Curtis Faith has written a book - The Way of the Turtle - you can read about it on his site.
But that's not why I'm here this morning. I'm here because he wrote an article on a subject that is near and dear to my heart. An article that could have been titled - "why doesn't this stupid system I got out of this stupid book work as well as it did in the book". Except that title is much too long.
He goes into several reasons why back test results are frequently bogus and I reviewed them and I think that my back testing meets all requirements for being honest and realistic. If you want to see what he has to say then scoot on over to his site and read the article. I'll wait.
I think I resolve all of the issues with my basic approach which is - keep it simple. If it is simple then people can understand it and understand why it works and so what if everyone else uses it. That's the beauty of simplicity.
But best of all, given my care to find the best possible methods to help me in my trading - the absolute best part of it is - I give it away - for free - I don't pretend that it is something really special or something really important or a cure for cancer, the crabs, or the clap - it's just trading folks.
If you have a good idea and your first thought is how can I make some money from this then your idea is "how can I make some money from this".
In order to make money you have to figure out how to convince people that your idea is good for them - good enough that they will be willing to pay to get to your thoughts. And trust me - you don't get what you pay for - you get what the other guy thinks he knows. And we all know what I have to say about that.
The way of the turtle is simple - take a lot of small losses and, then, when you have a winner ride it to the end of the trend - do you really need another book to learn that?
Showing posts with label Turtles. Show all posts
Showing posts with label Turtles. Show all posts
Sunday, March 18, 2007
Thursday, January 11, 2007
Turtles – Again
The Turtles original method was based on Donchian channels. As I remember it they went long when the price went above the upper channel and short when it went below. Then they used risk management to keep them in the game. In other words if the stock kept running in their desired direction they stayed with it and if it came back the stop loss would take them out. They had many small losses and a few large wins and that’s what kept them solvent.
I wrote a filter using Donchian channels and set it up like the Turtles would have done – then I back tested it - I hate “many small losses.” I take enough as it is.
So I changed it – I set the filter to look for stocks that came back up through the lower Donchian channel after being below it for at least one close. This is what that looks like –
This output is actually for Wednesday's close - you can see that Glass responded well to the Donchian channel.
When I back tested it there were some absolutely great results. 62% win percentage and 75% annualized ROI. But the net change over time was where the real story was told – This has the potential to be a real intermediate (10 – 20 day) term filter.
NET Change
1 day - .04%
4 days – 1.07
10 days – 2.39
20 days – 4.33
30 days – 4.86
Donchian channels Marlyn’s way – it might be a winner.
I wrote a filter using Donchian channels and set it up like the Turtles would have done – then I back tested it - I hate “many small losses.” I take enough as it is.
So I changed it – I set the filter to look for stocks that came back up through the lower Donchian channel after being below it for at least one close. This is what that looks like –
This output is actually for Wednesday's close - you can see that Glass responded well to the Donchian channel.
When I back tested it there were some absolutely great results. 62% win percentage and 75% annualized ROI. But the net change over time was where the real story was told – This has the potential to be a real intermediate (10 – 20 day) term filter.
NET Change
1 day - .04%
4 days – 1.07
10 days – 2.39
20 days – 4.33
30 days – 4.86
Donchian channels Marlyn’s way – it might be a winner.
Saturday, December 23, 2006
Revisiting the Turtles
We have written about the Turtles before this in mock admiration because they proved that if you followed the rules you could profit in the market. The method was largely - take a trade whenever a certain criteria was met and set a firm stop loss under it. As long as the trade continued to meet the criteria stay - with it and keep trailing the stop loss.
So now I'm going to say what I should have said then - this works - all you need to do is stick to it. And that is what the Turtles basically proved - if you have something that works don't fix it.
I can prove it. I wrote a filter that was a little more complicated than normal.
Show stocks where low 2 days ago < lower bollinger band(20) 2 days ago
and low 1 day ago < lower bollinger band(20) 1 day ago
and close > open
and Average Volume(90) is above 500000
and close is between 15 and 35
This requires the low of the day to be below the lower bollinger band on the previous two days and that the day before the trade is taken the stock price went up. Same basic price range and volume constraints.
I then looked back 30 days and pulled out this list of stocks (highest 10 by volume on that day):
NT, HANS, DHI, PHM, TOL, CC, SPF, INTU, HOV, and AQNT. I then put a mythical 5000 investment in each stock. As of Friday: CC and INTU were both stopped out at a 5% loss and the rest were profitable. What are the returns - over the 30 day period the SPX returned 1.018 and this little system - let's call it the Turtle system - returned 1.119.
The key factor is - take the losses as quickly and painlessly as possible and let the winners run. That is how you trend follow.
