Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

building multi-pair multi-timeframe scanner 47 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

Heg G,

a question on money management. What is your maximum "actual" (means in use) leverage for your overall position? And how much percentage is your SL when you open your position? 1%?

Thanks,

IBO
Quoting IBreakOut
Heg G, a question on money management. What is your maximum "actual" (means in use) leverage for your overall position? And how much percentage is your SL when you open your position? 1%? Thanks, IBO
On average my sl risk on any trade is about 0.5%. But I'll usually cut a loser off at less than that. 1% would be a large risk for me, and I try to not take large risks. The goal is to get rich slowly. The other option is to go broke quickly. Sadly, getting rich quickly doesn't seem to be an option, at least for me. -G

Edit: Oh! after I follow the procedures outlined earlier in my thread, taking off winning lots and putting new ones on, the trade becomes "free" in the sense that I have already taken off more profits than I am risking in total sl. At that point I may have 2 or 3 lots running at the same time. So even though the combined sl risk at that point in time may be 1% or even 2%, the overall trade can't be a loser since I have already taken off more pips in profit than is risked in the combined sl. We can discuss that more if there are questions.
Hi G,

thanks for the answers, they make sense. Hmmm I believe 0.5% could be useful, but only if someone has really large accounts. You might qualify for that. I agree, this is a slow getting rich game where hard work is rewarded.

I have 1 question to discuss. I tried yesterday your above mentioned entry technique on GA short. It was just great. So of course it is hard to say if I had luck wiht it or it really works good in the long run. Can you tell some of your experience and suggestions with this specific entry technique? I believe you chose it because all TFs line up into your direction. But, that would suggest me overbought conditions in many TFs I believe, on the other side it shows a pretty strong trending condition. This made me thinking what is the usual reaction from your trades when you enter a trend on this entry method (although you did not actually mention if you use it). This would be interesting to discuss as you might have many years of experience with it and can share some great tips.

Thanks,
IBO
Hi G,

I guess today is not the best day for trend followers!

So it would be a great time to ask how exactly you handled the situation with the existing positions? Did you get out at BE or did the gap increase your pre-defined risk level? Did the events change your overall fundamental bias on EUR and CHF pairs? What is your risk management tool for handling such situations? For gaps it is most likely impossible to be prepared in most cases.

Cheers,

IBO
Quoting IBreakOut
Hi G, thanks for the answers, they make sense. Hmmm I believe 0.5% could be useful, but only if someone has really large accounts. You might qualify for that. I agree, this is a slow getting rich game where hard work is rewarded. I have 1 question to discuss. I tried yesterday your above mentioned entry technique on GA short. It was just great. So of course it is hard to say if I had luck wiht it or it really works good in the long run. Can you tell some of your experience and suggestions with this specific entry technique? I believe you chose it...
Hi IBO. I've done well with this technique overall. Of course, it's possible that price is at a maximum overbought position just at the moment of entry with this method and price will reverse right after entry. All trends must end and a small number will end right after entry with this method, but most wont. It's probably impossible to predict which will reverse right after entry and which will continue the trend for 100's of pips of profit. The way to handle that is to exit with a small loss if price seems to be reversing right after entry. We never want to let these trades go too far against us since if it's the beginning of a reversal (which occurs about 1 in 5 times) it can go against us hundreds of pips, or more. Even if we get scared out of a good trending trade with a small reversal, we can always re-enter later and often at a better price. Generally, this method either works wonderfully or fails miserably. If it fails, and it will sometimes, it critical to take a small loss quickly. If it works, ride those winners and add to them in a very conservative and reasonable way. -G
Quoting IBreakOut
Hi G, I guess today is not the best day for trend followers! So it would be a great time to ask how exactly you handled the situation with the existing positions? Did you get out at BE or did the gap increase your pre-defined risk level? Did the events change your overall fundamental bias on EUR and CHF pairs? What is your risk management tool for handling such situations? For gaps it is most likely impossible to be prepared in most cases. Cheers, IBO
Yes, some big reversals today. I had done really well over the past week and so last night just before bedtime I took off most profits and tightened up stops on the trades I let run overnight. That is just a habit of reducing risks any time I can. I didn't have any problem with gaps this time, but I've been hit with them before. I did hit some stops while I was sleeping, but I only gave back a small portion of my profits. Overall, my trades were very profitable.

Typically, I'm running tight enough stops while I sleep that I'm out of reversals very quickly. I don't usually hold over the weekends, so I avoid those gaps. I have listed my risk management tools previously in this thread, but generally I will risk a small percentage of my account on any single trade and I keep sufficient free margin so any reversal will not generate a margin call. Basically, I am prepared for a complete reversal of ALL my positions and some slippage at any time, because the market can do anything at any time. If you look back after any big loss, you'll realize you could have prepared yourself ahead of time and avoided it. I've looked back at more large losses than I can name and I've tried my best to learn how to prepare for such reversals in fortune. It's worth thinking over, a lot. If we go back and look over our trading logs and remove all the large losses, we might find we're pretty good traders after all. My goal for next year is to take no large losses. Best success. -G
Hi Graviton and other readers,

First of all, great thread! I was referred here from the TMS thread and I have been reading all your posts with a lot of interest. I'm trying to digest everything and still have 20 pages to read. The power of your method seems to lie in the fact that it can be implemented in a lot of 'systems'. I would like to find a way to do this.

