Multi-Lot Multi-Timeframe Position trading | Page 32 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

building multi-pair multi-timeframe scanner 48 replies

Multi timeframe multi chart 13 replies

Timeframe inside current timeframe - is that possible? 2 replies

Seeing higher timeframe indicators on lower timeframe 4 replies

BUY run profit 1 lot, SELL LIMIT 2 LOT down again n BUY 1 lot - EA Reqd 5 replies

Attachments: Multi-Lot Multi-Timeframe Position trading
Exit Attachments

Multi-Lot Multi-Timeframe Position trading

  • #621
  • Jan 7, 2015 3:41pm Jan 7, 2015 3:41pm
Quoting AlbinaGrande
{quote} Hi G. My note evaporated. An occasional problem here with WiFi USB connections. But some more of my mumblings from out of the ether. I was happy to see your reply which confirms my thinking that the regular Harvesting of profit from a growing Floating Profit is the way to go. Looking at several of my back testing models I have come to believe that for longer running Trends taking profit, perhaps via Limit Orders somewhere between 500 to 750 pips for each position, results in the smoothest relation between FP and Balance. Above that only...
Good points for discussion AG. I usually only use limit orders or Take profit points to catch a very large quick move in my direction. Those don't happen too often, but when they do it makes for a very profitable day. Otherwise, I want to manually take profit on a local peak in profit as best as I can. It's part of my effort to get better at buying low and selling high.

I'm pretty good at cutting losses short and letting profits run, which are the other two tenants of good trading, though I have to allow myself minor mistakes in that area. I can always re-enter a trade if I fail to let profits run, so that's not too dangerous to my long term survival in the markets. As for cutting losses short, I have taken the pledge to never allow myself to have a large loss again, and I can't allow any mistakes there as it could wipe out my account. When trying to buy low and sell high, sometimes price goes to an even higher peak after I have taken off a lot for profit, but that's fine when I still have 4 more running, and I'll never get better at it if I don't try very hard to improve.

Generally, we have no control over which direction price will go or how fast it will go there. We do have control over which pair we choose to trade, which direction we choose to trade, where we enter and where we exit. I like to maintain as much control over those factors I can affect as possible to help offset the complete absence of control over direction and speed of price movement. That's why almost all of my entries and exits are taken manually even though I'm always trading with SL's and TP's. If I mess up a trade I can try to understand my mistake and do better next time. If the market does something unexpected and messes up my trade, I have no way to fix the market, so I am at a loss as to how to improve.

The reason to take profit off the bottom rung when the third lot is put on is mainly a risk management decision, as I discussed earlier. I will almost always do that to reduce the risk profile in the riskiest part of my trade, when I have just put the third lot on and I'm carrying unrealized gains and have trading capital still at risk. Additionally, that third lot usually goes on when price has made a nice move in my favor, so taking the first off at that point usually results in the "selling High" part of the winning equation. It's very effective.
  • #622
  • Jan 7, 2015 6:14pm Jan 7, 2015 6:14pm
Quoting Graviton
The Yen is recovering some of it's value this week. The Bank of Japan can't be happy about that. The BOJ announced increased QE for 2015 back in Nov of 2014, so we can expect them to move soon. Be patient, but be prepared to catch the next wave of Yen devaluation. GJ has given back exactly 50% of it's 2150 pip rise since October. Will it turn back up right on the 50% Fib line? We'll know in the next day or so. I have no problem riding it back up 1,000 pips to it's previous Dec high if it turns back up. Time will tell.
Hi Graviton,

I agree with you Yen analysis ... JPY should weaken (so xxx/JPY piars up). The monthly trend is up, but daily and lower is generally down.

I downloaded some xxx/JPY index data from a broker, scaled to 1 (so 1.03 is 3% move up, etc) and looked at what the index did historically and averaged over the past 3, 5, and 8 yrs:

http://i.imgur.com/m4y4YZS.png

Since 1/1 to the first week in April, historically xxx/JPY pairs have gone up. Historically we know they will move up and fundamentally we have reasons for these pairs to move up.

I entered my first position long since the trend is generally flat on 15 and 1 hr TF's. My reasoning is this move is near 50% retrace of the overall move, and I expect the historical trend up till April to take place.

