I sat down and thought long and hard about this system: I have come to the conclusion that this system can only be traded with some human intervention at key levels for example EURUSD right now: we have a range of 1.27 to 1.31 If either of these were broken, I would say that it would be prudent to hedge 100% of your position on the wrong end of the trade And I would also say that this can only be done with a short term option (say 1 month) because if the break proves to be a false break and the hedge (a spotFX hedge) reverses on you, you will suddenly be in a massive high leverage losing spotfx position This hedging idea will also eliminate the risk of the large "supermove" account wipeout type scenario, but it will eat profits because the option is obviously going to cost money Basically what I am saying is each currency would have to be broken up into large "blocks", where if a block is broken, the risk of a strong trending run is enormous- example if USDJPY broke 115 or 100 it mean something big, and a hedge should be implemented for protection or if EURJPY broke 142 on the high side, etc etc you get the idea If anyone has any better ideas, I am all ears
Thanks Guys, Well I did the transition...This was a lot of work tonight... I am now using the new spreadsheet. Please check in "now and then" to see how it is going. I am tired.... Good night.. Michael B.
Ezekiel, You are correct. And then if you keep going with that and it retraces back to the start. You end up closing all of those trades you left on the way up to create another 160 pips, but you are still -450 in the unrealized because you left a new trail on the way back down. This leaves your realized at 320 and your unrealized at -450 for a total change in NAV of -130 pips. So you need more than just one retrace (or a lot of small ones along the way) to get into the + NAV. This and the fact that the unrealized loss increases exponentially in a trend is what concerns me with this system. If we could just figure out a way to capture the swings without the potential exponential loss, we would really be on to something. Possibly playing around with the grid spacing and take profit ratio we could improve, or using something like the inverse grid (but I think it needs more work). Any way, there are some of my thoughts after trading this live for only 4 days. I think there is a lot of potential here, but even though is seems so simple on the surface it is really a very complex system. David p. s. ElectricSavant, Thanks for the welcome. I hope I can help in some small way. Quote from Ezekiel: Davidwillis Considering what you said below, something I've realized as I've been tinkering on a demo account is that say you place your buy/sells every 10 pips with 20pip targets like you state below. If it moves a straight 100 pips then you should have profited from 10-80 with the 90 and 100 pip order still open. This should total 160pips realized profit, right? 8 orders that hit targets. But on the flipside your not just down -100pips, because if you truly buy for every sell and vice versa, this would put you down -90 for the 10 order -80 for the 20 order -70 for the 30 order .... -10 for the 90 order which totals -450 pips. Obviously, if it retraces these orders will profit, which is what this system hopes to capture. This just made me realize how important the spacing/incrementing of orders and order size is.
i've been following this concept since the interesting trading style thread at mt, the posts at oanda and now here. i did an off the cuff backtest from jan 2003 to jan 2004 on eurusd daily data. this system made 90% realized and about 50% unrealized. ie with a $10 000 acc that's $9000 dollars realized and $5000 unrealized. i did that by trading $2500 lots. here are some ideas. the idea of using a pip increment spacing can be improved upon in my opinion. what about a time based entry system. ie make your trade at a certain time each day. that way when the market moves fast your "spacing" between trades is larger and when the market moves slower your "spacing" will be closer since the market hasn't moved much since the last trade because it is slow. this concept also has to be developed because i'm not going to let the cat out of the bag completely. another thing is that increasing your trade sizes as your realized grows is critically important. because what this does is make those lagging trades at the bottom of the pool less significant as the trades at the top of the pool are bigger. hence your realized grows faster than your unrealized. i tested the same data without increasing my trade size as my realized grew and the net result was 30% realized and 30% unrealized. a key thing here is how you increase your trade size. i won't reveal how i change mine but suffice to say that you should spend some time on this. in my opinion this system is basically a discretional one however mechanical it seems. i like to think of it as trading against the move using only my discretion and never taking the loss relying on it to come back later. the more i think about it the more i believe intuition/gut feel/discretion is a real important part of this system. what makes it even better is that playing the game this way is fun because you never really have to take the loss (gasp!? did he really say that!!!???) that way i always win when i play. there is only one thing that bothers me about this whole concept and that is that you have no real edge. kinda like playing a game of martingales. only other thing i can think of is es's statement that flow>capacity is infact the edge. faure ps. diversification is another concept that requires attention beyond what is given to it in this thread. "two are better than one, because they have a good reward for their labour. for if they fall, the one will lift up his fellow. but woe to him who is alone when he falls and has not another to lift him up." ecclesiastes 9-10
faure... I don't know who you are....your wisdom is familiar.. Michael B. P.S. You would not believe some of the PM's I am getting, from people I would be honored to meet.