One of the many posts I found interesting was the following:

Quoting Graviton
Continuing our discussion on improving our r/r ratio. A third risk to greater profits can be described statistically. The longer we stay in a trade, the more likely we are to experience a spike in price that will drive us out. I'm not talking about a major reversal due to fundamentals changing, like a central bank changing from lowering interest rates to raising them. I'm just speaking of the run of the mill news spikes that cause price to dip before it recovers and returns to following the main trend. In a week perhaps you would have only two daily...
I found this a very interesting statement. Do you have some kind of references on this topic? I'd be very interested to read more. But I found this a bit counter-intuitive. I would expect the >2-sigma events to be larger in the direction of the trend? I have been doing some own research also and I'll share when I have finished. But perhaps you are more referring to the time? The longer you are in a trend the bigger the chance becomes that you will have a large move against you?

I will definatly be following this thread and hope to contribute in the future!

Best,
Quoting tibovwh
Hi Graviton and other readers, First of all, great thread! I was referred here from the TMS thread and I have been reading all your posts with a lot of interest. I'm trying to digest everything and still have 20 pages to read. The power of your method seems to lie in the fact that it can be implemented in a lot of 'systems'. I would like to find a way to do this. One of the many posts I found interesting was the following: {quote} I found this a very interesting statement. Do you have some kind of references on this topic? I'd be very interested...
Hi Tibovwh. Excellent question! I agree that larger spikes are more likely to occur in the direction of the trend, but as we stay in a trade for longer periods, we get more larger spikes in both directions. Of course, those spikes in the direction of the trend, and hopefully also in the direction of our trade, are windfalls. We'd be wise to take those profits quickly and think about it later. But the spikes against our trade are also getting larger over time. Eventually, one of those will probably run us out of the trade, regardless of the size of our stoploss. That's ok though if you have a healthy r/r ratio and W/L percentage. You'll take some losses that can't be prevented, but the wins can be greater. In the end, it's a business of probabilities. We want to be the House Casino that loses some small bets quite often, but wins large over the long term because we have set the odds in our favor. Unfortunately, I don't have time at the moment to look up the data analysis, but I have done so before and the occurrence of the spikes roughly follows the bell curve or normal distribution over a long enough period of time. I'd be happy if someone else reading this thread could do that bit of research for us all. Best trading success to you! -G
Quoting Graviton
{quote} Hi Tibovwh. Excellent question! I agree that larger spikes are more likely to occur in the direction of the trend, but as we stay in a trade for longer periods, we get more larger spikes in both directions. Of course, those spikes in the direction of the trend, and hopefully also in the direction of our trade, are windfalls. We'd be wise to take those profits quickly and think about it later. But the spikes against our trade are also getting larger over time. Eventually, one of those will probably run us out of the trade, regardless of the...
G,

I have done some crude research on this topic, trying to dig deeper. A first problem really seems to be: How do we define a trend?

I took a sample of daily returns on 15 pairs for 15 years. The full sample of observations shows returns have large fat tails and don't resemble a normal distribution. When defining trending as being above or below the 10,50 and 200EMA we can plot the returns again. It shows us that a trend-following approach sets us up with a negative skew. Essentially, we're having a larger number of small wins it seems, but we're keeping the tails.

So, I don't really agree that the occurence of spikes resembles a normal curve over a long enough time. I think the only way to do this in by increasing our temporal aggregation to Weekly or Monthly returns. Then it would resemble a normal distribution more?

Best,
Attached Image(s) (click to enlarge)
Click to Enlarge Name: Schermafdruk 2015-12-14 15.57.51.png Size: 16 KB
Click to Enlarge Name: Schermafdruk 2015-12-14 16.09.05.png Size: 32 KB
Click to Enlarge Name: Schermafdruk 2015-12-14 16.32.15.png Size: 30 KB
Quoting tibovwh
{quote} G, I have done some crude research on this topic, trying to dig deeper. A first problem really seems to be: How do we define a trend? I took a sample of daily returns on 15 pairs for 15 years. The full sample of observations shows returns have large fat tails and don't resemble a normal distribution. When defining trending as being above or below the 10,50 and 200EMA we can plot the returns again. It shows us that a trend-following approach sets us up with a negative skew. Essentially, we're having a larger number of small wins it seems,...
Interesting tibovwh, but your conclusions seem to go against the standard definition of trend following having positive skewness and more here.