One thing I am doing different than you is entering a 'basket', so long 7 different xxx/JPY pairs. I spread my initial 1 lot into seven equal positions. Later as JPY weakens, I'll add to my JPY basket.

Regards,
EZ
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot3.png Size: 86 KB
  • #623
  • Jan 7, 2015 8:51pm Jan 7, 2015 8:51pm
Quoting ezcurrency
{quote} Hi Graviton, I agree with you Yen analysis ... JPY should weaken (so xxx/JPY piars up). The monthly trend is up, but daily and lower is generally down. I downloaded some xxx/JPY index data from a broker, scaled to 1 (so 1.03 is 3% move up, etc) and looked at what the index did historically and averaged over the past 3, 5, and 8 yrs: http://i.imgur.com/m4y4YZS.png Since 1/1 to the first week in April, historically xxx/JPY pairs have gone up. Historically we know they will move up and fundamentally we have reasons for these pairs...

Good Yen discussion. I also do the basket trades, but selectively. So if I have a very weak currency against the Yen, Like the Euro, I'd be less likely to include the pair in my basket. I took a small position in GJ long today and I'm very close to pulling the trigger on another UJ long trade. On UJ, it looks like the Daily is completing a small retracement, but it will take another day or two for me to see if it will head back up. The UJ is back to near 10 year highs now, so I'm not sure how much the BOJ wants to beat this lame horse. But, if it trends, I'll try to trade it.
  • #624
  • Jan 8, 2015 10:50am Jan 8, 2015 10:50am
Graviton,

Yes, good advice on the baskets (selectively selecting pairs). I have an AUD long basket for the past couple of days, but manually exited the GBPAUD short for profit since GBP index is now at support (see screenshot). Since GBP index at support and no big surprise from BOE, I think GBP can bounce here. Also GBPAUD was at WS3 and MS1 confluence of pivots (strong support).

I have also attached my research on historical movements for currency index baskets over the past 3, 5, and 8 yrs. As you can see, AUD and xxx/JPY are set to run up from basically now till April (historically this is the case in the past, but we will see what happens in 2015).

Regards,
EZ
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 155 KB
Attached File(s)
File Type: zip historical currency indicies.zip   1.1 MB | 372 downloads
  • #625
  • Jan 8, 2015 5:23pm Jan 8, 2015 5:23pm
Seeing some slight drifts up in the USD as Asian traders anticipate a strong jobs report in the US tomorrow. Bias is to the upside as a good report is expected to make the USD soar and a poor one should make less of a change. Nothing to do but wait for the news.
  • #626
  • Edited 10:05am Jan 9, 2015 7:46am | Edited 10:05am
Ha ha, nailed it! NFP numbers just as suspected. Nice way to start the day. UCad taking off. A little kickback as news traders take their quick profits, but trend remains strong. Terrorists in France crashing markets now. They will be dead by the end of the NY trading session, but it's playing havoc with markets at the moment.
  • #627
  • Jan 9, 2015 10:38am Jan 9, 2015 10:38am
Yes, and also weak fundamental CAD data and oil crashing will make CAD weak which will help your USDCAD longs. Also, CAD index historically down for all of Jan.

I am testing out a basket trading idea .... CAD short basket up 4% overnight (this demo is scaled to $10k act):

http://i.imgur.com/7O6iPyG.png

Regards,
EZ
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot2.png Size: 123 KB
  • #628
  • Edited 3:00pm Jan 9, 2015 2:37pm | Edited 3:00pm
Quoting Graviton
Here's a 4-Lot trade to think about. There is a lot of strategy built into this trade. Initial entry was small for 0.2 Lot and 18 pip risk. This is a test. If it goes against me I call it a bad one and bail out quickly. If it goes in my favor more than spread cost, about +5 pips, I call it a good one and quickly put on another lot for 0.3 lot and 18 pip risk. This is because the spread cost is what imposes a negative expectation on a trade. It's the "house advantage". Once spread cost is covered, most of the negative expectation is out of the trade....
I'm just reading the first few pages of your post and this is just EXCELLENT material... Exactly what I'm looking to do with my Millipede structure to cure the short term drawdown blues while building those long term legs. Thanks for detailing this!!

