You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
ronfalcone said:joined your group, but can't open any of the so-called Pre-Edge 'files' with any known program.
Not any great concern.
Have a merry,
Ron
Ron,
the link to the software that you need for this kind of files is on the group description page. I used this software to zip the files and it can also be used to unzip them. Please download this software and use it. If you still have a problem let me know. No one else mentioned any problem, so I gather that everything is OK. More files to come today.
Best,
Wally
Last edited:
ronfalcone
Member
- Messages
- 93
- Likes
- 1
Wally, articles
Got it. Thanks. Overall, not very impressed with the Vision stuff or 'Pre-Edge' concepts, though if someone's having success with them, I wish them much prosperity.
I will say, I WAS more impressed with the article The Daytrading Mind. Excellent piece! Yea, stuff we've all heard more or less, but the author captures real kernels of wisdom and puts them in an excellent perspective. So easy to see, but so hard to implement. For that reason, I tend to be leery of systems which generate many trades throughout the day---the more trades, the more potential for emotional stress and error. But that's just my opinion. For some traders, stalking the trade most of the day, and then pulling the trigger only very selectively and with much discretion is the better way to go. But again, MO.
Ron
wally_ said:Ron,
the link to the software that you need for this kind of files is on the group description page. I used this software to zip the files and it can also be used to unzip them. Please download this software and use it. If you still have a problem let me know. No one else mentioned any problem, so I gather that everything is OK. More files to come today.
Best,
Wally
Got it. Thanks. Overall, not very impressed with the Vision stuff or 'Pre-Edge' concepts, though if someone's having success with them, I wish them much prosperity.
I will say, I WAS more impressed with the article The Daytrading Mind. Excellent piece! Yea, stuff we've all heard more or less, but the author captures real kernels of wisdom and puts them in an excellent perspective. So easy to see, but so hard to implement. For that reason, I tend to be leery of systems which generate many trades throughout the day---the more trades, the more potential for emotional stress and error. But that's just my opinion. For some traders, stalking the trade most of the day, and then pulling the trigger only very selectively and with much discretion is the better way to go. But again, MO.
Ron
For what it's worth,
I've been following the Dow today, missed the opening move as I was at work, but I've looked back over it and been watching the last few hours...
With a 10pt leeway (as advised) on the barchart tops and bottoms on the Dow, and today's fairly small swings, the system is firing signals off like firecrackers on Chinese new year and you're hardly into a trade in one direction before it reverses or hits the MA - 10 pts on a 1 min chart is FAR too much leeway to allow on the relevant barchart formations... today, at least on the Dow, it's like being nibbled to death by goldfish. I have to say that this has been something I'm uncomfortable with - I don't understand how you can have the same stop/profit loss on 1 min bars as on daily charts and expect a sensible result, nor do I think it makes sense to have as much leeway on what you consider an equal top as you intend to use as a profit/loss stop exit. It's fair to say I'm seeing what my old fashioned TA says to expect - whipsaws in a market that is lacking in direction.
I'm prepared, as ever, to be corrected in this - but as I see it virtually every tiny little swing today since the big open move has been 10 pts or less, leading to many of these tops and bottoms being officially within limits for a signal... the indicators, poor dumb creatures that they are, have been making the requisite moves, and the result is bedlam. On the other hand, conventional boring TA would say we've been trading sideways for much of the day, with a couple of small jumps between levels, and being quick to trade a breakout of each range might have been worth a few points.
Dave
I've been following the Dow today, missed the opening move as I was at work, but I've looked back over it and been watching the last few hours...
With a 10pt leeway (as advised) on the barchart tops and bottoms on the Dow, and today's fairly small swings, the system is firing signals off like firecrackers on Chinese new year and you're hardly into a trade in one direction before it reverses or hits the MA - 10 pts on a 1 min chart is FAR too much leeway to allow on the relevant barchart formations... today, at least on the Dow, it's like being nibbled to death by goldfish. I have to say that this has been something I'm uncomfortable with - I don't understand how you can have the same stop/profit loss on 1 min bars as on daily charts and expect a sensible result, nor do I think it makes sense to have as much leeway on what you consider an equal top as you intend to use as a profit/loss stop exit. It's fair to say I'm seeing what my old fashioned TA says to expect - whipsaws in a market that is lacking in direction.