So now I'm going to say what I should have said then - this works - all you need to do is stick to it. And that is what the Turtles basically proved - if you have something that works don't fix it.
I can prove it. I wrote a filter that was a little more complicated than normal.
Show stocks where low 2 days ago < lower bollinger band(20) 2 days ago
and low 1 day ago < lower bollinger band(20) 1 day ago
and close > open
and Average Volume(90) is above 500000
and close is between 15 and 35
This requires the low of the day to be below the lower bollinger band on the previous two days and that the day before the trade is taken the stock price went up. Same basic price range and volume constraints.
I then looked back 30 days and pulled out this list of stocks (highest 10 by volume on that day):
NT, HANS, DHI, PHM, TOL, CC, SPF, INTU, HOV, and AQNT. I then put a mythical 5000 investment in each stock. As of Friday: CC and INTU were both stopped out at a 5% loss and the rest were profitable. What are the returns - over the 30 day period the SPX returned 1.018 and this little system - let's call it the Turtle system - returned 1.119.
The key factor is - take the losses as quickly and painlessly as possible and let the winners run. That is how you trend follow.
Friday, December 08, 2006
Risk Management
I was reading Dr. Brett’s BLOG and he brought up the Turtles – remember the Turtles - no - mother, not the Teenage Mutant ones. These Turtles were a group of people selected to take part in a noble experiment to see if absolute morons could be taught to trade in the stock market (sort of like the movie - Trading Places). Well that’s probably not true either and perhaps even a little harsh. Well, it’s a lot harsh but the Idea (and we always capitalize the “I” in “Idea” when it is a great Idea) was to take ordinary people not unlike you or I for example and teach them a trading method and then, if they followed that method to a “T” they would make fame and fortune. For some reason or other they called themselves “Turtles” probably because of the absolute slow and boring nature of the work they were doing.
Think about it people – if I give you a set of trading rules that I know has a probability of success of say 55%, plus a very large trading account and I set you in front of a computer terminal all day and you follow those rules perfectly then you should have a 55% probability of success. And if you don’t follow the rules, you won’t. Good grief - for this they needed a test? I’m probably missing something as usual. Anyway if you go to your favorite search engine and you type in Turtle you will probably find the rules. They ain’t much and they have been sold for thousands and thousands of dollars but are now available for free. And there is a reason for that. They basically say buy the trend and when it stops trending sell it - Duh. I can prove this and will a little bit later in the post.
But let’s not belittle the process itself. The most important part of Turtling is risk management. That’s because risk management is the most important thing a trader can do. Above all preserve capital until you hit the home run. Something we day traders sneer at and other people make fortunes doing.
Now the proof. I pulled out one of my favorite tried and true, time tested – oh crap – I threw together a filter that simply selected stocks based on closing price in relation to 2 moving averages and then listed them in volume order highest to lowest. It is a very prolific filter because there are so few moving parts but we only use the top several selections by volume in our back testing so it doesn’t really matter how many are output. I expect this filter to produce about 60% winners over a four-day period – the expectation is based on experience.
When I back test I use a simple baseline – One trade a day with no more than four stocks in the portfolio at any time. The way I have the back test exit criteria set is also simple – after four days exit the trade and take a new one unless you hit a 10% stop loss before 4 days in which case sell that stock and take a new one. Based on a virtual equity starting capital position of $100K this permits 25K per trade to begin. Note that I don’t care how many shares I’m holding just how much money I’ve got in them. After I run the test over an 80-day period or so I look at the equity results and for this baseline we achieved $53860 in profit with a 67% win rate – not bad for 80 days or so.
After establishing the baseline I started mucking with the internals one at a time. First number of stocks in the portfolio. I adjusted this to 1, 2, 8 and 16 stocks in the portfolio at any one time and here are the results.
1 – 47865
2 – 46147
8 – 24325
16 – 11570
This suggests that there is an optimum number of stocks in the portfolio and for this test it seems to be four. (If we doubled our starting cash to 200K would that change the number of stocks in the portfolio factor)? Let’s now look at more than one trade a day with no more than four stocks in the portfolio. (If you answered “yes” to our question above you are correct).
If we take four trades a day with four in the portfolio we can get to 60249 with a 71% win rate. For two trades a day with four in the portfolio we achieve 54891 and 68% win rate which is still better than one a day with four in the portfolio.
Let’s try one more factor – let’s take our best result, 4 x 4, and increase our holding period to 10 days. The result was 24357. If we add a simple measure such as a 15% profit exit criterion it goes to 34334. This is probably trying to tell us that stocks go up and down and unless you are a buy and hold forever type you need to attend to your portfolio a little more often than the broker recommended once per year.
So with a makeshift filter I have shown how money management – changing the way you buy not what you buy can make a large impact on your results.
