I wonder.. will this strategy work? | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

One hit wonder or am I on to something? 81 replies

One Candle Wonder 19 replies

wonder if u guys played those $0.005 penny stocks 18 replies

Ever wonder why your Bokers charts use different data?? 13 replies

Attachments: I wonder.. will this strategy work?
Exit Attachments

I wonder.. will this strategy work?

  • #1
  • First Post: Edited Jan 11, 2012 12:54am Jul 24, 2011 2:54am | Edited Jan 11, 2012 12:54am
I am really thinking of this strategy from many days... i trading it manually and its working fine.

But i really wonder will this work by autotrading...

It like this... [for this I use ea which closes all open trades when hitting target= $10]

1) Open pending BUY order with lot size 0.1 at somewhere around current price.

2) If pending BUY order become active , open pending SELL order with lot size 0.2 at same prize, and wait till hitting basket take profit target (ie $10).

3) Again If price don and pending SELL order become active , open pending BUY order with lot size 0.3 at same prize, and wait till hitting basket take profit target (ie $10).

ie we open buy/sell orders alternately at same price with incremental lot size (Not martingle). This is almost no loss strategy.

Now after step 3, if price go up ... we have 1+3=4 buy orders and 2 sell orders, so we are in profit.
But Suppose after step 3, if price go down we have to open SELL pending order at same price with lot size 0.4. Now we have 2+4=6 sell orders and 1+3=4 buy orders, again we are in profit.
  • #2
  • Jul 24, 2011 3:23am Jul 24, 2011 3:23am
Any strategy, whatever it is, only works if it can capture price movement from A to B

  • #3
  • Jul 24, 2011 3:33am Jul 24, 2011 3:33am
Quoting Password
I am really thinking of this strategy from many days... i trading it manually and its working fine.

But i really wonder will this work by autotrading...

It like this... [for this I use ea which closes all open trades when hitting target= $10]

1) Open pending BUY order with lot size 0.1 at somewhere around current price.

2) If pending BUY order become active , open pending SELL order with lot size 0.2 at same prize, and wait till hitting basket take profit target (ie $10).

3) Again If price don and pending SELL order become active...
if only you can find broker with no spread
  • #4
  • Jul 24, 2011 3:36am Jul 24, 2011 3:36am
Sounds really interesting. I can already see that this will not work during choppy markets. Could you post some charts and results?
Also, where do you put you pending BUY/SELL orders? Do you use standard technical analysis for that?
  • #5
  • Jul 24, 2011 6:59am Jul 24, 2011 6:59am
Quoting Password
i trading it manually and its working fine.
But i really wonder will this work by autotrading...
Any mechanical strategy that has a genuine edge will be profitable regardless of whether it's traded manually or automatically.

Quoting Password
ie we open buy/sell orders alternately at same price with incremental lot size (Not martingle). This is almost no loss strategy.
If I've understood you correctly, and at the risk of sounding negative, then my view is that no analysis = no demonstrative edge, and the risk here is whether price moves sideways for long enough to allow your position size to escalate to the point where the drawdown becomes unacceptably large (the "death trade").

I think you acknowledge this by your use of the word "almost". The system is guaranteed to deliver $10 profit per trade until the death trade occurs. Only you can decide whether or not you're willing to take that kind of risk.
  • #6
  • Edited 8:02am Jul 24, 2011 7:41am | Edited 8:02am
This is an interesting idea and something I have considered before.

I think the issue here is ranging... ie, you're ideally going to want your TP level to be inside the ATR because if you're having to increase your lot size every time your method fails to make the TP level (and for example moves down so you open another sell with an increment of +0.1 lot to use your example) you're going to be up for a margin call eventually... so money management is key... as is I suspect chosing the right times to trade this.

Having said that during times of reasonable volume with modest volatlity if you set your TP levels sensibly I think it could probably be a winner.

If I recall on the modelling I did the number of 'loops' inside a TP range of around 10 pips was about 10 cycles, which with your lot scaling can add up quickly!

