The finest in trend trading

[didilut]

Save the big gun of the 5 lot for the perfect opportunity to use it. I only see that perfect opportunity once or twice a week, certainly not every day and very rarely two days in a row. The rest of the time I console myself with small pips and wait for the obvious 5 lot trading opportunity. I believe the main problem some people have with 5 lot is having the patience to wait for the perfect opportunity that only occurs about once or twice a week to pull the trigger on the 5 lot. That is, they try to trade every entry as a 5 lot entry, and the vast majority of good entries are not a 5 lot opportunity. This is basically a form of over trading that can be effectively addressed in your trading plan. You do have a trading plan, right? Practice makes perfect

Graviton Hello, and thank you for the reply:)
In summary, I think the difficulty of “pyramiding” is defined as clearly as possible the strength of the PA, as pointed out rightly Dpaterso, why not use a% of ATR, or as Fibos Basic Input or retracement.
I am a newbie, do not hesitate to “correct” if I say a huge mistake.
Thank you again to all the “old” :eek: to guide us and make our minds work

[Merchantprince]

First (and stating the obvious): your BE point (when using multiple entries) is the AVERAGE price of all of your entry prices. Tymen is quite correct in noting that BE on those two entries (in the example) is NOT the BE point of the FIRST entry but is rather the MIDPOINT of the two entries.

And I’m thinking that since we are looking at managing the BE point of multiple trades, why not simply draw a fib from the opening price of the first trade to the furthest point price has moved in the direction of the current trades? The 50% level of this fib would be our “bail out” point to close all positions, should price fall back that far. For myself, I would use this ONLY while adding new lots on. Once the final lot in the series has been added, I’d likely move my stops up to the BE point of either the final lot or the second-to-last lot in the series (if 4 or 5 lots are in play).

In any case, what I have been doing on the few occasions this week I’ve tried multi-lot is to simply add the new lot on at the predetermined point, then move the stop of the previous trade to BE. I manage the trades by observing price action (usually in how it behaves relative to the bolls – including looking at the smaller TFs to see any dramatic changes early enough), closing out [I]all[/I] trades if it looks like a retrace is imminent.

I have also tried simply opening every new trade with a trailing stop equivalent to the distance of the next trade in the series. Once the final lot in the series is open (and I don’t always use 5 lots, sometimes only 3 or 4) and has moved at least far enough to equal the trailing stop value, I open each trade and un-check the “trailing” stop-loss feature in my platform. From that point forward, I manage the stops collectively based on chart analysis, moving them all forward should the trend continue for a longer period of time.

Second and as to HOW FAR each new entry should be from the previous one: what about using a percentage of Wilder’s ATR(Period) to determine the next entry price e.g. 25% ATR(Period) from the previous entry??? At least this way your new entries are taking volatility into account and are not simply chosen on a static ‘one size hopefully fits all’ approach???

Good reply; this is exactly where my head has been on this subject. I mentioned in a previous post of my own that it makes sense to employ ATR in some way to calculate stops, as this is the same value used to set entries in the multi-lot system. Certainly one shouldn’t use the same value to calculate stops/multi-lot entry distances for GBP/JPY as they do for EUR/GBP! :eek:

I wonder, would one use ATR values calculated on the fly (which would seem to be very challenging) or would it be sufficient to calculate the most recent ATRs, say for the past week, and then just differentiate between those ranges calculated during peak trading hours and those derived from off-peak hours? Grav has already stated he makes this differentiation himself.

This could get complicated and yield weird numbers; I’m a fan of nice, easily calculable increments and would probably get as close to those as possible (for instance, if the ATR during peaks hours for a pair is 13.8 I’d probably just use 15).

As a side-note, this value we are discussing - the distance of the stops in a multilot strategy, which also determines the distance to the entry point of the next trade in the series - doesn’t yet have a term.

Maybe we should invent one.

[dpaterso]

Hello,

Thanks for considering my input.

