BlueHorseshoe

Thanks for an interesting and useful post. I wanted to question the following points: The trading methodology that Richard Dennis taught his Turtles is arguably one of the most difficult trading methods for a typical trader to execute. Is it really? By all accounts, 'Joe Public' loves to buy breakout markets at new highs. Buying a falling market is surely much harder - you've only to take a look at people's thoughts in the 'Never Catch a Falling Knife' thread to see that. While all trend traders play for the outlier, Turtles play for the outlier of the outliers—typically just one trend a year. It's debatable whether 'Turtle' trading is true outlier trading. If you compare it to the sort of thing that JW Henry or Bill Dunn do, for instance . . . ? Theoretically, for true outlier trading, one would never be flat, and the Turtles were. These points aside, I agree with what you're saying. I traded with a mechanical EOD strategy that wasn't automated, and I had someone else place the trades for me - this meant that I was pretty detached from the whole process and no possibility of lack of discipline could creep in. There was no early profit taking, and no discretionary impulse trades. Every trade was a 'good' trade. Of course, whether a trader makes money at the end of all that still depends on the rules by which they enter, exit, and size positions . . . BlueHorseshoe

Hi DB, Do you do anything ahead of the session to try and anticipate whether the market is likely to be choppy and rangebound within that 1min timeframe, or just trade what you see as the session gets underway? Thanks, BlueHorseshoe

Hi DB, Try and abstract from the specific to the general . . . I'm not interested in ValueCharts - I am interested in the fact that someone has been able to earmark such a general and public concept as their own intellectual property. You have a golden 'C' next to your name - I'm sure you can figure out the implications. Anyway, if you're not interested, why bother posting on the thread? BlueHorseshoe

So this patent does exist: United States Patent: 8195553 The patent would seem to be for any indicator that generates a volatility derived normalisation of price. Slightly hard to believe. So, what does this mean? Does it mean that Mr Helweg could try and stop someone else selling an indicator which normalizes price within, say, Keltner Channels? I suspect if Gosu were still around at TL he'd have known the answer! BlueHorseshoe

A bit like patenting a video game rather than the underlying code that generates the graphics? But then the visual representation of a value chart is just OHLC bars. Maybe I should try and dig a copy of Helweg's patent application out from the web . . . Thanks. BlueHorseshoe

I'm re-quoting PrymeTyme from the first page, because nobody has really addressed this: BlueHorseshoe

I was recently made aware that Mark Helweg has received a patent for ValueCharts. Is there anyone with a legal background who would care to explain how this is possible? I consulted a corporate lawyer with this question a few years back (not about ValueCharts specifically, you understand), and was told that it was not possible to patent a mathematical formula, otherwise it would be possible to patent something like the theory of relativity and restrict its use by others. In fact, didn't George Lane try and patent the Stochastic oscillator and fail? BlueHorseshoe

Just so long as you don't forget to buy back your entire inventory at a massive discount a month later . . . BlueHorseshoe

Buyers don't neccessarily need to step in. The number of buyers willing to cross the spread just needs to remain the same while the number of sellers willing to cross the spread and the number of sellers at the ask fall. I think this effect is most pronounced in equities and equity indices. There is always a 'buy-side' bias to these instruments in my opinion, and selling is as likely to be associated with profit taking or loss cutting as it is active shorting. Strip away all the retail day-traders, the prop shops, the HFTs, and the billion dollar hedge funds, and you're left with . . . a residue of buy-and-hold investors who are only ever long an equity. I've had this debate elsewhere with DB, though in the context of volume analysis. The ES tends to bottom on high volume (as buyers step in at value), and top out on low volume (as buying interest is depleted) - but noteably NOT not on high selling volume. Obviously this is a tendency rather than a rule. Think about it this way: - I have an Uncle who has been long half a dozen stocks for about ten years; I don't have an Auntie who has been short half a dozen stocks for ten years. Look at something like a currency and you're just as likely to get ACTIVE short selling as you are ACTIVE buying, so that's a different ballgame. And all of this is just my opinion, and just something to think about. Mr Horseshoes

I think it's important to remember that there is a difference between knowing the answer to this question, and knowing how to utilise this understanding. Supposing the very simplest model of supply and demand were to pertain . . . How you'd exploit this is still a big problem. Essentially you'd still need to make some sort of 'prediction' ahead of the current time. Not that I object to the thread topic, of course! BlueHorseshoe

TradeStation is great, but there are limits to what can be gleaned from the reports that a strategy test generates . . . For other ideas for useful Excel trade analysis using the TradeRecorder function that JSwanson has kindly provided, have a look at 'Trading Systems that Work' by Thomas Stridsman. BlueHorseshoe

Surely the whole of Catholicism is just a dark pool for "futures", with physical delivery upon "expiration"? Although the Vatican market-makers do look damn smart in those cassocks . . . BlueHorseshoe

Even amongst those who have sufficient financial capital remaining to continue after the original strategy has failed, it's another matter whether they have sufficient 'emotional' capital (energy, momentum, enthusiasm, ambition, passion, drive) remaining. Having a business fail is a draining experience. BlueHorseshoe

Obviously that's essential - it just comes later. I'm saying that it's better to check whether the swimming pool is full of water or baked beans before you start working out the best stroke for staying afloat in it. Sorry, that might be the worst trading metaphor ever I agree about 50/50 in terms of up and down bars. I was talking about consecutive up and down bars though. So, given that the ES closed up yesterday, what is the probability that it will close up today? The probability that it will close down could be seen as a very crude measure of the likelihood that it will 'go back the way it came', or revert to the mean. The probability that it will close up for a second, third, or fourth consecutive day, is a crude measure for its capacity to trend. Hopefully that's a bit clearer on what I meant. BlueHorseshoe

