T101 basket trading system - math analysis - page 4

 

Update

Ok, I was quiet this week, mostly due to PC crash, as I had to reinstall whole system. Anyway, I'm back online.

Fortunately I did manage to do some manual trading using T101 and CCFp, the results so far are very very promising. With T101 I've started the week perfectly, having started like 3-4 baskets profitable, yesterday account reached +1200 pips. I was using too big lot size, so overall account went +30% by that moment. Please note, that I don't have much experience in manual trading, and it was my first week trading this system as well.

The lack of experience manifested yesterday - 2 baskets lost, down to +250 pips. You can find the account info at MT Intelligence - Stats for t101

My thoughts so far:

  • The system definitely have some potential. When even rookie like me can generate 1200 pips in 3 days (and I saw that more than 5000 was possible), with some training you can become consistently profitable.
  • CHF follows EUR very closely - it turns out that all CHF pairs should be reversed for better income. I reversed some trades already.
  • I found profit calc graph very useful and surprisingly smooth. Or maybe it was just a quiet period..
  • EUR/JPY is definitely leading, as in math expectations. Opening 14 eur/jpy trades instead of 14 irregular, would result in 5-10x bigger profits. Of course instead of trading 14 eur/jpy lots, you can reduce both the account size and lot amount.
  • I still lack of discipline to refine clear entry conditions - being involved in that whipsaw and losing 80% profits was just an example. Gotta work on it more.
  • Orest and CCFp seem to do their job well, I based on them heavily.
  • I've created small script for doing snapshots of mt4 chart at each candle and ftping it to the server, then some simple php script to display it.. see it at http://www.stressfreerobot.com/orest/i.php - just dont kill my bandwidth, please

PS. In meantime I've opened separate account for pure CCFp trading, seems to be very promising as well. I started with H4 timeframe, got only one good setup for whole week, earned like +400 pips using 4 trades only, with almost zero drawdown. Will try more, second entry is forming, so probably early next week I'll enter again.

 

basket trade

some of my basket

 

some of my basket

 

Hi, Nopik:

earlier than this post, you stated in Oanda with two snapshots that 1 lot (US$100000 worth) CHF/JPY is approximately the net exposure of opening a 14 pairs basket positions.

Now you mentioned 4 lots eurjpy is the net position.

so am i misunderstanding some of your comments?

Nopik:

As for the drawdown, there is no such thing like max drawdown here. If you buy or sell 14 pairs as described, you'll end up having about 4 lots eur against jpy. Asking for max drawdown is asking for how long eur/jpy can go.

 

Yes, indeed. chf/jpy is for IA account (7 buys+7sells), while eur/jpy is for basket (14 buys or 14 sells). Interestingly, eur and chf are very strongly correlated, so by measuring IA, you can trade baskets.

 

sigh...jumping from original thread to here...

read four pages till now, don't know what i've got so far except for chfjpy astonishing exposure (thanks for this finding).....

not saying that your inputs are meaningless, but feeling this T101 system is really a "personal" system takes a trader's own understanding and speculation that proved to be DIFFICULT to convey

my understanding of the 7 buy 7 sell IA positions are the keys of the system. That's the CORE.

if a trader tries so hard to find eurjpy is very close to 14 buy/sell the basket and then jump to watch eurjpy closely, it will be a tradgedy.

it is fine to instead, trade 14 lots of eurjpy. however, all the time, keeping our eyes on IA instead of chfjpy (or eurjpy).

the IA is showing multi-fold info, a trader often tend to find one fold and then think that is it. for example, an entry after spotting a pair's jump, turned out profitable, then declared that's it.

NO, That is NOT it.

IA presents these below, (see if your mind ignore any of these):

1. it suggests market moves from a week openning (week open, it is so essential...);

2. it keeps feeding the moves/jumps between 14 pairs (watching chfjpy won't tell you that)

3. it shows the pips (Monday morning, eurjpy +25 pips, this means nothing, Tuesday morning, eurjpy +150 pips, this means something!)

4. how much 7 buys' p/l is and 7 sells' p/l is? are these two numbers balanced always ( mathABS(7 buys pl) - mathABS(7 sells pl) = X, how X evolves day by day? -- i ignored this too myself)

...... (you help me list more...)

any efforts to simplify/change the basket, is a diversion from the T101 system, and mostly will fail, i guess.

****************************

As to 560 stoploss, i don't think it is an extremly appropriate set up.

when a stopout occurred, you will see usually GJ is 100~120 pips negative, with a couple of pairs with 10~20 winning pips. this is not what the original 40 pips each pair wants.....

I typed these crap not to discourage/criticize anyone, just hope to bring up more insights from you and others with my 1 cent.

many thanks.

Nopik:
My thoughts so far:.
 