PS My sample sets on trend following definitely support small losses and large wins so not sure what is going on in your data.
I'll add something a little tangential but which might be interesting to discuss. Graviton here has often talked about fundamental bases for trends, and to pick trends to trade which you are comfortable with fundamentally. That's hard to measure statistically.
My take is that maybe its the USD that drives the global currency markets, and the smaller currencies are more volatile, and are often trying to "catch up" to the USD.
If USD and Euro trend more statistically, maybe there is a connection there to these two being the largest currencies in terms of volume? If so, maybe the Chinese currency is also going to act in the same way?
Quoting tibovwh
{quote} G, I have done some crude research on this topic, trying to dig deeper. A first problem really seems to be: How do we define a trend? I took a sample of daily returns on 15 pairs for 15 years. The full sample of observations shows returns have large fat tails and don't resemble a normal distribution. When defining trending as being above or below the 10,50 and 200EMA we can plot the returns again. It shows us that a trend-following approach sets us up with a negative skew. Essentially, we're having a larger number of small wins it seems,...
Excellent original research Tibovwh. I'm aware of the effort needed to create such data and analysis. There is some deviation from a normal distribution, but generally the curve shape seems to represent the classic bell shape, at least to my eye. I suppose my original comment could be rephrased to something like, if the distribution of the size of spikes over time resembles a normal distribution, then the trader will be subjected to ever larger spikes as she/he stays in the trade for longer periods of time. Eventually, given enough time, it is probable that a large enough spike will occur to hit any reasonable TP or SL point. If you provide the post number of my original quote, I'll go back and look over the context in light of your data. Thanks loads for your work on this subject. -G
Quoting Copernicus
{quote} Interesting tibovwh, but your conclusions seem to go against the standard definition of trend following having positive skewness and more here. PS My sample sets on...
Thanks for the additional references Copernicus. If we are going to approach our trading in a logical and scientific manor, it is important that we base our decisions on actual data and analysis rather than trading myths. -G
Quoting lemna
I'll add something a little tangential but which might be interesting to discuss. Graviton here has often talked about fundamental bases for trends, and to pick trends to trade which you are comfortable with fundamentally. That's hard to measure statistically. My take is that maybe its the USD that drives the global currency markets, and the smaller currencies are more volatile, and are often trying to "catch up" to the USD. If USD and Euro trend more statistically, maybe there is a connection there to these two being the largest currencies in terms...
Yes, good point Lemma. The USD and Euro make up the majority of most currency "baskets". If we are looking at the value of the USD, we are mostly looking at it vs. the Euro, and visa versa. I don't currently trade the Chinese Yuan because it is so highly manipulated by the Chinese government, but I agree that it may become more influential in the future as it is more widely traded and the Chinese economy continues to grow. -G
With all that said, I should admit that my very best trades are those that have developed over months to years of a fundamentally driven trend playing out rather than some quick single day move. I don't necessarily hold the same entry for weeks or months, since I use the multi-lot method described in my thread to ride these trends for a long period of time. Usually I will make several entries and exits when conditions seem most favorable along the way to attempt to take multiple quick "bites", usually each lasting only a few days to a couple weeks out of that longer term trend that may run for months or years. An example would be the long term trend created by the weakening of the Yen by the BOJ that started around the end of 2012 and continued until the middle of this year. Another example would be the falling value of the CAD with the falling price of oil that has occurred over the past couple years. Even though I am trading with a strong fundamentally driven trend, I will still take some losses when my entry occurs right at a peak reversal point. Those bad entries are not completely avoidable regardless of the entry method we use, but we can limit their damage by cutting losses short and take advantage of the good entries powered by the long term trend by adding to winners and letting winners run. Best success to all. -G
Sitting on the sidelines all of today. Will be back in the markets tomorrow after the dust settles from the Fed rate hike.
Quoting Graviton
{quote} Thanks for the additional references Copernicus. If we are going to approach our trading in a logical and scientific manor, it is important that we base our decisions on actual data and analysis rather than trading myths. -G
Those weren't trading myths G. They are the very reason for the success of trend trading styles and result from applying a trailing stop and letting the profits run. The Kaminski link I referred to was a peer reviewed study resulting from extensive statistical tests of trend following across a wide variety of markets. There was something funky with conclusions of Tibovwh's hard work and like any statistical test, rather than simply blindly accepting them, I just wanted to question the validity of the underlying assumptions or the method of analysis. Tib might have something more to say about this so I will leave it to him.
Quoting Copernicus
{quote} Those weren't trading myths G. They are the very reason for the success of trend trading styles and result from applying a trailing stop and letting the profits run. The Kaminski link I referred to was a peer reviewed study resulting from extensive statistical tests of trend following across a wide variety of markets. There was something funky with conclusions of Tibovwh's hard work and like any statistical test, rather than simply blindly accepting them, I just wanted to question the validity of the underlying assumptions or the method...
Right. I understand. The Kaminski link agrees with university research I found long ago. I didn't mean to insinuate it was a myth. I was just speaking in general that I think it's a good practice to look into the facts, either statistical analysis or logical conclusions, rather than trade on rumors of what worked for someone else. No problem. -G
No probs G. Have a great Xmas mate and many thanks for maintaining a great thread on this forum. It has been an excellent resource for me.

All the best.

C
Merry Christmas and a Happy and Prosperous New Year to all. Spending time with family and friends during the holidays. Trading is usually thin and unpredictable during the holidays, so I'll be on the sidelines until after the new year. Wishing best success for everyone in the new year.
0 traders viewing now