EDIT: I just re-read this post. This is just so amazingly STUPID SIMPLE. I can't even begin to believe I've never seen this potential before. No more trading like a scared rabbit and getting OUT of a profitable trade IN FEAR when the market is moving in my favor. THANKS AGAIN for this HUGE paradigm shift...
  • #629
  • Jan 9, 2015 9:42pm Jan 9, 2015 9:42pm
Hi Graviton,

First of all, thanks for sharing your knowledge. I find it immensely valueable so are traders who stop by and ask questions. I have been reading your thread for a while now, haven't done just yet. There's so much to catch up still ! However thought I'd start asking questions instead of sitting on the sidelines.

I have been using your method, I've tried it out mostly on lower TFs. Indicator wise, I use the exact same. 5ema, 10ema along with RSI (9). Thus far I have had more losers than winners. The only few positions that were able to survive the initial few hour or so ended up hitting SL when I tried adding another lot to it. I did catch USDCAD long signal a day or two ago but ended up taking another signal and came back to find USDCAD had a good run while my position wiped out quickly.

With trial and error I have noticed few things that may improve the probability of the signals; 1. Instead of taking the technical signal right away,I would observe how price would react on minor S&R levels relative to TFs that I'm trading . When price creates a new high or a low breaking minor S&R relative to that TF I find those trades that are in line with my technical signal has a higher chance of producing higher probability trades.

I have often taken signals at the minor S&R levels only to find out my signal being invalidated at those levels. And second off, I find lower TFs contain more chaos than order thus signals that may seem legitimate may become invalid on higher TFs. For instance I have been trying to go up TF hr1 preferably. Which TF do you trade ?

The indicator itself could work as a decent scalping strategy, but building a position that will survive a day to a week to months and years is proving to be no easy task and would require one to be extremely patient.
...
  • #630
  • Jan 10, 2015 5:15pm Jan 10, 2015 5:15pm
Quoting Graviton
We've discussed trends a lot here in the past few days. I dearly LOVE long running trends that run for months or years. All the big successful traders I know of trade that way. They don't get shaken out of a good trade on a small dip following some stray bit of news. They "build a position" over time and ride it to riches over months or years. We've discussed how to find a good trend and drilled down into market dynamics a bit to examine the differences between a "real" trend driven by underlying fundamental factors and an "apparent" trend which...
I am reading the entire thread (great stuff by the way!) and wanted to test out this idea of an "apparent" trend vs. "real" trend. F

I used R (an open source statistical program) to do 1000 flips of a fair coin (50:50) ... +1 or -1 being the result, then added the result of the flip to a previous running sum. Below are 5 screenshots of this. It looks like a currency pair! Sometimes ranging, sometimes trending.

This is pure random movement ... next I will test with a 'fundamental bias.'
Attached Images
  • #631
  • Jan 10, 2015 5:22pm Jan 10, 2015 5:22pm
Now here is the same analysis, but lets say there is a fundamental bias of an upward trend, so now the probability of 1 (up) is 0.55, and probability of -1 (down) is 0.45. Nice uptrends!

This supports the idea that we want to have the fundamentals behind the reason for our trade! :-)

EZ
Attached Images
  • #632
  • Jan 10, 2015 6:02pm Jan 10, 2015 6:02pm
Quoting ezcurrency
Now here is the same analysis, but lets say there is a fundamental bias of an upward trend, so now the probability of 1 (up) is 0.55, and probability of -1 (down) is 0.45. Nice uptrends! This supports the idea that we want to have the fundamentals behind the reason for our trade! :-) EZ {image} {image} {image} {image} {image}
Wow this analysis completely blew my mind - and here i am sitting in front of my desk thinking about how to maximize my profit.

An uptrend only requires us to slightly tilt the scale in our favour - not even 60%.
  • #633
  • Jan 10, 2015 8:47pm Jan 10, 2015 8:47pm
Great thread Graviton and a marvellous post ezcurrency. This forum is such a big beast and I am not sure why I didn't see this thread earlier.

Your great examples ezcurrency clearly demonstrate the power that a subtle component of non-random bias has on an otherwise random price sample. I have been interested in this feature as on the long term timeframes, this bias cumulates to have a powerful effect on driving the longer term price moves. I suspect that this bias is not linear and is actually compounded due to the cause and effect nature of participants responding to economic announcements.