If the truth be told: there is another rather ‘famous’ trading system that pyramids entries at a certain percentage of the ATR(Period) so I cannot claim the ‘glory’ for this idea. Regarding WHICH value of ATR(Period) to use: this very same system uses the ATR(Period) value at the close on a Friday for the rest of the coming week. That keeps things simple.

Regards,

Dale.

Edit: just to be clear I’m talking about the DAILY CLOSING ATR(Period) at the end of the previous week NOT the WEEKLY CLOSING ATR(Period) (that would in most cases be HUGE)!!! LOL!!!

[tymen1]

Tymen, you are correct that you would have a loss, but looking at Graviton’s post above, the bold font says to stop the trade out manually before it goes back to the b/e point on the first trade.

Yes, [B]Graviton [/B]in an earlier post said he still likes to take the 1st contract out of the trade so that it is a confirmed profit.

I am still undecided whether to make that part of my GFT multiple system.
However, there are benefits for keeping the 1st contract in the trade.

[tymen1]

Yes, everyone is correct.

… In that case, as you correctly calculated, risk never exceeds the 25 pip initial stop loss value, but that risk is carried on the second lot and exists until the third lot is put on …

… an alternative … That is to let the 1st lot run, in this case, a full 50 pips up …when 25 pips of profit are locked in …put on the second lot with a 25 pip stop loss.

This method takes the risk out of the trade as soon as the second lot is on…

Ah, I am so glad that you have clarified that matter, Graviton.
It is important that I get my facts straight so that I can properly consider what I am doing in devising this GFT averaging multiple entry method.

As I said once before, it’s probably not optimum, but it serves me well.
…

You have also stated that you like to take out the 1st contract (lot) after it reaches a certain profit level.

This is a different senario from what you have explained above.
Could you clarify as to when you take out the 1st contract and when you leave it in to run.

Your answer here will again be of great assistance to me in my devising work.

[tymen1]

…
…

Second and as to HOW FAR each new entry should be from the previous one: what about using a percentage of Wilder’s ATR(Period) to determine the next entry price e.g. 25% ATR(Period) from the previous entry???

[B]Welcome to this thread, Dale!! [/B]

For those of you who do not know, [B]Dale Paterson, Honorary FX Member[/B] is the 2nd longest serving member on this forum after [B]Rhodytrader, Honorary FX Member. [/B]

There is probably very little time difference between the 2 people.

[B]Dale carries a very high degree of respect on this forum and it is a privilage and an honor to have him on this thread to contribute his great knowledge.[/B]

Now I am going back to devising the multiple entry method for GFT programs.

At present I am working with the notion of using the average true range to effect a first entry amount.

As you can see from my above comment on my post #3040 on page 304, I was already starting out with the ATR principle in mind.

I am definitely on the right track!!

[tymen1]

… why not simply draw a fib from the opening price of the first trade to the furthest point price has moved in the direction of the current trades?

Wait for it, [B]Merchantprince[/B]!!

I have already pre-empted this matter!!

It will become the stuff of the macro method - complete with the multiple entry method!!

The 50% level of this fib would be our “bail out” point to close all positions, should price fall back that far.

Actually the bail out (exit) will be a lot less than that!!

The Fibonacci levels should be adjustable in your charting program so that will be necessary.

[tymen1]

Hello everybody,
…

Many many thanks to Tymen and all the fantastic crew of contributors to this amazing thread, I look forward to being able to add what I can here.

Welcome back again, [B]Mastersafe[/B]!!

I like your avatar (very appropriate!!).

[Graviton]

Hello,

I’ve FINALLY reached the stage where I no longer to wish to ‘chop and change’ from what works for me nor do I need to ‘prove a point’ anymore. It’s quite boring actually but apparantely it’s ‘the place to be’!!! LOL!!!

The above being said: ‘just for fun’ I monitor this thread as often as possible and, up until now, have had nothing worthwhile to contribute. HOWEVER: this morning I feel that I MAY have something worthwhile to contribute regarding multiple entries (or ‘pyramiding’ as it is commonly known).