I'm prepared, as ever, to be corrected in this - but as I see it virtually every tiny little swing today since the big open move has been 10 pts or less, leading to many of these tops and bottoms being officially within limits for a signal... the indicators, poor dumb creatures that they are, have been making the requisite moves, and the result is bedlam. On the other hand, conventional boring TA would say we've been trading sideways for much of the day, with a couple of small jumps between levels, and being quick to trade a breakout of each range might have been worth a few points.
Dave
DaveJB said:For what it's worth,
I've been following the Dow today, missed the opening move as I was at work, but I've looked back over it and been watching the last few hours...
With a 10pt leeway (as advised) on the barchart tops and bottoms on the Dow, and today's fairly small swings, the system is firing signals off like firecrackers on Chinese new year and you're hardly into a trade in one direction before it reverses or hits the MA - 10 pts on a 1 min chart is FAR too much leeway to allow on the relevant barchart formations... today, at least on the Dow, it's like being nibbled to death by goldfish. I have to say that this has been something I'm uncomfortable with - I don't understand how you can have the same stop/profit loss on 1 min bars as on daily charts and expect a sensible result, nor do I think it makes sense to have as much leeway on what you consider an equal top as you intend to use as a profit/loss stop exit. It's fair to say I'm seeing what my old fashioned TA says to expect - whipsaws in a market that is lacking in direction.
I'm prepared, as ever, to be corrected in this - but as I see it virtually every tiny little swing today since the big open move has been 10 pts or less, leading to many of these tops and bottoms being officially within limits for a signal... the indicators, poor dumb creatures that they are, have been making the requisite moves, and the result is bedlam. On the other hand, conventional boring TA would say we've been trading sideways for much of the day, with a couple of small jumps between levels, and being quick to trade a breakout of each range might have been worth a few points.
Dave
Dave,
This is unfortunately real life trading that can and usually does differ from the 75-80% success rate advertised by the vendor of this system. As I said in my email to those who requested my materials, 'in hindsight everyone can be a visionary', but in real life things are not as smooth and easy.
This is true about many discretionary methods, free or commercial. For instance, I have seen quite a number of journals by CCI traders. Those poor souls too would end up with too many trades and little to show for. The journals in question were on some site that is now gone because Woodie did not like the negative impact the journals had on the image of the CCI as the leading indicator. Leading indicators are often more misleading than leading.
I, on the other hand, took only 3 swing trades today in ES and I am now done.
The reason I don't like much of the stuff out there is not because I derive some sick pleasure from bashing vendors, but simply because I think that many people will end up losing their time with this stuff and time can be more precious than money.
Best,
Wally
Last edited:
ronfalcone
Member
- Messages
- 93
- Likes
- 1
my results
tedious as it has been, I've been able to isolate a few good paper trades using my own standard deviation approach with seeming higher probabilities. It just takes so damn long to unravel.
Today, my standard deviation platform showed a top made at 10:33am on the YM with a subsequent MA cross beginning to occur at 10:46. Had you waited for confirmation and then entered short at the worst possible entry and accounting for maximum slippage (I always, when paper trading, take the worst-case scenario to try and simulate real time results as best as possible), then this trade would have conservatively yielded 18 points. This platform found another possible short at 12:30pm.
I'm convinced that statistical analysis---whether Bollinger or otherwise---is a must; indicators can be used within the context of SA (not TA). But having said that, there's still a long way to go.
Not to dismiss The Edge. I just am uncomfortable with 6-10 trades per day.
tedious as it has been, I've been able to isolate a few good paper trades using my own standard deviation approach with seeming higher probabilities. It just takes so damn long to unravel.
Today, my standard deviation platform showed a top made at 10:33am on the YM with a subsequent MA cross beginning to occur at 10:46. Had you waited for confirmation and then entered short at the worst possible entry and accounting for maximum slippage (I always, when paper trading, take the worst-case scenario to try and simulate real time results as best as possible), then this trade would have conservatively yielded 18 points. This platform found another possible short at 12:30pm.
