ed_inacloud

Hi, I've changed my email address in UserCP but still get notifications to my old one and not the new. Can anyone suggest a resolution? I can see that someone had a similar problem last year but there wasn't a reply to that. Thanks

Good point equtrader, which I think is what the OP was trying to do in the 1st place. To the wider audience......I do understand the sceptics of market randomness, I was one for many years until the evidence proved otherwise. What I dont understand is what they have to gain by rubbishing the views and flaming the posters. I appreciate that those making money from mug punters need to keep as many in the market as possible but hopefully we don't have too many of the former on the forum. Now what would be useful is that if those that had a similar view about market randomness were allowed to share findings to see if it really is possible to beat it. Challenge it by all means but in a constructive way. I for one would like to be proved wrong, if there's evidence that it's not random then post it and help us all out. Ed

Below is the simple code to show the randomness of the markets, it looks at the number of contiguous bars in each direction. Cut and paste it into ProRealtime or any of it's derived chart packages. Put the results into a excel spreadsheet and create a graph, you'll see the Normal distribution pattern. I only trade € / $ and has been tested on multiple time frames with the same pattern, also plugged it into UKX again with same result. I'm not a programmer but have developed many indicators for my my own analysis, It may not be elegent or optimised but it does the job. That's my evidence for randomness, now what was the evidence against again? REM start points c = 0 if open > close then a = a + 1 if a[1] = 0 then c = b b = 0 endif endif if close > open then b = b + 1 if b[1] = 0 then c = a * -1 a = 0 endif endif if c = 1 then v1 = v1 + 1 endif if c = 2 then v2 = v2 + 1 endif if c = 3 then v3 = v3 + 1 endif if c = 4 then v4 = v4 + 1 endif if c = 5 then v5 = v5 + 1 endif if c = 6 then v6 = v6 + 1 endif if c = 7 then v7 = v7 + 1 endif if c = 8 then v8 = v8 + 1 endif if c = 9 then v9 = v9 + 1 endif if c > 9 then v10 = v10 + 1 endif rem negative values if c = -1 then v1a = v1a - 1 endif if c = -2 then v2a = v2a - 1 endif if c = -3 then v3a = v3a - 1 endif if c = -4 then v4a = v4a - 1 endif if c = -5 then v5a = v5a - 1 endif if c = -6 then v6a = v6a - 1 endif if c = -7 then v7a = v7a - 1 endif if c = -8 then v8a = v8a - 1 endif if c = -9 then v9a = v9a - 1 endif if c < -9 then v10a = v10a - 1 endif v11 = v1+v2+v3+v4+v5+v6+v7+v8+v9+v10 v11a = v1a+v2a+v3a+v4a+v5a+v6a+v7a+v8a+v9a+v10a //mttav = v1 / mtt //mttch = mtt / (mtt+v2+v3+v4+v5+v6+v7+v8+v9+v10) return v1 as "1", v2 as "2", v3 as "3", v4 as "4", v5 as "5", v6 as "6", v7 as "7", v8 as "8", v9 as "9", v10 as ">9", v1a as "-1", v2a as "-2", v3a as "-3", v4a as "-4", v5a as "-5", v6a as "-6", v7a as "-7", v8a as "-8", v9a as "-9", v10a as ">-9", v11a as "-tot", v11 as "+tot"

So is the suggestion that someting that is random is not mean reverting? Random behaviour is normally associated with binary events (mean reverting around zero) and mean reversion to values. A bar on a chart comprises both. As soon as you take a postion then you set your own zero level, your decision before that is up or down ie binary. Thats the random bit. Averaging in resets the zero level which improves the chance of get back to BE but increase risk of greater loss. It's a Normal distribution curve, there will be outliers / black swan events that will cause great pain on occasion.

1a, as I've said before I agree with most of what you say (including mixed long and short positions) but the diagram you show doesn't make much sense to me. I can relate to the the averaging in on the long side but the combination is confusing, if you had the 1st short then why would anyone put in a counter position rather than closing the 1st? Also if we could identify any one of those shorts positions then we'd be sitting in the Carribean now (unless you are of course ). I'd be interest to know how you would actually trade that chart, personnaly if I'd taken the 1st long as the opening then I would average in to improve my opening and possibly taken a counter short position on the break of a previous low (but closed that in place of opening a further long).