.....but how to detect when non-random bias is present in an emerging trend to distil between a 'fake trend' of randomness and a 'real trend' with built in bias? That's a tough one to crack but I think the clues lie in looking for price moves of extended directionality in the longer term timeframes of say W1 or MN. The 'trend is your friend' is always our backstop however as if there is some small bias drift in the longer term trends, then as trend traders we are going to benefit from this bias if we use the longer term timeframes as our guide.

Furthermore, having an understanding of fundamentals is going to help us particularly the effect of how serial correlations of similar economic news appear to have a compounding effect to keep momentum in our favour. Also having a clear look at the price history during a trend may also offer clues as to whether a non-random driver is present. For example, I use MTF analysis to determine the possible significance of the trend and a standard deviation channel as a tool to assess price history during a trend. Visually observing price history during trending phases is used to assess when initial volatility in trend begins to reflect more ordered nature, and that's when I pull the entry trigger. I might be reading tea-leaves here however but so far, it seems to be working wonders for me.

Anyway, once again thanks for this very useful thread. :-)
  • #634
  • Edited 3:29pm Jan 11, 2015 9:53am | Edited 3:29pm
Hi Graviton,

I got a suggestion from the other site you were before, to come here and read your thread, because it would suite to my trading style. I am an active member of that other site. I read your whole thread, from the very first post until the last. I started last December and read strictly 20 posts/day so I can think about the issues and digest it all I think I have a good overview how you trade it and what is your methodology. Hopefully I did not miss out any important points.

Along the way I wrote down all my questions, thought about them and when they were answered until the last post then I just deleted them. I am done with reading and will post my observations which were not fully discussed, not answered or maybe not even mentioned.

I decided to post in every one of my posts only one issue/question of mine because if I post 10-12 in the same post then maybe some of them will get lost. So we can move slowly. There is no need to hurry. You asked often the readers for discussion so we might have one. Shall we start? Here comes the first question which will definitely surprise you.

1. What is the exact goal of this thread?
So after being surprised why does someone ask you this question after more than 600 posts, here comes the explanation. You did not really explain clearly the purpose of the thread. Of course it is not difficult to see the direction of the thread but if I want to participate here then some questions need to be answered.

You usually ask for "comments, questions or discussions". Well, if you ask for "comments and questions" these show me your goal is basically to get to know how people like your thread and if they have questions about your style then you can help them to master it and get better at it. However when you say "discussion" this would mean to me that you would like to work together with the readers here to make the existing strategy better. These are very important issues to clear because without them I am not sure what kind of a role readers have here. I fully agree with you that negative atmosphere is not welcome, but I have also seen if some readers were questioning your methods, then you were not so happy to hear the critics and did want to discuss long those issues with those readers.

So it is important to clear if we are here to learn a method and ask questions about it or we are here to learn and expert an already existing strategy to an even better level.

Continue the great work and I am looking forward for some good discussions,

IBO
  • #635
  • Jan 11, 2015 3:17pm Jan 11, 2015 3:17pm
Quoting Copernicus
.....but how to detect when non-random bias is present in an emerging trend to distil between a 'fake trend' of randomness and a 'real trend' with built in bias?
Excellent question! That got me thinking ... I have two different ideas I want to test out which can help answer this:

(1) Multiple timeframe Markov Chain model using a simple two-state space model of up/down based off of the smoothed Heiken Ashi indicator. Using this, we can calculate the transition probabilities from each state to each state. Ex: trending up to trending up or trending down to trending down. We do this for all TF's ... and when the TF's most or all align (uptrend for example), we can calculate the probability of the next smoothed HA bar to be up. That is different than the price bar ... but generally as long as the smoothed HA is up, the overall price movement is up (especially over longer TF's).

(2) Using Commitment of Traders currency futures data to model trends in futures (and hence spot FX which the futures get their value from). This can be useful for trends, but also in extremes (when everyone is on one side of the market, it may reverse).

I'm working on a few other projects, but will add these to the list!

In the meantime, I've basically just looking at sentiment, and how price data reacts to fundamental data. Ex: if bad data out and the currency goes down, that is what we expect .... but bad data out and currency initially goes down, but a day or so later is heading up, that's positive sentiment, so we may want to long.