Hi Charvo,

Thanks for your input. You seem to have experience in trading this system already (are you profitable with it? how much, if we can ask?) - any new opinion is always welcomed

charvo:

my understanding of the 7 buy 7 sell IA positions are the keys of the system. That's the CORE.

if a trader tries so hard to find eurjpy is very close to 14 buy/sell the basket and then jump to watch eurjpy closely, it will be a tradgedy.

You completely missed my points here First of all, I do watch buy/sell balance from IA account as well as Orest indicator myself. My point is, I want to undertstand what I am looking at. So, for example, the core of this method is to watch IA then trade regular long or short basket, right?

But what will happen if EUR/CHF will get big jump, like it did 1st of April this year? How this affect this trading strategy? Also, if EUR/JPY is close to strong support/resistance level, while CHF/JPY is not, does it matter to this strategy? The thing is - it does matter.

So, one can blindly follow IA account watching his 7 buys and 7 sells and profit since week beginning, then trade only based on this info. And, indeed, I believe that one can be profitable with this approach.

Yet, if trader knows what he is doing, he can easily get more info. For example this week I was seeing small trend reversal on IA account, from long to short. I was not sure if this is trend reversal or only small pullback, but decided to open short basket anyway. Guess what - eur/jpy was just above strong support level was not going down. If I would notice that earlier, I would not open my position. And so on.

charvo:

it is fine to instead, trade 14 lots of eurjpy. however, all the time, keeping our eyes on IA instead of chfjpy (or eurjpy).

My point is to keep your eyes *both* on IA *and* chf/jpy. And eur/jpy and eur/chf as well. Eur/chf only for noticing big moves, like 1st of April this year, nothing more. eur/jpy and chf/jpy are more important here. E.g. if there was important news in Swiss this day, this will skew your signal, and this is something what you will not know if you trade original method blindly.

charvo:

the IA is showing multi-fold info, a trader often tend to find one fold and then think that is it. for example, an entry after spotting a pair's jump, turned out profitable, then declared that's it.

Indeed. I am just adding more folds, not removing existing ones.

charvo:

1. it suggests market moves from a week openning (week open, it is so essential...);

Yes, I'm watching these as well, especially during mid-week. Thursday-Friday this info is less important, as trend already made the move and direction is unlikely to change. One may at most decide to not trade against weekly trend.

charvo:

2. it keeps feeding the moves/jumps between 14 pairs (watching chfjpy won't tell you that)

But watching chf/jpy can tell you something in addition to watching 14 pairs alone.

charvo:

3. it shows the pips (Monday morning, eurjpy +25 pips, this means nothing, Tuesday morning, eurjpy +150 pips, this means something!)

I think that usd profit here is more important than the pips.. I actually do not like the fact that Orest shows in pips, this is misleading in my opinion.

charvo:

4. how much 7 buys' p/l is and 7 sells' p/l is? are these two numbers balanced always ( mathABS(7 buys pl) - mathABS(7 sells pl) = X, how X evolves day by day? -- i ignored this too myself)

X evolves just as chf/jpy evolves, you probably already found that in my previous points. That was important finding for me, as previously I thought that it will oscillate much more.

charvo:

any efforts to simplify/change the basket, is a diversion from the T101 system, and mostly will fail, i guess.

No, I can't agree. If trader knows what is he doing, and does it properly, he will succeed. Though things might be just harder to read, thats all. As far as I know nobody tried that yet (at least nobody said so in both original thread and the one on this forum), and probably nobody will try. But I think it is doable.

For example, this system shows you correlation between chf/jpy and other pairs and based on that info you trade eur/jpy-like basket. It can be changed to watching usd/chf and trading eur/usd for example. Though the pair list wont be just 7 buys + 7 sells most likely, so you will still need to have 2 groups, but the groups will be mixed. Or maybe it is still possible to have 7b+7s pair list to have net on any pair? I'd like to see that. Theoretically if you get original pair list and change one buy to one sell and one sell to one buy, you'll change the final exposure. And by doing such 1+1 changes repeatedly, you might finetune your exposure to be strong on usd/chf, not on chf/jpy. Though opening just the same 14 buys later doesnt seem proper, anymore.

charvo:

As to 560 stoploss, i don't think it is an extremly appropriate set up.

when a stopout occurred, you will see usually GJ is 100~120 pips negative, with a couple of pairs with 10~20 winning pips. this is not what the original 40 pips each pair wants.....

The method needs some stop loss, for sure. Original approach of using account protect script looks interesting, by locking profit quickly it hides many errors from trading - if you enter on wrong moment you are still able to get like +$10 locked. So far I do not found any better exit than +560, but I'd like to learn about it.

charvo:

I typed these crap not to discourage/criticize anyone, just hope to bring up more insights from you and others with my 1 cent.