1a, I fully support most of your view and have trodden a similar path except that I moved from stocks to indices to forex as the spread is more consistent through the 24 hr period. I too have a well funded account and always make sure I never put myself in a position of being wiped out. I happily add to losing positions in order to bring a trade back, averaging in is not uncommon. However to the point that some are making I was long €/$ @about 1.4 just as Greece was kicking off, I did add to the position when things looked as though they were quietening down but the trade kept moving against me. I also took equivalent short positions to limit the drawdown for periods. I did end up cutting the trade for about £12K loss about October late last year, my averaged in position was 1.36 . If I’d stayed with it then it would have come back. However a lot of the entries were based around feel. I have tried to formalise a strategy with set points for averaging in and profit taking. I found that within the parameters I set I could get 88% chance of success or stop at break even, of the ones I let run they too would break even 88% of the time. Unfortunately the bad trades <2% of the time had the potential for massive drawdowns as with my €/$ 1.4 entry. I support your comments on randomness and have proven it conclusively to myself, as I’ve commented on your other thread it needs to be embraced, understood and worked with in order to be successful.

Well done 1a, the more times people start threads with their experiences then the more chance that newbies and strugglers will have to see that they are far from alone. I have been trading (well gambling) Forex for about the same time as 1a, since 2008 and have looked at every major indicator and the few strategies that are put out there. I have developed many of my own and have even tried trading them from both sides but everything loses in the end, primarily because market either ranges or is in a trend and entry points are 180 degrees apart for either. Randomness is very easy to test for, there are only two aspects of bar formation that is direction and size. A simple indicator that counts the number of contiguous bars in any direction will show that you will get a Gaussian or normal distribution as a result. It’s random, no arguement. Even I was surprised to find the same result when looking at a 500 pip up move in €/$. If you look at the size of the bar then the same curve is found although it does move slightly in favour of the direction of any move, but now you’re thinking about how much profit or loss you’re going to take whereas your primary decision is which direction (ie the random bit). Again a simple indicator to find how many bars hit certain levels is easy to put together. To the left of the chart shows lots of reason why it cant be random, but as we all know we cant trade that bit. It does raise the question as to why it appears one way but is clearly not. To my mind the markets move on far greater aspect than we try to see in the detail of the charts. Moves are based around Sentiment tempered by News which if great enough can reverse sentiment, although even there an element of randomness exists as shown this week when German PMI was down and shares and the Euro rallied (and please don’t give me the ‘built into the market bullshit’). Now consider that we all take sentiment as our decision on direction then that would be too easy to make money and so is complicated by randomness, as I say you can prove it for yourselves. Now as to how and why randomness occurs is up to you, I’ve spoken with LIFFE and LSE and was advised that the control algorithm would never be released and am happy in my own skin that there is a central control. There are two easy ways to make money in this game, either start a spread bet company or sell your trading ideas to the gullible. If you do either then you need keep as many people in the game as possible....and get rid this crazy idea that the markets have a random element. If you’re going to trade successfully then accept the randomness arguement as an element within moves and find ways to identify when it’s not in control of direction or make sure you set stops wide enough to take account of it. If you’re new to this thread then you’ll see lots of people before saying markets cant be random because they are successful individuals, not much of an arguement really. I’ve never really understood what people gain by claiming undocumented success other than trying to show others that whatever struggles they’re having they should persevere. If you really want to share your success then show how others can benefit from it, otherwise it’s those nasty SB co’s and Trainers that take the money. Well done 1a for sticking your head above the parapet.

The thing is Savant if you leave it long enough then everyones guess would probably be right, especially in trading as long as you have a system and as long as you can afford to do so. But there will be the couple of times (period high and period low) when you're not. If you had limitless amount of money I can give you a system that will guarantee results even in a random market. Trouble is non of us have the funds to trade it (and no I'm not trying sell anything, it's decribed earlier in the thread) So, getting this random stuff back to our (well yours at the moment) trading environment. If it is truely random ( as I believe I can show) then why do the trades feel as though they're under pressure so much of the time?