Todays Trades (Monday) First period of six currency trading, this is hopefully the first of many daily spreadsheets. Many little bugs fixed in the spreadsheet. Let me know if there can be any improvements. see attachment oops had to repost... mistake in carry unrealized
Last night at 3:00am I couldn't sleep, so I started thinking about this system (or was it that I started thinking about this system so I couldn't sleep). Any way, the hardest part is filling up the pool right? So why wait months for the price to move all over the place to pick up all the trades? Why not just look at the last 12 months to determine the high and the low, then buy all the longs and all the shorts at the current price. Then you just need to set sell orders at every grid space. You would have your entire pool filled (at least the last 12 month worth), and would have better priced that if you bought them as you went. Let me give an example, using EUR/USD and 100 pip grid spacing. 1- Look at the last 12 month high and low and current price. H = 1.3670 L = 1.1759 current price = 1.2850 Note: if you had been following this for the last year you would have short buys all the way from 1.1800 up to 1.2800 at an average price of 1.2300 and long buys from 1.3600 down to 1.2900 at an average price of 1.3250 and your net unrealized pip p/l would be 5500 for the shorts and 2800 for the shorts for a total of 8300 pip loss (this is just an approximation because your lot sizes would probably change over the last year). 2- Buy 10* (your lot size depending on your account size) Shorts and buy 7*(your lot size) Longs. Now you have the same positions as if you had been filling the grid all year, but at a better price (although you do not have all the gains from the year). 3- set targets to sell one lot size every 100 pips. 4- when a target gets hit fill in the hole with limit orders like normal. Maybe I am just tired from lack of sleep and am missing something. David
Faure, Obviously the use of Pivots calculated every day to draw lines (grid spacing) with is not the best use of time. The thinking was to capture the "natural flow". But, there are other ways to capture flow through "floating" static, fixed increments spacing from simply keeping track of the the 1Y range. This is done now in this "Forward Test" and called "minumum pip increment". Why use the Pivot Points at all then? Just use the minimum calculated from some method of accounting for # of entries with a rolling forward 1Y range. Use wider increments so that sleep and multi currency management is possible. You will notice that I have posted about the one hour overlap of the markets has a higher volatility. So there are three opportunities to set the alarm clock by, to put in 10 pippers between the "normal" time grids discussed in the above paragraph. All the big news release's and even smaller news release's and reports have not been discussed here in this Journal yet. This is "alarm clock" setting material, too. Then FRIDAYS from 6:00 est up until 15:45 est of course when everybody is getting flat, represents great extended high volatility! Faure, I appreciate you comments, but please don't start this "cat out of the bag" crap. This is not a cryptic hint giving Journal. So I am not going to spend my time picking your brains, when you are clearly intending to speak cryptically and not let the "Cat out of the Bag" So Faure or Sympatico or whoever you are, you are welcome here, but this is not the place for game playing. (now, I know this statement pisses you off, but I have "printed", yes... printed your one and only post to this Journal, so far) Michael B. P.S. The inverse thought might be, What if slow times were wide and volatile times were TPA's (ten pipper alerts)
FOLKS, WE WILL DISCOVER THESE HIGH VOLATILITY TIMES, I PROMISE YOU. I WILL PUT IN 24 COLUMNS REPRESENTING ONE HOUR RANGES AND HIGHLIGHT THE CELLS "DURING NEWS" WITH ANOTHER COLOR.