Martingale: Why does this not work? Any value to this?

Might as well do this now - whilst lunching. (grilled salmon, and half a bottle of Reisling :) )

Are markets more likely to stay in a narrow range, or move about away from the range?
I assume markets are more likely to move away from a price point, rather than stay.
If the move away from the range is large enough, you could bet on an outcome.
Even if the price stays within a range, you have an idea of the size of the range!
(this was an unexpected realization)
This can give you some idea of potential outcome.
I don't even have to know the direction of the move. (another plus, especially for automation)

1: Suppose, you pick a price point. Any price point.
1a: determine a pip range-size.
In my example, I have picked 20pips price-range. Zero-point = 1.1000.
1b: determine a trade range-size. I have picked 80 pips. (4 grids away from 1.1080).

2: Trade:
2a: I place a Long at 1.1000.
If all goes well, price surges from 1.1000 to 1.1080, and I cash out 4 grids.
Technically, I am "risking" 20 pips to gain 80. (L=4 on the Set1 image)
(Set1)

2b: Price goes against me.
If price goes at least 20 pips against me, I will take a Short at 1.0980.
I go Short 2 lots. (I leave the 1 Long running).
I am now gunning for price to go 4 grids lower. Price point 1.0900.

If the price goes to 1.0900:
2c: At level 5, price=1.0980, market has gone 4 in my favour.
I have a Long which is losing 5, but my 2 Shorts are giving me 8.
2d: I close out with nett=3 lots. (Set2 on image: L = -5: S = 8)
PS: I was holding a nett position of 1 lot. (1 Long, 2 Short)
PS2: I have decided here that any price above 1.1000 is a Long play, and anything below 1.0980 is a short play. The range within is a sort of DMZ.

2e: But, if, horror of horrors, price dupes me once again, and doesnt get to 10900, but decides to rally above 1.1000, I would:
2f: Take out 3 Longs.
My nett position now would be 2 Short, and now 4 Long.
If price goes to 1.1080, I would cash out.
Set3 shows my 2 Shorts would be -10, but my 4 Longs would give me 16, a nett plus 6.

At this point I am holding a nett 2 position. 2 Short and 4 Long.

This should now give some idea to you how Set4 and Set5 pans out.

Hope the above makes sense.
 
For clarity,
1: you can decide for yourself the pip-range. I would advise an ATR, to reduce whipsawing.
2: you can decide for yourself the numbers of levels. You would use some TA to gauge this.

This idea requires a differential number of lots Long and Short, and some notion of a target.
One side of Longs or Shorts "catches up" with the other side.
Also, you could decide, if you wished to cash out earlier with either a smaller gain, or a controlled risk.

In my example, the "losing" trade loses you 5 lots, and the winning trade gains you 4 lots.

So, Set5 has 6 Short and costs you 5 lots, thus -30.
Therefore, the Long trade, which has to cover 4 lots of ground, has to be equal to or greater than 30.
8 lots would give you 32. I chose 9.
At Set5, I have 6 Shorts, and 9 Longs, only haveing a nett risk of 3 lots at Set5.

So, if Set5 failed, I would have 9 Long, this would give me 45 (9x5) loss at 1.0900.
There I would look for a 4 multiple to cover that. 4 times 12 would give me 48.
So, if I had to go short, I would add another 6. Making 12 short.
If price hit 1.0900 I would close out with a 3 lot gain.
I would still only have a nett 3 lot Short/Long (12 - 9)

** I wanted to see if I could find a way of not doubling up the risk at each failed direction, whilst having a neutral direction bias.
** I wanted to find a way of defeating the Whipsaw Song!

 
Last edited:
Might as well do this now - whilst lunching. (grilled salmon, and half a bottle of Reisling :) )

Are markets more likely to stay in a narrow range, or move about away from the range?
I assume markets are more likely to move away from a price point, rather than stay.
If the move away from the range is large enough, you could bet on an outcome.
Even if the price stays within a range, you have an idea of the size of the range!
(this was an unexpected realization)
This can give you some idea of potential outcome.
I don't even have to know the direction of the move. (another plus, especially for automation)

1: Suppose, you pick a price point. Any price point.
1a: determine a pip range-size.
In my example, I have picked 20pips price-range. Zero-point = 1.1000.
1b: determine a trade range-size. I have picked 80 pips. (4 grids away from 1.1080).

2: Trade:
2a: I place a Long at 1.1000.
If all goes well, price surges from 1.1000 to 1.1080, and I cash out 4 grids.
Technically, I am "risking" 20 pips to gain 80. (L=4 on the Set1 image)
(Set1)

2b: Price goes against me.
If price goes at least 20 pips against me, I will take a Short at 1.0980.
I go Short 2 lots. (I leave the 1 Long running).
I am now gunning for price to go 4 grids lower. Price point 1.0900.

If the price goes to 1.0900:
2c: At level 5, price=1.0980, market has gone 4 in my favour.
I have a Long which is losing 5, but my 2 Shorts are giving me 8.
2d: I close out with nett=3 lots. (Set2 on image: L = -5: S = 8)
PS: I was holding a nett position of 1 lot. (1 Long, 2 Short)
PS2: I have decided here that any price above 1.1000 is a Long play, and anything below 1.0980 is a short play. The range within is a sort of DMZ.

2e: But, if, horror of horrors, price dupes me once again, and doesnt get to 10900, but decides to rally above 1.1000, I would:
2f: Take out 3 Longs.
My nett position now would be 2 Short, and now 4 Long.
If price goes to 1.1080, I would cash out.
Set3 shows my 2 Shorts would be -10, but my 4 Longs would give me 16, a nett plus 6.

At this point I am holding a nett 2 position. 2 Short and 4 Long.