Hi Colonel Would you be willing to share a little of how you identify OTF activity in any particular cash session? "Common sense" might suggest that volume would provide this information - what's your approach to identifying institutional order flow? Thanks, BlueHorseshoe

I was just suggesting bar count as one crude method for gaining an idea of how a particular market behaves. I think that the more tests of this kind one can design and perform, the better an idea one will have of the general behaviour of that market. It's just an extension of the notion that something's profitability should not be parameter sensitive; a general concept of market behaviour shouldn't be dependent on the measurement methodology of any single test. I'm not sure that I'd agree with that. If you're testing to determine general market behaviour, then there is no stop, entry, exit, profit target etc. There's just how the market acts, in complete isolation from how we might try and trade it. I can't understand why more people don't do this kind of research first, and then develop a particular strategy for exploiting the market behaviour that they discover to be there. While I have made what many would probably consider schoolboy mistakes with the strategies I chose to use (such as trading without a stop), I am pretty confident that I would never have gotten killed . . . . because I had done the research beforehand to ensure that I was trading in line with the underlying behaviour of that market. That, of course, is all pretty hypothetical though! Sure. How many people on here try and daytrade the ES? Now that can't be easy . . . As I said, why not look at both? And then anything else that you can think of? It costs nothing, and it's hard to see how it could possibly be anything other than beneficial, especially if you're kicking around waiting to fund an account, or trading in sim. BlueHorseshoe

I agree that the 'red candle, green candle' account could be misleading - I think it's just being used here by BobC as a shorthand for entering counter-trend (ie on a limit order). My guess is that this comes down to the differences between markets again. I suspect that Col B is coming at it from the point of view of trading the ES, in which case I would tend to agree with what he says. The momentum just isn't there most of the time, and buying the thrust from support/green candles won't provide the follow-through. In a market that trends more then this makes far more sense, and buying notional support/red candles is a bad plan. Anyone interested in all this can carry out a few simple tests in the markets they trade. What percentage of the time, say, is a red/green bar followed by another red/green bar? For all N, what most often happens at a retest of the N-bar High/Low - breakout or reversal? For these tests you will also need to look at consistency in results throughout the data set - standard deviation from the mean result should give an idea of this (what you're trying to avoid is an end outcome percentage that is dependent on where the data set ends). Do this over a large data set in all timeframes and you will begin to build an idea of whether that particular market tends towards breakout trends or mean reversion. Hope that's useful to someone. BlueHorseshoe

I can do this for you but I can't invest masses of time in helping you develop your strategy if it doesn't work. If you're just wanting a simple backtest running then PM me. You'll need to state clear system rules, instrument(s), timeframe(s), and whether you want any slippage or commissions deducting, and also how you'd like limit orders to be filled in backtesting (trade-through or at-limit). BlueHorseshoe

One of the great things about referencing these higher timeframes as a swing trader is that they're inherently less noisy. A weekly or monthly chart still tends to have the "cleanliness" that daily charts used to have back in the days when trend following funds were raking it in. Trading from these higher timeframes is impossible unless you have shed loads of capital, but incorporating clear trend signals from them into lower timeframe systems as Ingot suggests is a great idea. If you want to "prove" this to yourself, you could make the following comparissons between monthly/weekly/daily data: 1) Calculate the relative frequency (across a large data set) with which 'swings' occur (a low with a higher low on either side, or a high with a lower high on either side). In "cleaner" data you would expect fewer of these. 2) Calculate the percentage of each bar that occurs outside the range of the prior bar, averaged across the entire data set. In data that is "cleaner" and exhibits less mean reversion you would expect this figure to be lower, as price trends into new territory with greater frequency. Hope that's of some use to someone. BlueHorseshoe

No worries. I don't use the radar screen so I'm not certain of coding for that, but it should be fairly straightforward for what you require. You could try asking Onesmith or Tams very nicely via a PM - they're usually able to come up with an answer for anything EL related. BlueHorseshoe

If you have a search around on the forum I'm sure you'll find recommendations for free realtime futures charting. http://www.prorealtime.com is pretty good as it doesn't require any software installation, and you can sign up for a free trial with intraday data. If you're happy installing stuff, then you could get a trial SIM account with any broker that also offers charting - Infinity have Sierra Charts, which is pretty popular. Hope that's helpful. BlueHorseshoe

I haven't tested this, as I'm not at a computer with TS at the minute, but this should work okay. If not, then let me know. BlueHorseshoe Vars:Counter(0); If C>Average(C,20) and C[1]<Average(C,20)[1] then Counter=1; {or you could just use "If C crosses over Average(C,20) then"} If Counter[1]>0 and C>Average(C,20) then Counter=Counter[1]+1; If C<Average(C,20) then Counter=0; Plot1(Counter,"ClosesAboveAvg",Green);

Did you mean something more like this? BlueHorseshoe

There's already loads of good advice here. One thing that I do is, rather than be a passive reader (of books, forums, etc), I try and engage with the content of the material by 'testing' ideas. This doesn't neccessarily mean performing some full blown analysis - sometimes I'll just pull up a chart and look for a clear example of whatever the author is talking about. Not getting distracted and flitting from one thing to another is not something that I can offer any advice on - I'm terrible for it - in fact I came online just now to pay an electricity bill, not to come here on TL . . . Good luck, and hope to see more contributions from you around the forum. BlueHorseshoe