If an up + down is one loop, then 10 looks is:

1+2+3+4+5+6+7+8+9+10+11+12+13+14+15+16+17+18+19+20= 210

or 100 Long (1+3+5+7+9+11+13+15+17+19) and 110 Short (2+4+6+8+10+12+14+16+18+20).

Although I guess this is practically a hedge, so you're just 10 short. The other issues are slippage and spread if you're working on tight TP levels.

Hmm, getting interested again now
  • #7
  • Jul 24, 2011 8:19am Jul 24, 2011 8:19am
Quoting r2997790
........if you're having to increase your lot size every time your method fails to make the TP level (and for example moves down so you open another sell with an increment of +0.1 lot to use your example) you're going to be up for a margin call eventually.......

Having said that during times of reasonable volume with modest volatlity if you set your TP levels sensibly I think it could probably be a winner.
Have I misunderstood you, or does a contradiction exist between the two statements that I've highlighted?

The way I see it, all systems of this type are absolutely guaranteed to return profit (in this case, $10 per trade) because of the 'recovery mechanism' that's built into the escalating lot size.

The gamble is whether the margin call (that you describe) will happen during the system's lifetime. It only has to happen once, and all of the good work - maybe months', even years', accumulation of $10 profits - are all going do be undone, instantaneously. No amount of analysis (ATRs, trading times, etc) can make the system 100% guaranteed bulletproof against this one catastrophic event. An event that can happen unexpectedly, at any time.

I don't believe back-testing of any kind can forecast, without a self-defeatingly high level of statistical error, the likelihood of a 'black swan' event occurring. It's a bit like saying: here's a 10-year-old child, and here's a 50-year-old man, neither of whom has ever been struck by lightning. Based on that history, which one of them has the greater probability of being struck by lightning in the next X days/months/years/whatever?

Hence I don't believe that it's possible for anybody to say categorically whether systems of this kind are likely to be profitable, or not. They will always remain profitable, indeed handsomely so, so long as an event whose probability can't be meaningfully calculated, never happens. Hence it comes down to a personal decision: either one is willing to take that type of risk, or one isn't.
  • #8
  • Jul 24, 2011 8:34am Jul 24, 2011 8:34am
Hi Hannover, yes, perhaps I was hedging my bets with those comments you hightlighted

The point I was trying to make is if you had an over ambitious TP level, or trading in the middle of a dead zone, you might find yourself in an 'endless loop' inside you TP range and hence continually scaling up your lots until you got a margin call and death.

I think the point you raise is 100% correct.

I don't really agree though that there is one catastrophic that will kill you, in fact, nothing will kill you. Literally. No volaility and a consolidation is what kills you, so I think this method should definately not be traded 24 hours a day!

The issue would then become is it possible to set some 'rules' around when you might use this trading strategy.

If we're both agreed margin call is the problem, this would be overcome by trading micro or mini lots, having a large bank, a tighter TP target and trading in time of higher volume and volitality.

I would suggest the death of this method is when your lot sizing is too large, which is caused when you continually fail to TP and get trapped in an endless loop.

With this in mind do you think it is possible to indentify some optimium trading conditions? I think it probably is... based around time of the day and ATR for starters.
  • #9
  • Jul 24, 2011 12:01pm Jul 24, 2011 12:01pm
I've been thinking some more about this idea... the lot scaling is both a blessing and curse. As we've discussed at some point it's going to kill you through margin call... however it might be possible to exit and mitigate some of the risk..

Thinking out loud...

If we focus on cycles (one move up, one move down), after X cycles (let's say 10) you might wish to exit.

You'd be 100 long and 110 short based on my earlier post.

Now, let's say you there was a difference of 2 pips between the long and short entry points (the points where you add an scale up as the market yo-yos, if your TP isnt hit).

After 10 cycles we could shift out TP level in from 10pips to 2 pips.

From the Long Entry point, let's say the market moves down 4 pips (our long trade is now: 0.1 (lot size) x 100 x 4 = -40)

Our short would be 2 pips positive (0.1 x 110 x 2 = +22 )

So we could close out for -18.

During this 'close phase' lot sizing could be disabled to prevent any more margin risk, so then it's just a waiting game for the market to finally move and close out.