This should now give some idea to you how Set4 and Set5 pans out.

Hope the above makes sense.
Might as well do this now - whilst lunching. (grilled salmon, and half a bottle of Reisling :) )

Are markets more likely to stay in a narrow range, or move about away from the range?
I assume markets are more likely to move away from a price point, rather than stay.
If the move away from the range is large enough, you could bet on an outcome.
Even if the price stays within a range, you have an idea of the size of the range!
(this was an unexpected realization)
This can give you some idea of potential outcome.
I don't even have to know the direction of the move. (another plus, especially for automation)

1: Suppose, you pick a price point. Any price point.
1a: determine a pip range-size.
In my example, I have picked 20pips price-range. Zero-point = 1.1000.
1b: determine a trade range-size. I have picked 80 pips. (4 grids away from 1.1080).

2: Trade:
2a: I place a Long at 1.1000.
If all goes well, price surges from 1.1000 to 1.1080, and I cash out 4 grids.
Technically, I am "risking" 20 pips to gain 80. (L=4 on the Set1 image)
(Set1)

2b: Price goes against me.
If price goes at least 20 pips against me, I will take a Short at 1.0980.
I go Short 2 lots. (I leave the 1 Long running).
I am now gunning for price to go 4 grids lower. Price point 1.0900.

If the price goes to 1.0900:
2c: At level 5, price=1.0980, market has gone 4 in my favour.
I have a Long which is losing 5, but my 2 Shorts are giving me 8.
2d: I close out with nett=3 lots. (Set2 on image: L = -5: S = 8)
PS: I was holding a nett position of 1 lot. (1 Long, 2 Short)
PS2: I have decided here that any price above 1.1000 is a Long play, and anything below 1.0980 is a short play. The range within is a sort of DMZ.

2e: But, if, horror of horrors, price dupes me once again, and doesnt get to 10900, but decides to rally above 1.1000, I would:
2f: Take out 3 Longs.
My nett position now would be 2 Short, and now 4 Long.
If price goes to 1.1080, I would cash out.
Set3 shows my 2 Shorts would be -10, but my 4 Longs would give me 16, a nett plus 6.

At this point I am holding a nett 2 position. 2 Short and 4 Long.

This should now give some idea to you how Set4 and Set5 pans out.

Hope the above makes sense.

OK well a couple of things for now.

I did a walk through pretty much identical to this on a fixed 0 price point. ie longs above and shorts below and was horrified to discover that a sideways market could cross that 0 price point 22 times without ever travelling far enough to deliver an overall profit.

This type of system obviously requires a bias stake but doesn't need to be continually adding stake in order to achieve that bias. Dropping off stake has the same effect (creates a bias) does the same job and reduces both margin requirement and overall risk exposure.

Exploring and developing this idea could be quite worthwhile but guaranteed to be a total mind fek if it is to be a total tradable system. Then we have the small problem of automation :ROFLMAO:
 
OK well a couple of things for now.

I did a walk through pretty much identical to this on a fixed 0 price point. ie longs above and shorts below and was horrified to discover that a sideways market could cross that 0 price point 22 times without ever travelling far enough to deliver an overall profit.

This type of system obviously requires a bias stake but doesn't need to be continually adding stake in order to achieve that bias. Dropping off stake has the same effect (creates a bias) does the same job and reduces both margin requirement and overall risk exposure.

Exploring and developing this idea could be quite worthwhile but guaranteed to be a total mind fek if it is to be a total tradable system. Then we have the small problem of automation :ROFLMAO:
I am glad of your responses.
Good to know that someone else has considered such a thing.
The reason I thought ATR should be used to avoid excessive whipsaws if I used an arbitrary range size.
Another thing I looked at avoiding were Round Numbers for location of my Starting price.
Another addition was to only take a position at extremes of moves, such as Bollinger Band extremes, or RSI extremes, where some extreme flickback was likely.
One thing I thought about the rangey was to only take a price point at one end of a known range, to avoid whipsaws in the range.

I do like the reducing stake idea. But I found at some level, you are paying a high price for it.
For example, at Set5 in my example, where I have 6 Short and 9 Long. If I wanted to unravel a position, I would need to take a hit of -5 lots, so the next Set was 6 Short and 4 Long.

Really appreciate the considered responses.
 
I am glad of your responses.
Good to know that someone else has considered such a thing.
The reason I thought ATR should be used to avoid excessive whipsaws if I used an arbitrary range size.
Another thing I looked at avoiding were Round Numbers for location of my Starting price.
Another addition was to only take a position at extremes of moves, such as Bollinger Band extremes, or RSI extremes, where some extreme flickback was likely.
One thing I thought about the rangey was to only take a price point at one end of a known range, to avoid whipsaws in the range.

I do like the reducing stake idea. But I found at some level, you are paying a high price for it.
For example, at Set5 in my example, where I have 6 Short and 9 Long. If I wanted to unravel a position, I would need to take a hit of -5 lots, so the next Set was 6 Short and 4 Long.

Really appreciate the considered responses.

OK well in your first post you mentioned "dynamic" which is almost certainly the solution to the fixed 0 point range problem.
I have as recently as this week been looking at 3 line break charts and think I may have found a way of playing both sides dynamically whilst being able to incorporate booking profits along the way, where price action gives an opposite signal. We probably need to post some charts for illustration, but this will have to wait until later on.

Back later.
 
Oo’er, be careful you two. All those clever clogs will be after you soon ‘cos you can’t have longs and shorts open at the same time doncha know :)
 
Oo’er, be careful you two. All those clever clogs will be after you soon ‘cos you can’t have longs and shorts open at the same time doncha know :)

Yeah yeah we know, heard it all before :p

Err....is the Spanish flag just a protest or have you moved? :ROFLMAO:
 
Top