First (and stating the obvious): your BE point (when using multiple entries) is the AVERAGE price of all of your entry prices. Tymen is quite correct in noting that BE on those two entreis (in the example) is NOT the BE point of the FIRST entry but is rather the MIDPOINT of the two entries.

Second and as to HOW FAR each new entry should be from the previous one: what about using a percentage of Wilder’s ATR(Period) to determine the next entry price e.g. 25% ATR(Period) from the previous entry??? At least this way your new entries are taking volatility into account and are not simply chosen on a static ‘one size hopefully fits all’ approach???

Regards,

Dale.

Dale, I agree with your second point that using ATR to size increments takes market volatility into account and that is a good idea. I’m hoping someone will do some testing at some point to help produce a better formula. But I agree that some % of ATR is probably a good start. 50% of ATR would generally produce about the same stop loss settings as I usually use.

On the first point, yes, in retrospect the BE of the first two lots is the average entry point of the two. But that’s hindsight. Looking forward into a trade, you don’t know when the trend will end, so you don’t know if you will be putting on 1, 2, 3, 4 or 5 lots, so there is no way to predict where that final average will wind up.

I have explained NB’s “chicken” entry elsewhere, where he waits until the first lot has moved 2X stoploss increment, before moving the 1st lot’s SL to 1X and locks in 1X profit. He then puts on the second lot with a 1X SL. This assures a worst case BE of 0 in case of a quick reversal. NB also puts on later lots a little faster, so his end result, iff the trade runs the full 5 lots, is to achieve the same pips profit, but removing the risk from the trade faster. I’m sure he could explain what he is doing better than I.

There are many possible variations to this positioning system and I don’t know of any one that is best for all traders in all markets. I strongly advise everyone to test it out carefully in demo before trying it with real hard earned capital.

Thanks for your comments and congrats on finding your personal “Holy Grail”.

[tymen1]

My last post for this evening because it is very late here (2.30am)

I have just put a lot of pictures of Perth and Western Australia on [B]Merchantprince’s [/B]whiteboard.

If anyone is interested, go take a look.
[B]Gregart [/B]was there and he enjoyed them!!

[Graviton]

Ah, I am so glad that you have clarified that matter, Graviton.
It is important that I get my facts straight so that I can properly consider what I am doing in devising this GFT averaging multiple entry method.

You have also stated that you like to take out the 1st contract (lot) after it reaches a certain profit level.

This is a different senario from what you have explained above.
Could you clarify as to when you take out the 1st contract and when you leave it in to run.

Your answer here will again be of great assistance to me in my devising work.

I suppose I’ve tried every combination I can think of. These days, I am taking the first lot off with profit when I see the first clear sign of retracement. If the retracement continues I close all lots. I suppose I need to create a hard rule for that TP, but as yet I haven’t. I’m interested in MP’s suggestion though to use a fib to determine exit point.

[tymen1]

I suppose I’ve tried every combination I can think of. …
. I’m interested in MP’s suggestion though to use a fib to determine exit point.

Thank you for your prompt attention to my question, [B]Graviton[/B]!!

Fear not - [B]Merchantprince’s [/B]idea is not new and I have already tested it and it will be a staple diet in the macro method I will post soon.

[B]I am now to do the following…[/B]

1. Post the multiple entry method for GFT type programs where the entries are averaged.

2. Go onto [B]Merchantprince’s [/B]whiteboard and demonstrate a few DNA trades. ( I will probably use the original 2 contract entry method).

3. Post the macro method.
   This method, once outlined, will also be demonstrated on the whiteboard.

[tymen1]

My introduction pictures of Western Australia on Merchantprince’s whiteboard are now fully labelled - take a look if you wish.
([B]Green as Grass[/B] has seen them now too).

Its only a matter of time before we arrange to do a DNA demo trade on the whiteboard.

