Jump to content

Welcome to the new Traders Laboratory! Please bear with us as we finish the migration over the next few days. If you find any issues, want to leave feedback, get in touch with us, or offer suggestions please post to the Support forum here.


Market Wizard
  • Content Count

  • Joined

  • Last visited

Everything posted by BlowFish

  1. Isn't one of S.S's adages set it and forget it? Lends itself to not having always on internet.
  2. Despite the heavyweights already leaving the field I'll chip in a couple of points that don't seem to have been made yet. OP, Just because a random sequence (the coin toss of your example) exhibits similar patterns to a financial data series you simply can not conclude that the latter is random too. That's just logical fallacy. If you do even rudimentary statistical analysis of financial data series it seems pretty clear that they are not random. (Though IMHO certain market phases are more so than others). Benoit Mandlebrot has done a lot of work on the subject that you might find intresting. Of course just because the data is not random does not mean that it can be predicted. And to finish on a practical note it is not necasary to 'predict' to trade successfully. Finding good trade location (somewhere that you can have a tight stop with a decent target for example) is more important to me personally. So next time there is a six sigma event you might want to go in hard, you are not going to get another one for another 1.5 million years (if you are in the random walk camp). That's good trade location.
  3. Glad you are. I'm pretty enthusiastic about it but as I mentioned some people I have talked to found it a bit heavy going. Thinking about it a bit there are various 'high level conceptual models' (for want of a better term) that provide a broad framework of market operation. I am thinking of things like supply and demand or Market Profile whilst they are not exactly 'how markets work' they do give a consistent and plausible framework to trade within.
  4. Having said that a lot of participants can be placed in categories that have pretty similar objectives (or 'ideas'). Interestingly quite a few of these are not motivated by profit in a purely speculative sense. These participants will have similar modus operandi as other participants of the same category (though different to other categories). It is mind boggling (to me) that people start trading without having a rudimentary understanding of how the exchange they trade on works let alone a 'market'. I agree that the game is pretty complex but if you break it down into chunks (by participant objective is just one way) it need not be that daunting. Of course understanding how things work and profiting from that knowledge are quite distinct things
  5. A good book on market micro structure. I rather like Larry Harris 'Trading and Exchanges: Market Microstructure for Practitioners'. From what others have said to me you might find it heavy going, I really enjoyed it. Perhaps check it out in a library before shelling out or there used to be a couple of free chapters from an early draught on the interwebz too. This is not a how to trade book its a pretty dry textbook. It really does lay out who the (surprisingly numerous) types of players are and how they operate.
  6. Well, it's generally excepted that financial data series ('markets') do not have a normal (gaussian) distribution. Whilst there are various ideas what the actual distribution is, most (all?) agree that it is some sort of 'fat tailed' distribution. This would account for 'streaks' in systems. Both Mandelbrot & Taleb have done significant work on this (though I still haven't got round to reading Taleb). Fat tails suggest additional risk due to the increased likelihood of high sigma events (black swans if ytou like).
  7. The maths can help though for example you might see that trading n% of account equity less reduces the RoR from 1 in 500,000 to 1 in 50 million (completely made up numbers). The decision is based on emotional comfort the numbers just give you something solid to hang your hat on.
  8. MM, Yes its an 'emotional' thing as much as anything, but that's the nature of risk. If one is happy to take the same risks with a highly successful trading business with capitalisation of millions as with a start-up trading company with capitalisation of 10k go for it . The key thing is know exactly what that risk is!! Having says that as time goes by the chances of hitting that ruining streak looms ever closer. Larger sample size and all that. If you can significantly reduce your risk of ruin by reducing your bet size a little it's certainly worth considering. Btw it is not a straight correlation, halving your bet size does not half your RoR. Of course if you started very conservatively (which would likely need a large starting account anyway) then maybe you are happy with maintaining the same RoR. The other thing (assuming you have a reasonably profitable approach) is that due to the massive growth you get by compounding (maintaining the same risk levels) you are going to hit other issues anyway (liquidity for example). I also have a hunch that traders, trading systems, and markets themselves are 'streaky' in nature but that's just a hunch and a whole other story.
  9. That bolded bit is really smart imho. What you are actually doing is reducing your RoR (sorry couldn't resist) as your account grows. This has obvious advantages. One of the reasons experienced and successful traders still blow up is that they do not do this (or even are prone to over trade when on a wildly successful streak) as you play the game longer the probability of those NN tails in a row becomes greater and greater. And anyway would you really want the same risk profile on a multi million dollar account as you would on say a 25 grand one?
