nd indicator (on your previous chart, the 40/45 if I remember correctly), you understand when the trend is about to take off, but not its direction. On indices, you might consider trading with straddle options.
The theory works on any instrument, any timeframe. Instruments with real underlyings work better than e.g. Forex with its erratic swings. I look for confluence with natural cycles which brings me to classic auction theory and daily intraday swings (London Open, US Open, US Close). Here is the most potential but it's also difficult to trade correctly. I monitor the ZigZag balance and look to enter when balance is under water for some time and one can expect a shift in volatility (market open, news etc). The Donchian channel width helps me to notice possible entries more easy (difficult with ZigZag alone). However, that's manual and in my experience manual guessing isn't really consistent (at least for me). I find it very hard to trade correctly in a manual way.

Quoting engienmix
...in rare cases so long that no trader could sustain it...
Imho, even with mean reversion (no matter the strategy) one must never try to go against the trend. In your mentioned scenario the trader ignored the trend and tried to know better. Sooner or later it will kill you.
It takes blood, sweat and enough tears to wash away the blood.
  • #53
  • Sep 16, 2025 12:07am Sep 16, 2025 12:07am
Another observation: Even while every first view on a chart tells you the opposite, the real risk is never to the downside but to the upside. That's because fat tail events to the upside are much more violent and can last much longer than periods of stagnation. It's more safe to only trade from stagnation to volatility, not the other way around.
It takes blood, sweat and enough tears to wash away the blood.
  • #54
  • Sep 28, 2025 4:50am Sep 28, 2025 4:50am
Quoting engienmix
{quote} Searching into Phil's messages I found one, but he's not replying I would add that I never quitted reverse engineering his strategy, but I am stuck at hedging (far from the 1 bad case every 9)
i think that much of his strategy likely disappeared with him... i never attended his room but it was clear from the little contact i had that he wasn't only relying on his fractal indicators... also not sure why he stopped trading the strategy e.g. how much of a fundamental structural component it relied on (past practices of rate fixing, euro debt crisis etc)

in theory, hedging is more of a psychological prop, as if you have set point where you release the hedge, there is no different to quit the position where you applied the hedge and reenter at the release point... i think his main point was that you can leave a troublesome position hedged with zero margin utilization, whereas automation would deal with it as i described
  • #55
  • Oct 2, 2025 3:59am Oct 2, 2025 3:59am
Quoting anon007
{quote}i think his main point was that you can leave a troublesome position hedged with zero margin utilization, whereas automation would deal with it as i described
Eliminating the margin was actually the purpose of the hedge, to give a future possibility of closing both legs in profit/breakeven.
  • #56
  • Oct 2, 2025 4:04am Oct 2, 2025 4:04am
Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 16 KB


It is clear that at the birth of every new trend there are almost always more favorable entry points to scalp, as Phil stated
  • #57
  • Edited 3:23pm Nov 29, 2025 2:03am | Edited 3:23pm
A post in another thread where I explain my setup: BTC Scalper Hub
It takes blood, sweat and enough tears to wash away the blood.
  • #58
  • Dec 26, 2025 3:41pm Dec 26, 2025 3:41pm
Quoting emmzett
A post in another thread where I explain my setup: BTC Scalper Hub
you did well to share, thanks
  • #59
  • Dec 27, 2025 6:08am Dec 27, 2025 6:08am
Quoting emmzett
A post in another thread where I explain my setup: BTC Scalper Hub
have you tried to overlay a bollinger band on the ZZ blue indicator?
  • #60
  • Dec 28, 2025 3:37am Dec 28, 2025 3:37am
Quoting engienmix
{quote} have you tried to overlay a bollinger band on the ZZ blue indicator?
No, I didn't and for following reason: BollingerBand is a lagging indicator and I prefer to use only realtime indicators. You can see that the Donchian Channel always signals before the breakout (that's a fundamental principle of this signal). A BollingerBand signal would many times come too late. Again, that's a principle of a lagging indicator.
It takes blood, sweat and enough tears to wash away the blood.
0 traders viewing now