[IronHeart]

Hi Tymen, can you please provide a link to the whiteboard? I can’t seem to find your pictures on the site!

Also, I’m sure the there already is a name for the pip-difference between the multi-lot entry points but…I would like to call it “[B]Pyramid Depth[/B]”.

What do you guys think?

[Merchantprince]

Ironheart, a link to our whiteboard can be found on the BB DNA chat home page.

[IronHeart]

Thanks MerchantPrince.

Greg has kindly messaged me about this already but I’m sure your post will be helpful to many still to come!

[tymen1]

Here is my considered approach to multiple contract entry trading.

It combines many features >>>

Overview

This approach has very much a war zone mentality about it.
A profit is established as soon as possible and is held onto.
I have considered the establishment of an early profit to be mandatory because a retracement could come at any time.
Also with the DNA method, the trade distance is relatively short.
A slight delay is made before more contracts are added.

After much consideration, I have stuck with the 2 contract entry… I find that it simply cannot be beaten. (obviously designed by experts).
Instead we shorten the initial stoploss to 20 pips.
This should be enough to trade your belief - that the trade is going forward.

A loss here would generate a 40 pip loss which is not too bad considering the reward potential is very great.

The extra contracts are added on every 20 pips. These short distances allow for a maximum profit before a retracement occurs. This setup aims to generate many pips and do so as quickly as practicable.

The whole thing is carefully designed to allow for the averaging of the entries.

Explanation

The blue lines are the levels of price action.
The red lines are the trailing stoplosses.
The pink lines are the averaged entries with the entry number listed alongside.

The trading steps are given large black numbers and the corresponding numbers are also given to the trailing stoploss.

The trading steps and their maximum possible losses are now considered…

1. At price action = 0 we enter 2 contracts and set the stoploss at -20.
   If the price action drops to the stoploss we have…

2 contracts = -20 x 2
TOTAL = -40 pips.

2. At price action = 20 we exit and bank 1 contract for a profit = 20 pips.
   We immediately reset the stoploss to zero.
   If the price action drops back to zero, we have…

1st contract = 0
Bank = +20
TOTAL = +20 pips.

3. At price action = 40 a new 2nd contract is added.
   The stoploss is immediately moved to price = 20 (trailing).
   If the price action drops back to 20, we have…

1st contract = +20
2nd contract = -20
Bank = +20
TOTAL = +20 pips.

(averaged entry in pink = 20).

4. At price action = 60 a new 3rd contract is added.
   The stoploss is immediately moved to price = 40 (trailing).
   If the price action drops back to 40, we have …

1st contract = +40
2nd contract = 0
3rd contract = -20
Bank = +20
TOTAL = +40 pips.

(averaged entry in pink = 33).

5. At price action = 80 a new 4th contract is added.
   The stoploss is immediately moved to price = 60 (trailing).
   If the price action drops back to 20, we have…

1st contract = +60
2nd contract = +20
3rd contract = 0
4th contract = -20
Bank = +20
TOTAL = +80 pips.

(averaged entry in pink = 45).

It is possible to keep adding contracts if you wish.

(checked and corrected for accuracy, 25 May, 2010).

[gregart]

I take off my hat to master Tymen ! great teacher !

I understood at the first time when I had read, it’s an achievement

thanks a lot Tymen

Greg

[IronHeart]

3. [B]At price action = 40 a new 2nd contract is added.[/B]
   The stoploss is immediately moved to price = 20 (trailing).
   If the price action drops back to 20, we have…

1st contract = +20
Bank = +20
[B]TOTAL = +40 pips. [/B]

(averaged entry in pink = 20).

Tymen, this is fantastic stuff. Thanks very much for the time you have invested into this.

I have spotted one small mistake, however!

Regarding the quote above, the contract that is entered at 40 will lose 20 pips, causing the total to be +20 pips and not +40 pips.

[Hordane]

I like this entry method, thank you for putting it out there.