Hi SIYUA, I dont generally get involved in forum discussions as I seen too many threads end up with varying views leading to arguments and worse. I accept that clarity is writing is important, but interpretation in reading is too. My quote you include above is incomplete and implies that I'm saying that the chance of flipping a tenth head after nine previous is small, whereas what I actually wrote was "I would disagree that the chance of flipping a tenth head after nine previous are small, it's 50/50" I dont really see that as being too different a statement to the one I supposedly corrected it too. I will apologise if I confused anyone by introducing to possibility of 10 following 9, but I'm pretty sure it's correct... I've counted it on my fingers halve a dozen times (6) now and 10 keeps coming up after 9. Anyway, here I am perpetuating an argument thats becoming irrelevent to the thread (which is why I supposedly dont normally get involved) :doh: For the record, I'm not saying I'm correct in my views but just sharing my experience, findings and thoughts. People can either ignor them, discuss them or take them and try to build on / destroy them as they see fit. Cheers Ed PS I only responded to your comment 'is there a prize for flipping a tenth head only after flipping 9 in a row ---- I only ask as such a run is only a 50/50 chance....right?' because you appeared to understand the single action following the previous sequence but seemed to be questioning the odds and implying a single run of ten flips. If that wasn't the case and I mis-read that then I again apologise... but then again if you'd written a little more clearly......

Don't think anyone is saying a run of ten is 50/50, I'm sure you don't need me to explain it's 1/2^10. But because it's 1024:1 against flipping 10 in a row doesn't mean its going to be 1024:1 on flipping the 10th after nine previous. It's 1:1 or 50/50. If you want to place the 1024:1 on bet for the final flip when you did get nine others in a row, drop me a line and I'll take you up on it (and I'm not a gambling man ..... which is why I'm sitting on the sidelines trying to disprove my growing market randomness concerns). When I started trading a colleague said I was a fool, the market is random. I wish I'd listened harder and investigated price movement rather than trying to find the elusive indicator. I just hope someone starting out reading this thread doesn't make the same mistake...which I believe was the original goal of the thread. Just glad I could contribute

Thanks BlueHorseshoe, one of your comments has sparked a light. So for anyone that wants to listen...... There is no gausian distribution in the results that I have for size and frequency of moves, it's a ramp. The lower the value of the move the higher the frequency, with a pretty good (but not perfect) relationship between measured values and predicted frequency. As I said earlier the I believe the results I have are a long winded way of doing an FFT analysis on the data. The ramp is what I'd expect from white noise or random data, the question now is how is the random nature generated. 1) Is it just randomness created by so many participants with varying views. In which case in extreme news conditions the random nature should disappear. I need to go and do some more analysis but should be easy to spot. 2) Is it created by the Liquidity Generators (and conspiracy theory or not they are there, CME trade a huge futures market and have a published list of Primary Market Makers). I need to better understand their MO before I can assess that. 3) Is it an underlying Algorithm? Will only find out by ruling the other two out. But could it be done and how would they do it will be an interesting arguement to have. I have two analogies for could it be done. a) I want to take a train from Portsmouth to London, I can buy my ticket from a number of places and could use a couple of operators depending on route I wish to take, so sounds a competitive scenario to hit my goal of getting to London. Unfortunately the track is controlled by Railtrack and they can disrupt my journey at any point. b) Oil is a naturally occuring resource found in many parts of the globe. The producing countries (that in a normal market would be competing) have found that by controlling the flow and so price they can maximise profit. So dont be fooled into thinking competing exchanges wouldn't work together if it's their interest to do so, I'm not saying they do but there is no reason that they couldn't. How could they do it (well how would I do it if it was mine to control)? I would modulate the natural order flow with a pseudo random algorithm, that would account for the anomolies I've seen on the ramp discussed above. It would need to be sophisticated as simple Amplitude or Frequency Modulation would be detectable and beaten. I'd adopt something along the lines of an FHSS model for security. So how does this help us? I dont know yet. But lots of people come to these boards looking for how to get an edge and are all dissapointed because there is no-one out there saying here you go, this works, use this. Just the usual comments that no-ones is going to hand it to you on a plate and you have to work to find the answers. Well my comment those people looking for the answer is be careful, there may not be an answer. I think I'm reasonably well quallified to give that advice now. I've looked at price movement for four years and from all angles now, I've stripped it down to the bone and taken some off the wall approaches at it. Money mangement is not the answer that will get you results, using it will mean you wont wipe your account out so quickly but you will wipe it out in the end. If you use a balanced ratio of profit and stop/loss method it's a 50/50 gamble from where I'm sitting at the moment. If you move the ratio in favour of less than 1:1 then you'll win more often but the greater losses will kill you. I developed a strategy that will win or break even 88% of the time, if you run without stops the loss scenarios will come back to BE 88% of the time, so less than 2% loss. If I ran it real time it would wipe me out and while it was wiping me out I'd be paying significant overnight charges to the broker. This is a very simple game we are trying to play, all you've got to do is find the right entry and right exit. Good Luck;) Ed PS Am I just a beaten up and twisted loser that's blown his account? I don't think I am. I've 'invested' about £20K or roughly 12% of my account. I've had times that I've felt I've made money but have never been able to draw profit from the account. It keeps me occupied and I'll keep looking for answers.