  10. That of course is your prerogative. It is just a term (albeit an emotive one) to describe the maths. Are you not the slightest bit interested what the probability of depleting that capital is for any given approach and position size? There is a big difference between affording to wipe out the account and setting out on a path where the probability of doing so is higher than you guess (I use the term 'guess' advisedly). All I am saying is it is naive to not pay attention to this. Incidentally a naive approach can still yield acceptable results. Chances are you will need quite a large account (relative to your bet size) though. I wonder how one can truly accept risk unless one objectively knows what that risk is? The other thing that running the (simple) numbers will give is a much better understanding of not only the the effects of different bet size and starting capital but how (drastically) R:R and % winners impact things. It does not take long to get an intuitive feel for RoR but until you do running numbers is a much cheaper and quicker way than trial and error.
  11. Not too hard really. The key metric is RoR (Risk of Ruin) as I mentioned early in the thread. This little site tells you all about why you should know it and how to calculate it. TradersCALM - risk of ruin menu Of course you are right that you still have to determine what is an acceptable level to you but this is the way to calculate it. It is also why the original question posed in the thread (and some of the answers) are pretty naive and rather 'dangerous'.
  12. A bit more on the turtles:- The turtles used a relatively sophisticated (for the time) volatility adjusted position sizing algorithm. i.e. A 'unit size' was determined for the dollar volatility of each market. They then 'pyramided' up to full size in four (if memory serves) steps. There where limits (set in units) for a single market, closely and loosely correlated markets and direction. Each turtle also traded on a notional account size that was determined (and periodically adjusted) for each trader by Rich. The management of trades was really the heart of the turtle system, entries where a simple Donchian breakout. For those interested theres a pdf knocking about on the interwebz that describes it in detail.
  13. It is worth mentioning that the correlation between aggressive buying/selling and the direction of order flow is probably only about 80% or so. There have been lots of studies on trade classification algorithms in general and Lee & Ready's algorithm in particular (which is essentially 'market delta'). Still, worth trying to get the best data you can.
  14. One last thing. I'd of probably have completely agreed with you in the past Gosu. I do recall having beers with him in the dim distant past and arguing the other side of the coin pretty vehemently. As I get older and maybe less ambitious (or perhaps less 'driven') I can understand him better. Of course having all the material things one actually need changes your perspective too. When you ask people why they want to trade the first answer is usually 'money' when you scrape a little deeper it turns out it is actually the freedom that money affords. Steves post seemed pretty focused on the money, which is fine, it is more important to some people than others and in the commercial world there always investors or shareholders to keep happy. I guess that's what prompted my post...think about what is really important to you and don't loose sight of that. To try and get back on topic I guess that influences the type of Mentor one would choose? (not that you really get a choice of course).
  15. As far as I am aware most months he takes out more than the 10k that he leaves in sometimes significantly so. I don't really know more than that. Things might have changed a little (what with inflation over the last 20 years)! The point is it supports him and his family. His lifestyle is pretty comfortable if not opulent, he is certainly not a high net worth individual. The key thing is trading provides all that he wants from it. He certainly is not 'still figuring things out'. postscript: Just spoke he is favouring currencies at the moment, EUR in particular. He is trading 3 full units (300,000 euro) of that instrument. Inflation I guess. Still the principles remain the same postscript2: The account size he maintains is simply to provide margin. Most would consider him over leveraged looking at that alone but of course the reverse is true as that does not represent his total capital. Size is not everything. In the context of this discussion it is (imho) nothing. You make the point about trading the minimum which is fair enough. If he had traded the maximum while maintaining the same level of risk (RoR ) and compounding his gains, there would not be any market on the planet that could accommodate the size he would now need to trade to maintain that level of risk. Are you saying you would rather learn from some one who risks more? That would be ironic as the chances of surviving to be a veteran (which is a measure of longevity btw) diminishes frighteningly at higher levels of risk. Not having a thorough appreciation of risk is why veteran traders become blown up veteran traders. A surprisingly common phenomena after a few years of 'great success' and the main reason jobbing traders don't get to set their own risk limits or even the instruments they can trade.
  16. One of the many great anecdotes from 'Reminiscences of a Stock Operator' is the trader that asks advice of a 'veteran' trader. It goes something like this 'I am long of XYZ and it's is bothering me. There is this and that happening and frankly I am loosing sleep over it'. The sage advice offered is 'reduce your position to sleeping point'. The importance of position sizing and risk should not be underestimated. Neither should it's effect on trader performance.
  17. Actually more importantly is to ask yourself is it someone who you would want to learn with. Veteran is not a term I particularly like (with its war connotations) however it's meaning is well defined. I would say that 20 years earning a living from trading constitutes 'long service in an occupation'. Just to re-iterate he earns his living like this and has done for 20 years, it is not a hobby. (though arguably one can be a hobbyist and a veteran concurrently). it is hard to come up with a persuasive reason to get someone to increase size who has achieved everything they want with their tradng. He has nothing to prove and has an abundance of money. (Abundance & high net worth are not the same thing as I hope I made clear btw).