EZ
  • #636
  • Edited 3:53pm Jan 11, 2015 3:41pm | Edited 3:53pm
Quoting IBreakOut
Hi Graviton, I got a suggestion from the other site you were before, to come here and read your thread, because it would suite to my trading style. I am an active member of that other site. I read your whole thread, from the very first post until the last. I started last December and read strictly 20 posts/day so I can think about the issues and digest it all I think I have a good overview how you trade it and what is your methodology. Hopefully I did not miss out any important points. Along the way I wrote down all my questions, thought about them...
Not to answer for Graviston, but probably both. Why does everyone want to see statements, trade explores and such to figure if it's worth wild to read or participate? Why do so many thread starters who put so much time into there threads have to EXPLAIN the Purpose of the their thread to anyone? ANY QUESTIONS? So many good threads die because of the bullshit they have to endure in the process.
  • #637
  • Jan 11, 2015 3:57pm Jan 11, 2015 3:57pm
Quoting renegade
{quote} Not to answer for Gaviston, but probably both. Why does everyone want to see statements, trade explores and such to figure if it's worth wild to read or participate? Why do so many thread starters who put so much time into there threads have to EXPLAIN the Purpose of the their thread? ANY QUESTIONS?
Hi renegade,

I never asked anyone to post any statements, trade explores etc. I have been in many threads so probably I know why I ask. I asked Graviton because his answer might have an affect on what I will ask later. I do not think the question offended or criticized anyone.

Thanks for the nice invitation into the thread Hopefully G will not take the question to his heart as much as you did.

IBO
  • #638
  • Jan 11, 2015 4:13pm Jan 11, 2015 4:13pm
Quoting IBreakOut
{quote} Hi renegade, I never asked anyone to post any statements, trade explores etc. I have been in many threads so probably I know why I ask. I asked Graviton because his answer might have an affect on what I will ask later. I do not think the question offended or criticized anyone. Thanks for the nice invitation into the thread Hopefully G will not take the question to his heart as much as you did. IBO
Not my thread to invite you into to. If I would have would have been drilled like your last post, I would be very reluctant to share any more! It's free don't see where anyone contributing needs drilled for their efforts. My hat goes off to contributors like Graviton.

Renegade
  • #639
  • Edited 5:10pm Jan 11, 2015 4:29pm | Edited 5:10pm
Quoting ezcurrency
{quote} Excellent question! That got me thinking ... I have two different ideas I want to test out which can help answer this: (1) Multiple timeframe Markov Chain model using a simple two-state space model of up/down based off of the smoothed Heiken Ashi indicator. Using this, we can calculate the transition probabilities from each state to each state. Ex: trending up to trending up or trending down to trending down. We do this for all TF's ... and when the TF's most or all align (uptrend for example), we can calculate the probability of the next...
Ta ezcurrency. Nice thoughts. I am going to have a look more closely at the Multiple Markov Chain model. Thanks :-)
  • #640
  • Jan 11, 2015 6:48pm Jan 11, 2015 6:48pm
Just want to share my analysis here after learning so much from Graviton

So out of all the currencies that are "trending/ranging"

I believe that USD CAD has the best trending potential (even tho it has already trended for 120 days - 4 months)
Maybe if we can hop on and get more pips - we should.

Here's the chart for USD CAD
(The way I spot a trend is to use BB Band SD 1 on weekly and daily, if price is outside of the BB band then we have a trend)

Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot1.png Size: 84 KB


After that, I zoom into H1 chart to see if it is actually above BB bands as well.
As you can see, it is trending upward.

So I decide to enter when the market opens this week.
with a stop loss placed where the RED LINE is.

Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot2.png Size: 84 KB


So now, I wait and wait till the market forms another bottom that's higher than the one I entered. Then, I will put on Lot 2.

And I enter Lot 3 and Lot 4 in a similar fashion.

Once I Enter Lot 4 then I would exit my oldest and most profitable lot (adopted from Graviton) and take some profit.
That's my plan.

This follows the rule of:
buying with the trend
Buy low and Sell high
Minimizing my cost because I keep my stop loss underneath the support on H1

Downside:
My stop loss is greater than what I would like.
I suppose it's not as EXCITING...

In addition to USD CAD, (which is increasing because oil price is dropping)
I also agree with Graviton that EU GU are good trending currencies. Espeically when ECB is thinking of quantitative easing.
I am paying attention to AJ as well in case there are opportunities to stack on that trend.

Happy Trading.


Eric
0 traders viewing now