siuya, I'm familar with arbitrage but didn't appreciate that it might be carried out between the index and it's underlying shares. Spotting opportunity between two tracking entities is supposed to be hard enough, trading 101 shares with varying weighting that are moving in all directions in order to bring them back in line with a moving FTSE Futures contract would be tough for a single entity. Having multiple entites in the mix and holding a tight relationship sounds impossible? I assume these entites are not liable for SDRT when they trade FTSE shares, that 0.5% would be a killer in that environment. I'm really not sure who is and isn't liable but just know I hated paying it before moving to CFD's and then SB. Do I think it's a global consipiracy? I thought it was a well documented conspiracy at the exchanges by the Specialists / MarketMakers Anyway I've hogged this thread enough, seems to be only 3 or 4 of us failures in the world, so much for 97%

BlueHorseshoe, I appreciate your views ....it's good to get some balance to arguements that I try to make to myself in this lonely vacuum we choose to work in. I've looked at random strategies and would agree that outcomes of consecutive coin flips are probabilistically independent, although I would disagree that the chance of flipping a tenth head after nine previous are small, it's 50/50. The coin doesn't remember whats gone before. There's been a couple of good science programs here in the uk in recent years that supports that. The problem with waiting to trade an outlier in a bell curve is that it takes for ever for the opportunity to arrive but it does it's still 50/50. I had a eureka moment a few years ago when I litterally woke up one morning and thought 'don't try to predict the market, it's either going to go up or down just be on the right side whatever'. Seemed so obvious and simple. Idea was to set a line whenever you start trading and if price crosses over it then buy, if it crosses under then sell. soon found it can't be done in a retail environment with the spreads and order restriction we have. Price can cross 30/40 times in each direction so guaranteeing being on the right side at the right time is impossible. So I set my own spread of 10 pips and to offset any loss of an incorrect direction call then the next trade was set 10 pips away in the opposite direction and doubled. As soon as it moved in any direction outside the spread by 10 pips then I would be in profit. The average number oscillations between two points 10 pips apart is 2.7 times, I started at £1 per point and doubled up all the way to £128/ point before I lost my nerve and gave up on the trade. If I'd held on then I would have been successful, but fear got the better of me. I started to believe that someone was trading directly against me... but even I can understand that the global value of the dollar isn't going to be determined by someone trying to take a few thousand pounds from little 'ole me. Another mistake shared :-) Interestly if I halve the spread I get twice as many oscillations, if I double it then I get half as many. Turns out the strategy is already widely used in the casinos, it's called the martingale strategy. It too can be easily simulated on a spreadsheet, It's not long before a simple 50/50 strategy will wipe out the average account. I do trade a pullback strategy now and is why I'm trying to develop my own zigzag indicator so I can look at optimum entry points for various timeframes (and fib points). On average there is more movement in a pullback than there is in the extension to new highs or lows, so if you dont take advantage of the pullback then you will likely be put under psychological pressure of a negative position very quickly. However trading against the current move in the expectation that the real move is in your direction is also tough psychologically. Something for readers to ponder, I looked at relationships between bars on 1m, 10m, 1Hr, Daily and Weekly. Measuring O-C, H-L, O-H, O-L on each bar and Overlap between two consecutive bars. The ratios between each of the average measurements stayed fairly consistant over all of the timeframes (and the best place to be is in the o/lap). My gut says that there is a fractal based pseudo random algorithm which is why common patterns appear across all time frames, and it's designed to put us under psychological pressure where either fear or greed will take your money. I realise it all may sound off the wall and not what someone wanting to trade wants to hear, but it's four years of studying the markets from many angles. I think I'm reasonably intelligent, logical left brained and with qualifications in Electronic engineering. I'm just sharing for people to either use or ignor. If you have a way of proving it's not random then I'm keen to learn. I have considered using FFT analysis where random white noise patterns can be identified, but my programming skills aren't up to that. I also believe the the findings I have through other means show the same patterns that FFT analysis would give. Cheers Ed PS My understanding is that arbitrage is a trading opportunity rather than linkage between two derivatives. FTSE cash is directly derived from the values and weighting of the underlying shares, FTSE cash cant be traded as far as I'm aware and anything claiming to be is actually derived from and hedged in the futures market. FTSE Futures is FTSE cash + Fair Value and am advised the link is updated through the trading day. What I cant get explaination to is how market driven moves in the FTSE Futures feed back to FTSE cash value (as they claim there is no feedback link) and if there was a link how any changes in the FTSE cash is distributed back down to the underlying shares.