Moving in the opposite direction the trade would be:

Long (0.1 x 100 x 2 = +20)
Short (0.1 x 110 x 4 = -44)
Loss -24

This advantage of this is you know exactly what your maximum exposure and margin will be, so then you can concentrate on optimimising entry times and/or making sure you can have enough margin, and hence cycles, to maximise your profitable trades.
  • #10
  • Jul 24, 2011 12:59pm Jul 24, 2011 12:59pm
Password, can you explain how you take a trade in both directons at the same price? It is not possible due to the spread. Can you clarify your method re: entry?
  • #11
  • Jul 24, 2011 1:04pm Jul 24, 2011 1:04pm
Quoting Password
1) Open pending BUY order with lot size 0.1 at somewhere around current price.

2) If pending BUY order become active , open pending SELL order with lot size 0.2 at same prize, and wait till hitting basket take profit target (ie $10).

3) Again If price don and pending SELL order become active , open pending BUY order with lot size 0.3 at same prize, and wait till hitting basket take profit target (ie $10).
Simplified example of what will happen when sh*t happens ;-)

1) your pending buy 1 lot gets activated at price 1000, price then goes to 1008 (up 8 pips)

2) you put pending sell 2 lot at 1000, price goes down to 0998 and activates the sell orders

3) price suddenly jumps from 0998 to 1048, you have no chance to put pending buy 3 lot at 1000 (too close to market price), meanwhile buy (1 lot) +48 pips and sell (2 lots) -96 pips

4) you're so screwed
The tone of your words reflects the quality and result of your trading ;-)
  • #12
  • Jul 24, 2011 3:40pm Jul 24, 2011 3:40pm
I've looked and cant see anyway to exit this trade without wiping out your profit. You're going to need to find an way to exit when you're at your lot scaling limit without moving you into the red.
  • #13
  • Jul 24, 2011 3:49pm Jul 24, 2011 3:49pm
Quoting r2997790
I don't really agree though that there is one catastrophic that will kill you, in fact, nothing will kill you. Literally. No volaility and a consolidation is what kills you, so I think this method should definately not be traded 24 hours a day!
.........
If we're both agreed margin call is the problem, this would be overcome by trading micro or mini lots, having a large bank, a tighter TP target and trading in time of higher volume and volitality.
Thanks for your reply.

The first questions that I ask myself with all systems is "what is the worst possible downside?" and "am I willing to take that risk?".

Without going into more detail (in this case, TP points) I'm not sure how catastrophic the drawdown would be if a margin call ever occurred. As I indicated in my last post, while I agree that the risk of a margin call can be reduced, I don't believe that the possibility can be entirely eliminated. With the markets, anything is potentially possible, including prolonged sideways movement.

The argument with trading micros or minis on a large account is frequently used to justify Martingale, but the inescapable fact is: you cut your position size by a factor of X, then you cut your returns by a factor of X also; or even more so if you're compounding your gains. Obviously the returns have to be acceptable, or the whole exercise becomes self-defeating.

Again at the risk of sounding negative, whenever I see something that seems too good to be true, I also ask myself "if making profit is this easy, why isn't everybody trading this way?". Forex has been around for a good few years now, and there are apparently more than 2 million traders currently using MT4, let alone all of the dozens of other platforms. So whenever I see some interesting "new" idea, I have to wonder why somebody hasn't already tested something that's very similar, and why it isn't more widely traded.

However, I'll definitely continue to follow this thread. Call me a hypocrite who hedges his bets, but I don't want to miss out on an opportunity, just in case I've been asking myself the wrong questions all along. LOL.
  • #14
  • Jul 24, 2011 3:58pm Jul 24, 2011 3:58pm
Good post Hannover I like your approach.

At it's most fundemental with this concept of Password's we're asking how long is it going to take the market to move 10pips + spread, and each time we get a yoyo we're adding positions.

To me, that instantly says dont trade 24hrs, but focus a period between Frankfurt Open and New York Close, or part thereof.

In a trending market this strategy also obviously would work fine.

In a consolidation zone the risks are higher.

Perhaps the TP level could be dynamic, relative to the volality/trend to reduce the number of loops?