I'm convinced that statistical analysis---whether Bollinger or otherwise---is a must; indicators can be used within the context of SA (not TA). But having said that, there's still a long way to go.
Not to dismiss The Edge. I just am uncomfortable with 6-10 trades per day.
Hi,
back again - I left the PC for a few hours there (anyone watching 'currently active users' then my PC was here, but I've been watching TV <g>).
I'd not like anyone to get the idea I'm intending to knock Edge without a fair shake, and with my intermittent attendance at the screen you do have to make allowances for good trades (or otherwise) missed, I'm just reporting what I get on the Dow in my bumbling way as there seems to be little on this index. I'm curious about whether the system works on multiple markets or not. Naturally I'm also curious about whether I see 75% winners from it...
I'll keep watching, but so far it seems to be doing what I'd expect - which is 'nothing magical'. As for leading, it looks to me to be lagging in today's sort of market... when the market manages to get a halfway decent move going in one general direction then the signals seem to be quite timely, they seem to be a bit more sluggish if the move dies a death quickly... I'll keep an eye on that, in case it transpires that this helps identify better trades from those that fire. Some of the signals today were visible as a large single bar completed, that large bar often turned out to comprise rather a lot of the complete move.
Still, interesting to watch as it gives a better idea of whether the signals are visible in RT as readily as when looking back.
back again - I left the PC for a few hours there (anyone watching 'currently active users' then my PC was here, but I've been watching TV <g>).
I'd not like anyone to get the idea I'm intending to knock Edge without a fair shake, and with my intermittent attendance at the screen you do have to make allowances for good trades (or otherwise) missed, I'm just reporting what I get on the Dow in my bumbling way as there seems to be little on this index. I'm curious about whether the system works on multiple markets or not. Naturally I'm also curious about whether I see 75% winners from it...
I'll keep watching, but so far it seems to be doing what I'd expect - which is 'nothing magical'. As for leading, it looks to me to be lagging in today's sort of market... when the market manages to get a halfway decent move going in one general direction then the signals seem to be quite timely, they seem to be a bit more sluggish if the move dies a death quickly... I'll keep an eye on that, in case it transpires that this helps identify better trades from those that fire. Some of the signals today were visible as a large single bar completed, that large bar often turned out to comprise rather a lot of the complete move.
Still, interesting to watch as it gives a better idea of whether the signals are visible in RT as readily as when looking back.
ronfalcone
Member
- Messages
- 93
- Likes
- 1
signing off
just wanted to wish everyone the best of prosperity and luck with the Edge. I will be signing off this thread, moving on to other pastures, and continuing the development of my own system. As the song says "I gotta do it my way."
I'm always welcome to e-mails and discussion by fellow traders.
Toodles,
Ron
just wanted to wish everyone the best of prosperity and luck with the Edge. I will be signing off this thread, moving on to other pastures, and continuing the development of my own system. As the song says "I gotta do it my way."
I'm always welcome to e-mails and discussion by fellow traders.
Toodles,
Ron
I just bought the system a few days ago, cuz I was curious to know what it was all about.
Of course, hoping that I could learn something new.Over the years, I haven't read an ebook that contains
some useful refreshing info (apart from a few exceptions). The edge is no difference.
Small review:
The BAD
-------
Waaaaay overpriced.For $500 you expect a little bit more. I remember that the first fx course I took from
Doug Schaff(Schaff trend cycle anyone?) was in that price range. With support included from the master himself.
I traded and developed divergence systems for small time frames several years ago. With great success, as I placed a few times top5 in several trading competions incl the king of mini and forex olympics.
I posted one of my systems on moneytec several years ago, and even opened a private chatroom to give support to anyone who wanted to give it a try.
And I never, ever charged a dime for it!!!
The GOOD
--------
I did a 2 week backtesting with the adviced parameters. And almost all the signals I got matched with the ones posted on the charts.
Moreover, apparently they post 2 times a week the charts with comments about the trades they took.
Can be very helpful for beginning (divergence) traders.