Marc, If patterns work for you then great. Patterns are clearly there in all timeframes and I looked long at hard at some of the popular ones such as wedges and H&S. I could spot them easily after the event but to trade a developing pattern didn't work for me. I gave up when I developed a spreadsheet to generate four interlinked random numbers to give high, low, open and close and then linked them to the following row using another random number. I managed to generate most trading patterns in a very short time. As I said earlier I've traded E/$ since late 2008, it tends to track indicies based on $ weakness = index strength. I've hear about the santa rally but in that time the only one I've seen is in 2008 and to go long at that time based on a pattern would have been brave. Looking back it looks they did occur from 2000 to 05 and there have been 6 in the last 12 years, so 50/50? I am trying to program my indicator to look at fibs but am struggling to get the basic zigzag pattern programmed and need that to set start points etc. As outlined previously S&R does make some sense to me, certainly on the smaller timeframes where order flow seems to support the patterns, however I dont get S&R over long time periods when the financial markets are as turbulent as they are today. I keep coming back to a conversation I had with a guy at LSE, I was trying to understand why the markets moved so as to determine how they moved, I was also talking to a couple of people at Euronext/Liffe. The guy at LSE said they would never release the algorithm. Algorithm what algorithm I thought it was all order driven? Everyone I was speaking to shut down when I pressed that question, maybe 'cos they think I'm a mad fool. Does that account for the same patterns seen over all time frames....who knows. Another question I asked but never got a sensible answer to was the interlinking of indicies and their base shares. The index is derived from the value of the base shares and there is a direct link between the two which is based on fair values, the index is then also traded independantly but there is apparently no feedback of index movement back into the share prices. Anybody with any engineering experience at all (especially electronic) will see that cant be true, else people buying shares could drive the index in one direction and those selling the index could drive it the other way. So either the Index Futures aren't really traded as a separate entity (as LIFFE claims it is) or there is a feedback loop, possibly The Algorithm..... Not wishing to rain on anyones parade, and I'm desparately trying to prove my own findings wrong, but it seems to be a very well controlled random environment we seem to be working in. I'm off to take my pills and lay down :-) Ed

Could be an interesting thread if you can get people out from behind their egos, but my guess is that the result would be more of the BlueHorseshoe stories to which I can certainly relate. I've been in this for four years and have 'invested' a significant sum in the spirit of learning the trade. I trade Eur/$ In the early days tried many of the popular indicators, all of which worked when the market conditions met the parameters that the indicator was designed for but every one failed when the conditions changed from trending to ranging or vice versa. Indicators are great at telling what has been but I guess we all know that by now. I also started to write code to develop my own indicators with not surprisingly the same results, but it did enable me to develop a mechanism to track price movement and look at the odds of certain potential outcomes. At any point in time the odds are 50/50 or there abouts, the only way to improve the odds was to change the ratio of S/L to profit in favour of greater stops which resulted in more wins but not enough to offset the increased cost of the losses. The only thing that that makes any sense to me is that news is king, it drives sentiment and the market in an overall direction until the next bit of news offsets it. Most major turning points in the E/$ on the longer timeframes can be attributed to news. If you can swing trade off the back of that then well done. Like all things it's easy to look back and attribute direction changes to events, not so easy in real time. Support and Resistance is something I keep coming back to and looks like it should work based on simple price movement mechanisms, but it too is destroyed by big news events. The things that stand out for me over the years are: 1) There seems to be lots of people who are keen to learn to secrets of successful trading and a few people who will take their money to give advice or a call service. Having spoken to a number that have taken up these services, none have matched the results of the 'mentor'. Also begs the question as to why the mentor spends valuable trading time time where he could be making even more money on his own, and if he really did want to give something back then why is he charging in the 1st place. 2) Based on what I've found over the last 4 years I'm not convinced that there are many successful people out there. I used to own a satellite box which relied on clever people breaking the encryption codes and showing how clever they were by publishing them on the internet for others to use. With trading we have lots of people telling us how successful they are but not a single successful strategy, I often wonder why. I like BlueHorseshoe am sitting on the side at the moment. I'm using my software to see if I can find an entry condition that will give reasonable odds in my favour, I'm not hopeful and expect to walk away if I'm honest. Ed