This all comes back to minimising the number of loops and also finding a way to exit elegantly without wiping out any gains.

Any ideas?
  • #15
  • Jul 24, 2011 4:22pm Jul 24, 2011 4:22pm
Quoting r2997790
Any ideas?
This thread uses a very similar concept - at least it did at the beginning, until I gave up following it. I don't know where it went after that.

It's another reason why I asked the question about whether an idea has already been discovered elsewhere. To be fair, some of my earlier (skeptical) comments were based on conclusions that I jumped to, because I'd already been influenced by the early posts in that thread.

Anyway, hopefully you'll be able to glean some ideas from the 1,300+ posts there.

I should perhaps also point out that as I'm happy enough with my existing trading strategy, I don't have a huge incentive to spend a lot of time in further research. Having said that, though, I'm still immature enough as a trader to continue to keep half an eye open for a 'lazy' way to forex riches. But I doubt that I'd feel comfortable risking a significant amount of my own capital in a system that could suddenly and unexpectedly margin call. For better or worse, I just don't have the stomach for it.
  • #16
  • Jul 24, 2011 4:26pm Jul 24, 2011 4:26pm
Hannover, I guess we're all giving Password some feedback. When he pops back up he might share with us how he intended to trade this over the longer term, as at face value without a number of entry rules and clever exiting it's not that viable as is for as a profitable EA in my opinion.

Perhaps he's best to trade it manually.
  • #17
  • Jul 24, 2011 5:00pm Jul 24, 2011 5:00pm
More thinking out loud...

Let's assume for a moment you CAN enter a LONG & SHORT position at the same level... eg,

1. Pending Long Order at 1.2000 Lot Size 0.1
2. Market order filled and market level moves up (we'll come what happens if market moves down later)
3. Pending Short Order at 1.2000 Lot Size 0.2
4. Market moves down and pending order is filled.

This is as Password described. Now lets fast forward to 10 loops (assuming your TP of 10 pips hasn't been hit)

You have the equivlent of 10 lots (Long) [0.1+0.3+0.5+0.7+0.9+1.1+1.3+1.5+1.7+1.9] vs 11 lots (Short).

If we move the TP into 2 pips from 10 pips because we want to get out after 10 loops...

If the market moves to 1.1998 our short trade is +2 pips x 11 lots = $2200 and our long trade is -2pips x 10 lots = -$2000 and close for +200 profit.

If the market is under 1.200 but above the 1.1998 TP and moves up through 1.2000 we should add 1 lot to balance the trade (during our exit phase) so we are long 11 and short 11. {I know the numbers are no pratical I just use them for example purposes}

We can now exit at breakeven.

Back to point #2 above, obviously we cannot assume that the market will move up when you place the first trade, so for that reason it might be wise to mirror this process, by opening a short position at the start also.

Does this make sense or have a lost my marbles
  • #18
  • Edited 10:08pm Jul 24, 2011 9:56pm | Edited 10:08pm
Quoting hanover
Any mechanical strategy that has a genuine edge will be profitable regardless of whether it's traded manually or automatically.



If I've understood you correctly, and at the risk of sounding negative, then my view is that no analysis = no demonstrative edge, and the risk here is whether price moves sideways for long enough to allow your position size to escalate to the point where the drawdown becomes unacceptably large (the "death trade").

I think you acknowledge this by your use of the word "almost". The system is guaranteed...
Yaa, its absolutely correct… market moves sideways and this may lead to uncontrollable drawdown and also lot size.

But one thing need to notice is that as the lot size increases take profit target in terms of PIPS decreases ie less pips required to hit TP. Its because TP is in $. So it create bottleneck for trades to exit.

Well all other queries can be answer only if we can autotrade his strategy. I am not a coder so I can’t do this. If any coder is here... then please try to code this one.

I am trading this manually and getting some good results…

Regards
Password
  • #19
  • Edited 2:42am Jul 25, 2011 2:27am | Edited 2:42am
Gets even more interesting Password. Good idea.
  • #20
  • Jul 25, 2011 8:33am Jul 25, 2011 8:33am
Quoting r2997790
Gets even more interesting Password. Good idea.
thanx
0 traders viewing now