According to song and story those Turtles who became successful adhered to the process like glue. The ones who deviated from the rules were not successful. It is that simple, kiddo.
There are a couple of exceptions to the above regarding “what to trade”. For those of you who are enamored of low price stocks (less than $10) – same system with a dollar requirement of $2 to 10 – holding four stocks in the portfolio – one trade a day nets 27784 and four trades a day nets 17255. Both produced about a 55% win rate. Two points – with one trade a day you are sometimes holding 20000 shares of some low price piece of crap and I can pretty much guarantee you aren’t going to get out of it alive unless it goes way, way up. Second, it is obvious that low priced stocks are low priced for a reason.
Exception two – note that the trades selected in the above tests were all “highest volume of the set that was output by the filter daily.” Volume matters when selecting trade candidates. Low volume stocks are low volume for a reason.
Also no Turtles were injured during the filming of this report.
Think about it people – if I give you a set of trading rules that I know has a probability of success of say 55%, plus a very large trading account and I set you in front of a computer terminal all day and you follow those rules perfectly then you should have a 55% probability of success. And if you don’t follow the rules, you won’t. Good grief - for this they needed a test? I’m probably missing something as usual. Anyway if you go to your favorite search engine and you type in Turtle you will probably find the rules. They ain’t much and they have been sold for thousands and thousands of dollars but are now available for free. And there is a reason for that. They basically say buy the trend and when it stops trending sell it - Duh. I can prove this and will a little bit later in the post.
But let’s not belittle the process itself. The most important part of Turtling is risk management. That’s because risk management is the most important thing a trader can do. Above all preserve capital until you hit the home run. Something we day traders sneer at and other people make fortunes doing.
Now the proof. I pulled out one of my favorite tried and true, time tested – oh crap – I threw together a filter that simply selected stocks based on closing price in relation to 2 moving averages and then listed them in volume order highest to lowest. It is a very prolific filter because there are so few moving parts but we only use the top several selections by volume in our back testing so it doesn’t really matter how many are output. I expect this filter to produce about 60% winners over a four-day period – the expectation is based on experience.
When I back test I use a simple baseline – One trade a day with no more than four stocks in the portfolio at any time. The way I have the back test exit criteria set is also simple – after four days exit the trade and take a new one unless you hit a 10% stop loss before 4 days in which case sell that stock and take a new one. Based on a virtual equity starting capital position of $100K this permits 25K per trade to begin. Note that I don’t care how many shares I’m holding just how much money I’ve got in them. After I run the test over an 80-day period or so I look at the equity results and for this baseline we achieved $53860 in profit with a 67% win rate – not bad for 80 days or so.
After establishing the baseline I started mucking with the internals one at a time. First number of stocks in the portfolio. I adjusted this to 1, 2, 8 and 16 stocks in the portfolio at any one time and here are the results.
1 – 47865
2 – 46147
8 – 24325
16 – 11570
This suggests that there is an optimum number of stocks in the portfolio and for this test it seems to be four. (If we doubled our starting cash to 200K would that change the number of stocks in the portfolio factor)? Let’s now look at more than one trade a day with no more than four stocks in the portfolio. (If you answered “yes” to our question above you are correct).
If we take four trades a day with four in the portfolio we can get to 60249 with a 71% win rate. For two trades a day with four in the portfolio we achieve 54891 and 68% win rate which is still better than one a day with four in the portfolio.
Let’s try one more factor – let’s take our best result, 4 x 4, and increase our holding period to 10 days. The result was 24357. If we add a simple measure such as a 15% profit exit criterion it goes to 34334. This is probably trying to tell us that stocks go up and down and unless you are a buy and hold forever type you need to attend to your portfolio a little more often than the broker recommended once per year.
So with a makeshift filter I have shown how money management – changing the way you buy not what you buy can make a large impact on your results.
According to song and story those Turtles who became successful adhered to the process like glue. The ones who deviated from the rules were not successful. It is that simple, kiddo.
There are a couple of exceptions to the above regarding “what to trade”. For those of you who are enamored of low price stocks (less than $10) – same system with a dollar requirement of $2 to 10 – holding four stocks in the portfolio – one trade a day nets 27784 and four trades a day nets 17255. Both produced about a 55% win rate. Two points – with one trade a day you are sometimes holding 20000 shares of some low price piece of crap and I can pretty much guarantee you aren’t going to get out of it alive unless it goes way, way up. Second, it is obvious that low priced stocks are low priced for a reason.
Exception two – note that the trades selected in the above tests were all “highest volume of the set that was output by the filter daily.” Volume matters when selecting trade candidates. Low volume stocks are low volume for a reason.
Also no Turtles were injured during the filming of this report.
Subscribe to:
Posts (Atom)