Conclusion
----------
1) if you don't have a consistent system yet, and can trade for about 7 hours non-stop then:
1.1) If you're confident enough that you know what divergence signal to take or to leave,
go for the original one posted by Wally
1.2) If yet you're not too sure, and need a little guidance, and have $500 you'd like to spend,
you can consider the-edge
1.2.1) but if you don't want to spend $500, then use the original system in a simulation account.
I'm pretty sure you will figure out yourself after a while what signals to take.
Piece of advice. When the market is ranging in a tight range, both stoch and macd will most of the times signal a divergence. That's because of the way both indicators are calculated. ignore those signals, as at that moment it's a pure coin flip wheter the mkt will move up or down.
2) If you have a consistent system or can't trade for 7 hours non-stop:
2.1) Completely ignore the-edge.
Hope this review was helpful :cheesy:
EDIT: some typos
Of course, hoping that I could learn something new.Over the years, I haven't read an ebook that contains
some useful refreshing info (apart from a few exceptions). The edge is no difference.
Small review:
The BAD
-------
Waaaaay overpriced.For $500 you expect a little bit more. I remember that the first fx course I took from
Doug Schaff(Schaff trend cycle anyone?) was in that price range. With support included from the master himself.
I traded and developed divergence systems for small time frames several years ago. With great success, as I placed a few times top5 in several trading competions incl the king of mini and forex olympics.
I posted one of my systems on moneytec several years ago, and even opened a private chatroom to give support to anyone who wanted to give it a try.
And I never, ever charged a dime for it!!!
The GOOD
--------
I did a 2 week backtesting with the adviced parameters. And almost all the signals I got matched with the ones posted on the charts.
Moreover, apparently they post 2 times a week the charts with comments about the trades they took.
Can be very helpful for beginning (divergence) traders.
Conclusion
----------
1) if you don't have a consistent system yet, and can trade for about 7 hours non-stop then:
1.1) If you're confident enough that you know what divergence signal to take or to leave,
go for the original one posted by Wally
1.2) If yet you're not too sure, and need a little guidance, and have $500 you'd like to spend,
you can consider the-edge
1.2.1) but if you don't want to spend $500, then use the original system in a simulation account.
I'm pretty sure you will figure out yourself after a while what signals to take.
Piece of advice. When the market is ranging in a tight range, both stoch and macd will most of the times signal a divergence. That's because of the way both indicators are calculated. ignore those signals, as at that moment it's a pure coin flip wheter the mkt will move up or down.
2) If you have a consistent system or can't trade for 7 hours non-stop:
2.1) Completely ignore the-edge.
Hope this review was helpful :cheesy:
EDIT: some typos
Bye Ron,
best of luck mate, and a very Happy Christmas.
Sidekick - not a world away from my thoughts <g> One problem is that of ignoring signals during directionless periods, as you'll 'lag' by a move if you wait to confirm trend/no trend, and to be truthful I don't find the ADX that brilliant a guide. (Personally I think it lags too much, it confirms a trend long after my dog has spotted it - this is usually a bad sign).
I'd say the system performs as you'd expect it to, if you have some understanding of indicators - nothing unusual in what it produces. Signals do match - BUT when would you enter/exit? W£hile a bar builds the indicator is still moving - on esignal you see the line changing angle as the bar progresses. The signal 'fires' sure enough, but you are always the odd bar behind - instructions are to be sure the indicators are doing their thing, a bottom or top involves a definite peak or dip, not a temporary flattening out. In backtest I'd program the enter as the open of the bar following the green/red bar that confirms entry, as it's when that bar completes that you are supposed to go in. Is that what you did?
(Curious, not arguing, by the way <g> I backtest stuff quite a lot too, and getting it to match what you'd do in real life can be a chore at times).
Thanks for the input - it's good to see well reasoned criticism (good or bad).
Dave
best of luck mate, and a very Happy Christmas.
Sidekick - not a world away from my thoughts <g> One problem is that of ignoring signals during directionless periods, as you'll 'lag' by a move if you wait to confirm trend/no trend, and to be truthful I don't find the ADX that brilliant a guide. (Personally I think it lags too much, it confirms a trend long after my dog has spotted it - this is usually a bad sign).
I'd say the system performs as you'd expect it to, if you have some understanding of indicators - nothing unusual in what it produces. Signals do match - BUT when would you enter/exit? W£hile a bar builds the indicator is still moving - on esignal you see the line changing angle as the bar progresses. The signal 'fires' sure enough, but you are always the odd bar behind - instructions are to be sure the indicators are doing their thing, a bottom or top involves a definite peak or dip, not a temporary flattening out. In backtest I'd program the enter as the open of the bar following the green/red bar that confirms entry, as it's when that bar completes that you are supposed to go in. Is that what you did?
(Curious, not arguing, by the way <g> I backtest stuff quite a lot too, and getting it to match what you'd do in real life can be a chore at times).
Thanks for the input - it's good to see well reasoned criticism (good or bad).
Dave
ronfalcone said:just wanted to wish everyone the best of prosperity and luck with the Edge. I will be signing off this thread, moving on to other pastures, and continuing the development of my own system. As the song says "I gotta do it my way."
I'm always welcome to e-mails and discussion by fellow traders.
Toodles,
Ron
Good luck Ron and Merry Christmas and a Happy New Year to you and others here.
Keep in touch with this board though. Yes, eventually we all have to develop our own stuff, but one can always benefit from what others have to say... Esentially you need to develop your own trading style more than a method or system. Once you have it, you will be able to use more than one method. For instance, I don't have just one method/system, I have a few of them, but they all belong to the same category/style: selective trading.
Plausibility of Mechanizing?
This goes out to the trader/programmers here.
What do you guys think about the plausibility of mechanizing either the Edge or VisualTrader strategies?
Of course, this would require removing all subjectivity and establishing precise criteria by which the computer could act on the set ups.
Is this realistically possible with divergence-based systems?
Have others tried and succeeded or failed? (Well I'd venture it's been tried, just looking for insight on what did/didn't work.)
Thanks
Bill
This goes out to the trader/programmers here.
What do you guys think about the plausibility of mechanizing either the Edge or VisualTrader strategies?
Of course, this would require removing all subjectivity and establishing precise criteria by which the computer could act on the set ups.
Is this realistically possible with divergence-based systems?
Have others tried and succeeded or failed? (Well I'd venture it's been tried, just looking for insight on what did/didn't work.)
Thanks
Bill
Hi Dave:
Any divergence-based systems you'd recommend that have already been mechanised?
How do you go about programmatically defining peaks and valleys in the oscillators and measuring their difference (of course in addition to measuring against price)?
I'm interested in programmatically backtesting divergence-based systems. And, If nothing else, I'd like to learn a little bit about how to overcome some of the programmatic challenges presented in strategies involving visual pattern recognition. With respect to trading/development platforms, I'm particularly interested in Metatrader 4 (for Forex) and TradeStation (emini futures and possibly Forex). Any experience with systems which accurately define and measure oscillator peaks/valleys as utilized in divergenced-based systems on these platforms?
Thanks in advance for pointers.
Regards,
Bill
Any divergence-based systems you'd recommend that have already been mechanised?
How do you go about programmatically defining peaks and valleys in the oscillators and measuring their difference (of course in addition to measuring against price)?
I'm interested in programmatically backtesting divergence-based systems. And, If nothing else, I'd like to learn a little bit about how to overcome some of the programmatic challenges presented in strategies involving visual pattern recognition. With respect to trading/development platforms, I'm particularly interested in Metatrader 4 (for Forex) and TradeStation (emini futures and possibly Forex). Any experience with systems which accurately define and measure oscillator peaks/valleys as utilized in divergenced-based systems on these platforms?
Thanks in advance for pointers.
Regards,
Bill
DaveJB said:Mechanising it etc is entirely possible, not hard to be subjective as th basic system is prescriptive enough.
I'd not do it however, as I don't think it's any better than other systems I've looked at.
One way is to run a moving average calc of the indicator values, and look for points where the difference between MA values changes sign... eg you've got a 5 period MA of your MACD say, values are 1,2,2,3,3,4,2,1,0,-1 ... differences between consecutive values tells you if it's going uphill, flat, or downhill - set your 'bump/dip' routine up to ignore changes of less than a suitable value (by altering this you'll make it more/less sensitive, use a percentage change to avoid scaling issues). Store the peak and dip values in an array, compare array values to determine whether 'this dip' is lower than 'last dip' etc.
Another way would be to pick a percentage you'll define as 'big enough change to call it a turn' and store 1st array value = 1st indicator value. If next value is greater than stored value then increase this first array value to match. If the next value is (scurrent array value - set percentage for turn...or less) then call your first array element 'top pt 1' increment to next array element and then keep lowering it as the indicator drops until the indicator hits (current array value + set percentage for turn) at which point you store the 'current array value', increment the array, then start running up towards the next peak.... you'll end up with a series of alternating peak and dip values for the indicator's swings.
Dave
Hope you followed that <g>
Another way would be to pick a percentage you'll define as 'big enough change to call it a turn' and store 1st array value = 1st indicator value. If next value is greater than stored value then increase this first array value to match. If the next value is (scurrent array value - set percentage for turn...or less) then call your first array element 'top pt 1' increment to next array element and then keep lowering it as the indicator drops until the indicator hits (current array value + set percentage for turn) at which point you store the 'current array value', increment the array, then start running up towards the next peak.... you'll end up with a series of alternating peak and dip values for the indicator's swings.
Dave
Hope you followed that <g>
Thanks Dave. I think I follow. One question: Would determining the peaks and valleys based on an MA of the indicator values contribute to more lag time on generating a clean entry signal? Yes, as you say adjusting the sensitivity would be key here. But, I'm just pondering whether or not increased lag time (if applicable here) would cause some of the short-term signal opportunities here to not work out as intended.
Not intending to put you (or anyone for that matter) on the spot, but, has this technique been actually applied by anyone or is it theorhetical? Any working examples out there you might be aware of?
Thanks again for your informed insight on this.
Regards,
Bill
Not intending to put you (or anyone for that matter) on the spot, but, has this technique been actually applied by anyone or is it theorhetical? Any working examples out there you might be aware of?
Thanks again for your informed insight on this.
Regards,
Bill
DaveJB said:One way is to run a moving average calc of the indicator values, and look for points where the difference between MA values changes sign... eg you've got a 5 period MA of your MACD say, values are 1,2,2,3,3,4,2,1,0,-1 ... differences between consecutive values tells you if it's going uphill, flat, or downhill - set your 'bump/dip' routine up to ignore changes of less than a suitable value (by altering this you'll make it more/less sensitive, use a percentage change to avoid scaling issues). Store the peak and dip values in an array, compare array values to determine whether 'this dip' is lower than 'last dip' etc.
Another way would be to pick a percentage you'll define as 'big enough change to call it a turn' and store 1st array value = 1st indicator value. If next value is greater than stored value then increase this first array value to match. If the next value is (scurrent array value - set percentage for turn...or less) then call your first array element 'top pt 1' increment to next array element and then keep lowering it as the indicator drops until the indicator hits (current array value + set percentage for turn) at which point you store the 'current array value', increment the array, then start running up towards the next peak.... you'll end up with a series of alternating peak and dip values for the indicator's swings.
Dave
Hope you followed that <g>
Hi,
if you used a short enough period for your MA of indicators I don't think the lag would be too bad, especially as in real life you have to be sure a dip or peak is actually in, and this takes a bar or two to decide otherwise you'll be taking minor trembles that (in hindsight) are seen not to be signals. The indicators ought to precede the candle chart patterns for it to work anyhow...
Using the second method would avoid lag, as in this any change above n% in the indicator, that is contrary to its trend up to this point, will trigger a peak/bottom signal. Yes, Iv'e programmed very similar to both of these in my Pfscan program - the MA method I use to decide the 'shape' of the indicators in the automated analysis routine, by comparing what a variety of MAs of indicators are doing over a period I find dips/peaks, U bends and so forth, and on the bar/candle charts I use the second method to locate swings in the price chart.
Before anybody asks I will point out that my program is not for sale any more, so I'm not touting for business surreptitiously <g>
if you used a short enough period for your MA of indicators I don't think the lag would be too bad, especially as in real life you have to be sure a dip or peak is actually in, and this takes a bar or two to decide otherwise you'll be taking minor trembles that (in hindsight) are seen not to be signals. The indicators ought to precede the candle chart patterns for it to work anyhow...
Using the second method would avoid lag, as in this any change above n% in the indicator, that is contrary to its trend up to this point, will trigger a peak/bottom signal. Yes, Iv'e programmed very similar to both of these in my Pfscan program - the MA method I use to decide the 'shape' of the indicators in the automated analysis routine, by comparing what a variety of MAs of indicators are doing over a period I find dips/peaks, U bends and so forth, and on the bar/candle charts I use the second method to locate swings in the price chart.
Before anybody asks I will point out that my program is not for sale any more, so I'm not touting for business surreptitiously <g>
Billworld,
To add something on whether divergence coding has been done...
TradesStation code library has multiple snippets, designs, systems etc for divergences between price and indicators... ie coding divergence is nothing! Coding up measures of the 'quality' of the swings leading up to a divergence is the real coding trick.
Strategically, stop placement ( win size : loss size ) is far more important than pretty divergence entry setups particularily if you are not filtering for 'quality' swings (ie when you're taking ALL the divergences). With proper stop loss size, you can drop divergence as a requisite condition and just trade the 2nd and 3rd thrusts (or hooks) of indicator and do just as well as with 'edges' created by divergences.
All the best.
zdo
To add something on whether divergence coding has been done...
TradesStation code library has multiple snippets, designs, systems etc for divergences between price and indicators... ie coding divergence is nothing! Coding up measures of the 'quality' of the swings leading up to a divergence is the real coding trick.
Strategically, stop placement ( win size : loss size ) is far more important than pretty divergence entry setups particularily if you are not filtering for 'quality' swings (ie when you're taking ALL the divergences). With proper stop loss size, you can drop divergence as a requisite condition and just trade the 2nd and 3rd thrusts (or hooks) of indicator and do just as well as with 'edges' created by divergences.
All the best.
zdo
Last edited:
Dammit!
Just typed 32 chapters and it went Kppphhht on me....
Okay, as ZDO says some platforms have this divergence stuff built in - Omnitrader for example. eSignal could be coded to do it, mainly by a masochist I'd suggest, but it's doable. First put the horse before the cart though - a significant programming effort will carry some baggage with it - it's hard to write your code, get it working properly, and then find you might as well be flipping coins. Write 'the edge' etc over a period of time and you'll be convinced it works, even while you keep losing money... you'll blame anything but the basic logic to the system, and that will be the EXACT spot it's going wrong.
Try your chosen system out on chart printouts, allow a sizeable slippage after entry/exit - allow for being able to react to a signal 1 or 2 bars later than they end up appearing, many systems take 1 or 2 bars to actually show the signal you were supposed to pounce on, and you need to add the spread to that. (These two factors alone will turn many 'winners' into losers). Once you KNOW you have an ace idea, THEN program it.
Dave
Just typed 32 chapters and it went Kppphhht on me....
Okay, as ZDO says some platforms have this divergence stuff built in - Omnitrader for example. eSignal could be coded to do it, mainly by a masochist I'd suggest, but it's doable. First put the horse before the cart though - a significant programming effort will carry some baggage with it - it's hard to write your code, get it working properly, and then find you might as well be flipping coins. Write 'the edge' etc over a period of time and you'll be convinced it works, even while you keep losing money... you'll blame anything but the basic logic to the system, and that will be the EXACT spot it's going wrong.
Try your chosen system out on chart printouts, allow a sizeable slippage after entry/exit - allow for being able to react to a signal 1 or 2 bars later than they end up appearing, many systems take 1 or 2 bars to actually show the signal you were supposed to pounce on, and you need to add the spread to that. (These two factors alone will turn many 'winners' into losers). Once you KNOW you have an ace idea, THEN program it.
Dave
Similar threads
- Replies
- 0
- Views
- 3K
- Replies
- 4
- Views
- 2K