  18. Which is exactly why the sort of people you describe are not likely to be financially motivated Steve. Those that still are will not waste there precious time teaching. In fact those that are (still financially motivated) have pretty much by definition 'not arrived yet'. Of course there are other motivations that have a by product of accumulating wealth. These type of educators will have all sorts of excuses; alternate revenue streams, I get something out of it, It shows commitment from the pupils etc. but the long and the short of it is they have found a nice little earner. Now these guys might even be OK traders but for sure they are not the 'high net worth' individuals you describe. The flip side of this is that not all 'successful' traders are high net worth individuals. I know a couple of guys that are happy to consistently take a handful of points here and there and still only trade 3 contracts. One guy I know trades singles and has lived off that for close to 20 years, he empties his account every month (apart from margin which used to be 10k last I asked him). Certainly not a high net worth individual, but most would agree a veteran. Fortunately there are people that realise that 'money isn't everything' (to coin the old cliché), ironically this is an easier perspective to get when you have an abundance of it. Anyway, the reason I say fortunately is that these individuals are likely to be motivated by other qualities than a trainees ability to pay.
  19. Welcome back DB. Always good to see a quality poster return.
  20. It has (been a while). A far as I am aware TSSuport have a kind of 'interim solution' for CD that is restricted to IQfeed and reads their (IQfeed's) historical data rather then the MC database. This is rather limiting not least because IQfeed does not supply enough historical data for a complete CD calculation. (Depending on how far back you want to go). A shame really but I guess better than nothing and possibly good enough for what you require. To be honest when it became apparent that was as good as it was going to get for the foreseeable future I stepped back a bit. Automation seems to be an area that they are more receptive to working on
  21. I have noticed this and mentioned it a couple of times over the course of the threads. (probably rather labouring the point). Price will move in the same direction for most of the day and volume will increase with each push attracting new buyers and sellers. On days like this the PVP will frequently jump ahead and form in the new high/low territory (as there where so many buyers sellers at that point). Price often will not hang round there and continue in he same direction. You will get the PvP 'stair stepping' along with the new extremes. There are several things you can do. Stay out of the market. This would possibly be the best choice, it follows Jerrys principles too. Take breakout trades at the 1 SD. In practice I have found on these days the PVP will usually be above the 1SD so this probably won't trigger a BO trade as described here. You could take trades at the 1SD even if the PVP is not between 1SD and VWAP. This pretty much runs against what Jerry has shown us here. You might not even have much chance of that as price leaves the VWAP and 1 SD behind. This is a little hard to describe in words but I have previously posted charts of this type of price behaviour/
  22. Yes Master the Markets as advised. It has a few errata you'll work that out as you digest the info. There where 3 mammoth threads here. Whilst there is quite a low signal to noise ratio in places there are a couple of insightful posters that contributed in other places. You'll probably work out who, there where also a couple that did not really stand up to close scrutiny (be more wary of the hindsight only guys).
  23. Possibly. A couple of the later threads where somewhat shorter but I seem to remember this one being longer than 5 pages.
  24. It's simply a data sampling issue. It 'means' exactly what it means. The raw data is each and every tick. If you sample that data every 29 ticks, 243 second, 53 contracts or 5.5 points you are holding one variable constant and sampling with respect to that. People choose (or should do) sampling periods to reveal characteristics in the underlying data that suits there trading objectives. The most obvious one is the types of move they are trying to capture.The main characteristics of 'type' are magnitude and length. So for example if you are looking for 2-3 point moves that generally take less than an hour to complete you probably would not use a weekly chart (a weekly sampling) to trigger your entries. Now having said that there are some 'special' considerations. I think the significance of most are being eroded. For example daily bars, the close represents the end of a session however with some instruments trading 24/5 I guess that is less meaningful than it once was. The weekly close probably still has some sort of tangible 'meaning' as markets still close at weekends. 30 minute periods meant something when a lot of people where using market profile (30minute TPOs), probably still do a bit though arguably there are better ways of profiling markets. I also remember a time when I could have sworn that 5 minute closes where 'gamed' by shorter term participants. Anyway short answer is that it is a data sampling thing. A 'doji' on a 10,000 tick chart 'means' price has not managed to maintain any advance or decline within that sample period. A 90minute bar moving above the previous 90 minute bars (sample) high means that price is trading at a higher price than in the whole of the previous sample. Of course attaching special meaning to magic numbers is dubious (imho). A 150 tick chart is likely to be just as effective as a 144 tick chart.
  25. Is it an array out of bounds error you are getting? If so simply increasing the array sizes should fix it 3000 should be ample. Currencies have a lot of pips in a day and so require larger array sizes.
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.