Sure, this is discussion, and we discuss different opinions, in a peaceful manner

 

my live trading is a little worse than demoing.

not profitable so far. however, i feel many loss are avoidable (caused by my absent mind or careless). the potential is definitely good. i was the 4th or 5th comment in the original thread, followed it, quit it, and come back.... it's been two years almost, and i think that this should be the best method i've seen all these years.

this method, i feel, ultimately is to:

spot theoverall market trend at anearly stage.

sort of like a trend-follow system, (i said sort of because this T101 can do some trend reversal trades too, but mainly it is a trend system)

therefore, it is very simple to say, when there is trend, you earn money, when there is no trend, you lose. that is pretty much why you lost 1200 pips last week--no trend.

you might want to look back the past week, with hindsight even, to see if yesterday once more could you do it better? yes, i think you may lose only 600 pips or even making a few hundred pips, but it is absolutely much harder than the end of March, or two weeks prior, right?

these few days, i gradually treat this thing as a multi-currency meter, sparing me from jumping among varied charts, hohoho. it feels like that the more i use this, the less i know how to.

Nopik:
Sure, this is discussion, and we discuss different opinions, in a peaceful manner
 
Nopik:
Hi Charvo,

My point is to keep your eyes *both* on IA *and* chf/jpy. And eur/jpy and eur/chf as well. Eur/chf only for noticing big moves, like 1st of April this year, nothing more. eur/jpy and chf/jpy are more important here. E.g. if there was important news in Swiss this day, this will skew your signal, and this is something what you will not know if you trade original method blindly.

But watching chf/jpy can tell you something in addition to watching 14 pairs alone.

I think that usd profit here is more important than the pips.. I actually do not like the fact that Orest shows in pips, this is misleading in my opinion.

Nopik,

Your observations are very refreshing indeed. I have tried to trade this system a few times before, but not with much success. Just now GBPUSD and GBPJPY had a rally but basket did not respond much and the change was marginal. CHFJPY did not move at all. So there is a possibility of misjudging the trend if we go by the major movement in GJ and comment on usd profit for pip instead of pip profit is also very true, as I have found out to after suffering some losses some time back.

Keep it going.

 

Virtual Trade Monitor v2.1

Hello all,

Attached you'll find the Virtual Trade Monitor v2.1 (indicator).

It is based on the great work of the A. Samirs indicator. I wanted to apply some small changes to A. Samirs indicator and ended up in an complete rewrite, featuring some nice extras like:

1. Auto-Detection of supported pairs by your broker

2. Support for odd broker symbols (like "USDCHFm" etc.)

3. Visual jumping pair display (really cool)

4. Jump pair info (which par, from slot/to slot, time passed since jump)

5. Calc takes now spreads into account

6. Sums of Buy/Sell pairs

Although it is still not perfect, I decided to release it to the public. I did test it extensively on Alpari UK, but not on other brokers. If you find any problems, I'll do my very best to solve them.

Installation:

==========

1. Copy the indicator (Virtual Trade Monitor v2.1.ex4) into the folder experts/indicators

2. add the Virtual Trade Monitor to a chart (any timeframe).

3. make sure you have activated Chart Shift

Configuration:

===========

There are a few new options in the config dialog:

1. AutoDetectPairs (true/false). If set to true, the Virtual Trade Monitor will test, if your broker supports the original currency pair set or the alternative and selects the appropriate one. If set to false it will use the alternative (IBFX) pairs, but you can overwrite them under "Custom Sell Pairs" and "Custom Buy Pairs". Which pairs are currently used is shown in brackets in the line below "Virtual Trade Monitor" (near the top right corner of the chart)

2. ShowJumps (true/false). If set to true, the jumps will be shown visually by lines and arrows. This is maybe one of the most important features of the Virtual Trade Monitor, as it shows how the pairs are moving after the jump!

3. DisplayJumpInfo (true/false). If set to true Virtual Trade Monitor will show below the sums of the sell/buy pairs extra informations about the last jump. Especially the time that passed since the jump happened (e.g. if you don't sit in front of your computer all the time). In combination with the ShowJumps setting, it is a nice tool to monitor the movement of jumping pairs.

4. ShowJumpAlert (true/false). If set to true, you'll will be informed about every jump through an alert box (with sound).

5. ShowBaskedBackground (true/false). If set to true, the background of the basket indicator will be colored in red (sell pairs) and green (buy pairs). This is just to give a better visual impression and makes it easier to differ between the upper "sell zone" and the lower "buy zone". You can of course set it to false to turn the background shading off.

6. Custom SELL pairs and Custom BUY pairs can be used only if AutoDetectPairs is set to false! By default the alternative set of pairs is used, but you can put in your own pairs of course.

Credits go to A. Samir for his great work on the A. Samir-UNIVERSAL indicator, which I have used as base for the Virtual Trade Monitor - respect Samir!

Expect some additional features very soon...

Cheers,

Andy

Files: