Don't Panic!!

[0007]

There are undoubtedly other ways rather than a load of hard-graft self-inflicted analysis projects to extricate money from the markets (as Sal's £300k demonstrates). But Oh! what you learn by going back to basics and working it out for yourself. Once you start confronting these things it makes you think and question your understanding which often in turn leads to greater understanding. Has helped me loads even though the more I think I know, the more I realise how much I don't know.

DT and Alan5616 both make very worthwhile points but I'm intrigued by Sal's progress. So go to it Sal!

 

[Cable Sal]

Thank you 0007. I'm glad you find it interesting watching what I'm doing

I maybe should have added to my clarification above, that I have no intention of going back to trading pure fundamentals, but that's not to say that I'm entirely ruling out fundamentals (or anything else for that matter). They may yet have their place in my plan. DT has a great link in his signature, made me think about TA throwing up that BP shares were looking a bit undersold earlier this year. TA may have said buy, fundamentals would have said they've got a disaster on their hands and prevented a stopped loss. Fundamentals look like a good additional way to stay out of a trade, and possibly a way, as Leonarda stated earlier, to tell you which stock to watch out of the thousands available. I'm just not at that stage yet,

Before I went off to university I worked for a while as a car mechanic. One lesson that I learned then, which I have applied throughout my somewhat diverse career is this:

Have the biggest toolbox you can buy and fill it with the most diverse range of tools that you can afford, don't just stop at a set of ring spanners and a pair of pliers. When you're on a breakdown at 3am in the pouring rain, you find that you're suddenly glad that you bought that funny shaped widget in the bottom of the box that you've never used before.

Why rule out anything? Why not learn everything - or as much as I can - and learn how to use each tool to best advantage for each individual situation. After all, the market is always changing. I'm the first to jump on to a new bandwagon to investigate it, and the last one to experiment with money to test it. Helps to keep me rich that way.

Sal

 

[DionysusToast]

Sal

From my perspective, it would be a shame if you really put a lot of effort into this but restricted yourself to 'TA or nothing'. The chance of you making money as the result of a pure analysis of textbook TA is close to zero in my opinion.

There's more to trading than textbook TA or fundamentals. It appears that you believe the alternative to a pure fundamental approach is the use of pure TA/indicators.

In my opinion, you are making the mistake of many engineers who attempt to resolve trading with their engineering skills. If you were attempting to resolve any other problem with your engineering skills, you'd take what works (or a hypothesis) and test it out. With this problem, you don't know what the solution is but are hoping you can conjure one up out of a set of limited and questionable tools. There are trillions of permutations and the chances of you stumbling on one that works without actually doing some trading are very slim.

There's a couple of things that would help you move along in this project. The first is to get to know some people that are making money trading and are willing to show you the ropes. The second is to watch a couple of markets real time, preferably with the DOM and T&S up and running so that you can at least see the games being played when the market approaches key levels. Once you realise games are being played, it puts a slightly different slant on things. I'd recommend the ES & the Bund as they move in different ways and aren't correlated.

There's a free ebook floating around by someone called "DBPHOENIX", in particular the sections on distribution and accumulation may switch on some lights for you in terms of what is actually happening behind that chart.

Anyway - good luck whatever you do.

Cheers

DT

 

[Splitlink]

Sal

From my perspective, it would be a shame if you really put a lot of effort into this but restricted yourself to 'TA or nothing'. The chance of you making money as the result of a pure analysis of textbook TA is close to zero in my opinion.

There's more to trading than textbook TA or fundamentals. It appears that you believe the alternative to a pure fundamental approach is the use of pure TA/indicators.

In my opinion, you are making the mistake of many engineers who attempt to resolve trading with their engineering skills. If you were attempting to resolve any other problem with your engineering skills, you'd take what works (or a hypothesis) and test it out. With this problem, you don't know what the solution is but are hoping you can conjure one up out of a set of limited and questionable tools. There are trillions of permutations and the chances of you stumbling on one that works without actually doing some trading are very slim.

There's a couple of things that would help you move along in this project. The first is to get to know some people that are making money trading and are willing to show you the ropes. The second is to watch a couple of markets real time, preferably with the DOM and T&S up and running so that you can at least see the games being played when the market approaches key levels. Once you realise games are being played, it puts a slightly different slant on things. I'd recommend the ES & the Bund as they move in different ways and aren't correlated.

There's a free ebook floating around by someone called "DBPHOENIX", in particular the sections on distribution and accumulation may switch on some lights for you in terms of what is actually happening behind that chart.

Anyway - good luck whatever you do.

Cheers

DT

dbphoenix based his calculations, like many people, on the use of volume. I have never been able to use it. For one thing, I have an idea that it is manipulated, or not even reported by some brokers. Perhaps, at EOD, it is all put together in a more accurate amount but during the day? I am sceptical.

I admit to not being the best of students and you can hit me over the head with the most obvious ideas and they won't enter. It's something that each one of us must study for himself. He had a famous thread, the name escapes me, but it must be archived and a monitor will get it for you.

He should be read by students interested in looking at all sides of the question. He has written a book, too, on the subject.

 

[timsk]

Hi Sal,
I suspect that the 'famous' thread referred to by Split' is this one:
Price, (Volume), Support, Resistance, Demand, Supply . . .
There are quite a few other threads of note by dbphoenix which can by found by doing a search of his name. He wrote a book, for which he made a modest charge of around U.S. $30.00, although that was quite a few years ago. He used to post some pdf sample chapters which, from memory, include the one that DT refers to on accumulation and distribution. However, they are protected by copyright, so I don't want to re-post them here (I have his complete book). However, if you're interested, you can contact Dbp at Traders Lab' where he hosts and moderates a forum (with the same username) devoted to Richard Wychoff - whose work has inspired much of his output.

One small thought to keep in mind . . .
You've observed already how something as seemingly trivial as a moving average can spark off very extreme reactions in some members! Needless to say, Dbp was a controversial character and, for every one of his devotees, there were some equally vociferous critics who maintained that his ideas are out of date and irrelevant in today's markets. As with anything you read about trading, there's precious little that one can accept verbatim and without question.

Tim.

 

[DionysusToast]

dbphoenix based his calculations, like many people, on the use of volume. I have never been able to use it. For one thing, I have an idea that it is manipulated, or not even reported by some brokers. Perhaps, at EOD, it is all put together in a more accurate amount but during the day? I am sceptical.
.

The DBP book is not for day trading for sure. The chapter I mention is relevant because it is about gamesmanship. There are no calculations involved. It is not technical per se. Perhaps someone could find a way to detect the accumulation phase programmatically but it would need to be someone that traded it & understood the nuances.

Shorter term - the exercise I mentioned of watching markets (chart/DOM/T&S) as they approach key leves should suffice. There's fun & games afoot. Mostly fun for those with size who sucker people in and then run over them.

The key here is that you need to resolve a problem before programming it. The path of automating something you don't know how to do is a painful one. Setting yourself a task such as this will end up a humbling experience, no matter how smart you are at work.

In my opinion, it's better to jump in and place some trades than to attempt to model the markets with no experience in that type of trading.

 

[Cable Sal]

A lot of food for thought in the past few posts, many thanks to everybody for contributing. One thing that I've decided to take on board whole-heartedly as a result of DT's posts in particular is that if I don't get the results that I'm looking for from textbook TA by October, I won't necessarily bin the whole project. Maybe there is another way, not quite so formulaic as pure indicators and not so discretionary as pure fundamentals.

I've not got much scope to widen my activities at the moment. The day job is running at 70 hours a week or more, which leaves me with litte more than the 4 hours between 10pm and 2am to do any research, number crunching or market watching. I get little snatched moments like right now, 10 or 20 minutes to catch a bit of sanity (so, obviously I head for the T2W forums). Towards the end of October it should ease off a little and I can look at scheduling in a bit of market-watching time.

New plan

Aug-Oct Research, learn, play with spreadsheets and indicators and simulate what I can
Oct-Dec IF I get something that looks vaguely profitable, I'll start fleshing out the trading plan
Dec-Jan I'll paper trade the plan using publicly available data and refine it
Jan-Feb If the plan looks encouraging I'll open account(s) and learn the platform(s)
Sometime afterwards - kinda thinking April - I'll start small-time trading live

If by October, when my time frees up somewhat, everything I've looked at has come up pants, I'll develop a new strategy based on a different approach rather than just give up.

If by the time I'm comfortable trading small-time and decide to scale up a bit, I'll be looking to investigate new strategies anyway.

I'm very, very intrigued by DT's message and I will, at some point, do exactly what you recommend, I just don't think I can do it right now with my current schedule. The exercise that I'm going through at the moment will stand me in good stead either which way though, so nothing to lose by playing with spreadsheets.

An update on the spreadsheet stuff while I'm here, in the interest of sharing the bad with the good. I knocked up a quick alternative to the comedy strict set of repeatable entry criteria based on the ATR fudged data that I already calculate. It's improved my losses so dramatically by changing my stop management, hey, it might just do the same again by replacing comedy entry with structured entry. Nope, my small profit of £36.97 turned into a £69 loss. Nice try though. :whistling

Sal

 

[timsk]

Sal,
I found the preview pdf by dbphoenix:
View attachment Previewr.pdf
Enjoy!
Tim.

 

[DionysusToast]

Sal - Good to hear you won't throw in the towel if an engineering approach doesn't get you to the end goal.

Good luck with it.

DT

 

[Cable Sal]

Many thanks DT

Can you believe I've only just noticed the "historical prices" hyperlink on Yahoo. Doh and woo-hoo, all in one moment! So, I can easily lay my hands on highs, lows, opens and closes for the 14 days before I started recording the spread. I can easily calculate an accurate ATR(14), amongst others. This may take a little while, but could be interesting...

I've been using a few spare moments looking at the correlation between spikes and volume in the context of the pdf that Timsk posted, prompted by DT and Split. It throws an interesting light on proceedings. Like a game really. I'm good at games, especially once I've got an idea of the rules and how to make them work for me.

Gonna be a bit manic for a week or three now, but I'll keep posting as much as I can in the meantime.

Sal

 

[Splitlink]

It seems to me that you are going to need the three-day bank holliday in a couple of weeks time for the increasing pile of research on hand!

Something that I use and have not seen mentioned is the line of "fair value". This was, really, the only piece of useful information that I dredged from Market Profile. Well, everything has something useful!

Please do not go into MP on my recommendation but, in a nutshell, the period's price range, when used with MP, takes the form of a statistical bell shape, with the price of fair value cutting the peak.

If one has (in fact, one has to have) a clear idea as to the trend direction he can, then, enter the market below the line (in a bull), above the line in a bear and this helps me in timing and entering correctly.

I quickly learned that if that was all I needed from MP, then I could more easily use the line that cuts the maximum number of bars on a bar chart.

Remember, though, that if the direction is wrong, nothing will work and IMO that is why so many fail in trying to spot turns. They are trying to spot a change of trend. Far better to get it established.

 

[Cable Sal]

That's an interesting thing for me to consider Split, I shall tuck it away in the corner of my mind for an appropriate moment.

I appreciate what you say about the mountain of research, lol, I'm cool with it. I'm on a huge learning curve, and immersion helps me a lot. It's all about awareness at the moment, my focus on indicators is a pragmatic way to analyse and understand the data and the drivers behind the change in trends. The more my awareness grows of other non-discretionary aspects of trading, the sooner I'll be picking all the right tools for the right job.

Unfortunately I can't devote the bank holiday to playing with spreadsheets as I've gotta go to the Beatles festival in Liverpool. It should be fun though, and it's about time I got blind drunk and slumped in a corner somewhere. I don't think I've done that since Mardi Gras in Louisiana last year. :innocent:

In the meantime, I should have finished structuring all the historical data into my spreadsheets from Yahoo by tomorrow, and be able to rebase my ATR calculations on actual highs and lows. It'll be interesting to see whether that pushes profitability up or down. A whole universe of indicators will open up to me once I've got the structure set up, I'll be like a kid in a sweet shop. (Honestly, I'm such a nerd at times.) I've nearly finished the pdf that Tim posted, it's an enjoyable read and an interesting light to shine on market behaviour. I've hardly scratched the surface of the famous thread though, I'm just a few pages in.

Anyway, enough of the little grey squares tonight, I'm gonna catch 40 winks.

See ya later,

Sal

 

[Splitlink]

That's an interesting thing for me to consider Split, I shall tuck it away in the corner of my mind for an appropriate moment.

I appreciate what you say about the mountain of research, lol, I'm cool with it. I'm on a huge learning curve, and immersion helps me a lot. It's all about awareness at the moment, my focus on indicators is a pragmatic way to analyse and understand the data and the drivers behind the change in trends. The more my awareness grows of other non-discretionary aspects of trading, the sooner I'll be picking all the right tools for the right job.

Unfortunately I can't devote the bank holiday to playing with spreadsheets as I've gotta go to the Beatles festival in Liverpool. It should be fun though, and it's about time I got blind drunk and slumped in a corner somewhere. I don't think I've done that since Mardi Gras in Louisiana last year. :innocent:

In the meantime, I should have finished structuring all the historical data into my spreadsheets from Yahoo by tomorrow, and be able to rebase my ATR calculations on actual highs and lows. It'll be interesting to see whether that pushes profitability up or down. A whole universe of indicators will open up to me once I've got the structure set up, I'll be like a kid in a sweet shop. (Honestly, I'm such a nerd at times.) I've nearly finished the pdf that Tim posted, it's an enjoyable read and an interesting light to shine on market behaviour. I've hardly scratched the surface of the famous thread though, I'm just a few pages in.

Anyway, enough of the little grey squares tonight, I'm gonna catch 40 winks.

See ya later,

Sal

As I said, you have to have a clear idea of the direction. With that in your head, what is the maximum number of 5 minute bars , or whatever TF, cutting the fair value price today? Think about it before the bank holliday, then!

I'm referring to the FT.

 

[megamuel]

How are the spreadsheets going Sal?!

 

[Cable Sal]

Hiya Megamuel,

Struggling a little bit for time and internet at the moment, but I'm managing to keep on top of spreadsheet maintenance - just about. I'm collecting my numbers and adding them into the spreadsheet each day. It'll be another week before I'm back into the usual routine, and able to try some new ideas out.

I'm very pleased with the results that the spreadsheets are showing. I'm running 9 simultaneous methodologies across the basket of 11 shares, operating at twice the minimum bet. Overall, I'm showing significant positive expectancy on 4 of the 9 methods, with running account profits ranging from £250.87 to £523.34.

The thing that's particularly catching my eye at the moment is that there's 4 of the basket of 11 shares that are making a consistent loss - all the time - on all 9 methods. I can see that if I remove those four from the basket, the philosophy of cutting losses short and letting profits run on the remaining 7 is working really well. The basket of 7 shows a positive expectancy on 6 of the 9 methods (including the 4 methods that stay in profit with all 11 shares), and account profits range from £570.23 to £1,298.57.

I've not had a chance to consider the similarities between the four loss-making shares, I think that 0007's three bears philosophy might show why they don't work for me. It shouldn't take long to figure it out, and it'll help me to define the characteristics of which shares I should and shouldn't bother watching. I think share choice is probably my next focal point.

The other 7 shares lose a bit, gain a bit, and overall gain more profit than they lose. By coincidence of choice at the outset, I'm usually going short on half the basket and long on half, so peaks and troughs in the FTSE are hedged quite nicely. An accidental lesson learned there!

I still want to compare more indicators. Once I have a little time to spare it'll be easy enough to drop a few more columns of formulae into the spreadsheet and see if I can beat the profitability of the existing methods. I think my entry criteria could do with a bit more focus, as I rushed that a bit. I wouldn't be confident enough (even with the profitable methods) that I wouldn't be tempted to second-guess the entry conditions if I was trading for real.

Setting stops and targets seem okay, I think I'd happily choose one or other of the most profitable methods for the plan and stick to it. I need to run the methods for a while yet and see which ones work best over a longer period / a greater number of up-down price cycles. I'd still like to refine my exits a bit, I'm sure I could eek out a little more profit, but it's good enough to work with.

So, all good stuff, just not enough time free at the moment. Must dash now.

Sal.

 

[megamuel]

Hiya Megamuel,

Struggling a little bit for time and internet at the moment, but I'm managing to keep on top of spreadsheet maintenance - just about. I'm collecting my numbers and adding them into the spreadsheet each day. It'll be another week before I'm back into the usual routine, and able to try some new ideas out.

I'm very pleased with the results that the spreadsheets are showing. I'm running 9 simultaneous methodologies across the basket of 11 shares, operating at twice the minimum bet. Overall, I'm showing significant positive expectancy on 4 of the 9 methods, with running account profits ranging from £250.87 to £523.34.

The thing that's particularly catching my eye at the moment is that there's 4 of the basket of 11 shares that are making a consistent loss - all the time - on all 9 methods. I can see that if I remove those four from the basket, the philosophy of cutting losses short and letting profits run on the remaining 7 is working really well. The basket of 7 shows a positive expectancy on 6 of the 9 methods (including the 4 methods that stay in profit with all 11 shares), and account profits range from £570.23 to £1,298.57.

I've not had a chance to consider the similarities between the four loss-making shares, I think that 0007's three bears philosophy might show why they don't work for me. It shouldn't take long to figure it out, and it'll help me to define the characteristics of which shares I should and shouldn't bother watching. I think share choice is probably my next focal point.

The other 7 shares lose a bit, gain a bit, and overall gain more profit than they lose. By coincidence of choice at the outset, I'm usually going short on half the basket and long on half, so peaks and troughs in the FTSE are hedged quite nicely. An accidental lesson learned there!

I still want to compare more indicators. Once I have a little time to spare it'll be easy enough to drop a few more columns of formulae into the spreadsheet and see if I can beat the profitability of the existing methods. I think my entry criteria could do with a bit more focus, as I rushed that a bit. I wouldn't be confident enough (even with the profitable methods) that I wouldn't be tempted to second-guess the entry conditions if I was trading for real.

Setting stops and targets seem okay, I think I'd happily choose one or other of the most profitable methods for the plan and stick to it. I need to run the methods for a while yet and see which ones work best over a longer period / a greater number of up-down price cycles. I'd still like to refine my exits a bit, I'm sure I could eek out a little more profit, but it's good enough to work with.

So, all good stuff, just not enough time free at the moment. Must dash now.

Sal.

Interesting stuff indeed Sal! Glad things are going well. Just out of interest, these 9 methodologies you are testing. Are the entries based on technical analysis, indicators, price action.... Or are they random entries and you are just using money management and probabilities to show profit on your account? Sorry if you've already answered this.

Also I'd be interested to know which 4 shares are causing the losses. I'd like to have a look at the charts and see if I can spot any similarities.... Let me know the tickers and I'll check them out... If you don't mind of course. I must have missed the "Three bears philosophy", I'll have to check that out!

As for going long on half and short on the other half, I paper traded something similar a while ago in which I'd go short on 5 of the more poor performing shares of the last 3 months and long on the best performing and update every week. In fact, looking back over my notes the other day it seemed to work quite well so I have no idea why I didn't stick with it. Although I am guessing it probably had something to do with roll over charges and spreads but I'm not sure. I guess its a bit like pairs trading on a bigger scale.

Anyway, sounds like things are going well and I am looking forward to the next update already!

All the best,

Sam.

 

[0007]

Interesting stuff indeed Sal! Glad things are going well. Just out of interest, these 9 methodologies you are testing. Are the entries based on technical analysis, indicators, price action.... Or are they random entries and you are just using money management and probabilities to show profit on your account? Sorry if you've already answered this.

Also I'd be interested to know which 4 shares are causing the losses. I'd like to have a look at the charts and see if I can spot any similarities.... Let me know the tickers and I'll check them out... If you don't mind of course. I must have missed the "Three bears philosophy", I'll have to check that out!

As for going long on half and short on the other half, I paper traded something similar a while ago in which I'd go short on 5 of the more poor performing shares of the last 3 months and long on the best performing and update every week. In fact, looking back over my notes the other day it seemed to work quite well so I have no idea why I didn't stick with it. Although I am guessing it probably had something to do with roll over charges and spreads but I'm not sure. I guess its a bit like pairs trading on a bigger scale.

Anyway, sounds like things are going well and I am looking forward to the next update already!

All the best,

Sam.

interesting comments. Re the three bears, I think Sal is referring to my earlier post on the influence of volatility. To my surprise I have found that volatility is very important to my way of trading and if I keep it (i.e. trade suitable shares-I have no control the market) within suitable bounds I can control my risk management much more easily. Maybe it's a bit like riding a bucking bronco - you need a bit of fun but not so much that he throws you off. And of course, what you count as fun depends very much upon your riding (trading) ability! I've found that noting ATR and beta to be quite useful.

 

[Cable Sal]

Hey everybody, I'm back. I've no idea where anything is up to, but I've managed to keep collecting the data daily. All I've gotta do now is transfer it into the models and see what happens...

By the way, one of the basket of 11 shares that I was modelling was suspended the other day. If I'd have had a bet on that company what would have happened? Has anybody experienced spreadbetting a stock that gets suspended?

 

[Cable Sal]

Interesting stuff indeed Sal! Glad things are going well. Just out of interest, these 9 methodologies you are testing. Are the entries based on technical analysis, indicators, price action.... Or are they random entries and you are just using money management and probabilities to show profit on your account? Sorry if you've already answered this.

Also I'd be interested to know which 4 shares are causing the losses. I'd like to have a look at the charts and see if I can spot any similarities.... Let me know the tickers and I'll check them out... If you don't mind of course. I must have missed the "Three bears philosophy", I'll have to check that out!

As for going long on half and short on the other half, I paper traded something similar a while ago in which I'd go short on 5 of the more poor performing shares of the last 3 months and long on the best performing and update every week. In fact, looking back over my notes the other day it seemed to work quite well so I have no idea why I didn't stick with it. Although I am guessing it probably had something to do with roll over charges and spreads but I'm not sure. I guess its a bit like pairs trading on a bigger scale.

Anyway, sounds like things are going well and I am looking forward to the next update already!

All the best,

Sam.

Hi Sam,

The entries of 6 methods are based on the previous 3 to 4 days of price action, I've cooked up a little method which is easy to model and apply without needing discretion. The other 3 methods are based on indicators - lol, I can't remember which, I need to familiarise myself with what I was doing. I thought I'd pay attention to my neglected journal before I pay attention to my neglected spreadsheets. One exit is based on price action alone, the rest are a combination of money management and indicator triggers - whichever is triggered first determines the sell/buy point.

The four shares that were making consistent losses (I assume they still will be once I've plugged in the rest of the data) are: Dignity (DTY.L), Genus (GNS.L), ICG (ICP.L) and Jar LLoyd (JLT.L). Dignity, Genus and Jar Lloyd are all very low volume stocks, so I can see what's going wrong there. The ATR varies significantly from one day to the next on these three, and as my stops are based on ATR(14), the spikes cause false stops. The value of these 3 are the only ones in my basket that are over the £5/share level, so the value of price movements tends to be quite high, and stops are expensive. ICG I haven't figured out. There's nothing extraordinary about this stock, all of its vital statistics fall within the same order as stocks that are making me a profit. Somehow, it always manages to trigger the wrong signal, or trigger the right signal too late, or not trigger anything when there's a profit to be made. All of the stocks and all of the methods do this to some degree, but ICG does it consistently, and I'm not sure why yet.

More than happy for you to have a look at the charts.

Cheers,

Sal

 

[Cable Sal]

interesting comments. Re the three bears, I think Sal is referring to my earlier post on the influence of volatility. To my surprise I have found that volatility is very important to my way of trading and if I keep it (i.e. trade suitable shares-I have no control the market) within suitable bounds I can control my risk management much more easily. Maybe it's a bit like riding a bucking bronco - you need a bit of fun but not so much that he throws you off. And of course, what you count as fun depends very much upon your riding (trading) ability! I've found that noting ATR and beta to be quite useful.

Yes, sorry, that's what I was referring to. I loved the analogy, there's some stocks with behaviour that's just too big, just too small and others that are just right. It made me view my trade plan differently. Instead of trying to find methodology that would work with anything and everything, I'm now trying to find methodology that will work with stocks that are just right for me, and methodology to identify which stocks those are.

I've been working successfully with ATR quite a bit, but not yet beta. It's on my hit list of things that people have mentioned that I don't know about.

Cheers,

Sal