X-Ray Your Trading System

Chart Forum » Trading - Systems » X-Ray Your Trading System

«« Previous Next »»

Thread Start New Thread
Last Poster Posts Pages Last Post

Author

Message

alann

Member

Post Number: 51
Registered: 05-2004

Rating:
Votes: 6


Thursday, January 20, 2005 - 03:45 am:

X-Ray Your Trading System

Just as an x-ray tells your doctor what lies below the surface of your skin so does the Profit Factor (PF) tell the trader what lies below the surface of his trading system. At the first layer below the surface of your trading system the Profit Factor tells whether your trading system is profitable or not and is calculated by dividing the total gains from all your trades by the total losses from all your trade. If the Profit Factor then equals 1 or less you should not be trading your trading system. A Profit Factor of one (PF=1) tells you that your trading system produces an overall loss that is equal your overall gains and as a result, there is no profit to be made. A Profit Factor less than one (PF<1)>1) it tells you that your trading system generates a profit. A Profit Factor of one and one-half (PF=1½) tells you that you have an average trading system. A Profit Factor of two (PF=2) tells you that your trading system is above average, and a Profit Factor of three of greater (PF=>3) tells you that you have an exceptionally good trading system and everyone would love to get there hands on it.

Besides informing the trader how good a trading system he or she has, the Profit Factor tells the trader how much money he or she is going to make by using this trading system. A Profit Factor of two tells the trader that for every dollar his or her system losses it makes a profit of two dollars, or a gain of one hundred percent (100%). A Profit Factor of three tells the trader that for every dollar his or her system losses it makes a profit of three dollars for a gain of two hundred percent (200%). The Profit Factor also tells the trader what the drawdown is going to be. For a system with a Profit Factor of two, for every two dollars of profit the system makes the traders trading account will suffer a one-dollar loss or a drawdown of one dollar.

For a first level of X-Ray below the surface of your trading system the Profit Factor tells the trader whether he has a good trading system, an excellent trading system or a poor trading system that should not be used. It also tells the trader how much profit his or her trading system will generate and what the drawdown on his or her account will be.

By increasing the power of the X-Ray it is possible to look deeper below the surface of the trading system where we can determine what parts of the system are healthy and what parts of the system are sick and in need of medication. By increasing the power of the X-Ray even higher it is possible to look even deeper into the trading system to determine things such as how many losses in a row a trader will have to endure, what the worse case drawdown will be, as well as how many profitable trades in a row the trader can experience. However, this will be left for another post.

Take an X-Ray of your trading system to determine its health; you owe it to yourself as a trader. Take all your individual trades and calculate their gain or loss in percent, then add up all your gains and add up all your losses. Finally, divide the total gains from the total losses to determine the Profit Factor. Save all these calculations, as we will use them later when we X-Ray further below the surface of our trading systems. If you would like to Post your Profit Factors please feel free to do so and they will be tabulated into a table.

(Message edited by Alann on January 20, 2005)

Alann,

I cannot predict the future I can only follow the trend. Why? Because the trend is my friend.

perler59

Member

Post Number: 506
Registered: 09-2003

Rating: N/A
Votes: 0


Friday, January 21, 2005 - 06:54 pm:

Since June 2004, my profit factor is 2.4 including open trades.

I can NOT control the markets, so I MUST control myself.

alann

Member

Post Number: 52
Registered: 05-2004

Rating: N/A
Votes: 0


Sunday, January 23, 2005 - 02:28 am:

This thread is probably more of a tutorial than a normal chat so I do not expect many contributions in the way of POST's to this thread. However, I would like to know the overall interest level. If you are interested in this thread and would like to see it continue then please respond by your vote. If there is no apparent interest I will drop this thread. Thanks!

Alann,

I cannot predict the future I can only follow the trend. Why? Because the trend is my friend.

perler59

Member

Post Number: 509
Registered: 09-2003

Rating: N/A
Votes: 0


Sunday, January 23, 2005 - 08:04 am:

Alan, yes, more please :-)

I can NOT control the markets, so I MUST control myself.

ken

Member

Post Number: 88
Registered: 04-2003

Rating: N/A
Votes: 0


Sunday, January 23, 2005 - 10:34 am:

alann,

I have analysed mine since June also - 1.70 was the answer.

I have recently been reading Elder so I analysed each trade to see whether the 13 week and 26 week ema's were both rising - for those that were, the performance was 2.42. Not hard to decide what a permanent part of my setup will be from now on. I will analyse on other criteria and may post again.

Keep it coming.

Ken

Price is the leader of the market crowd. (Elder)
Members of the crowd follow the leader and experience the same emotions as each other.
To be independent of the crowd we must not change our behaviour with price.

stevo

Member

Post Number: 253
Registered: 01-2003

Rating: N/A
Votes: 0


Sunday, January 23, 2005 - 12:18 pm:

Alan
On trades closed in the last 2 years, plus current open trades, profit factor equals 4.55.

stevo

perler59

Member

Post Number: 510
Registered: 09-2003

Rating: N/A
Votes: 0


Sunday, January 23, 2005 - 12:57 pm:

Stevo, that is amazing! Care to share any of your secrets? I understand if you don't :-)

I can NOT control the markets, so I MUST control myself.

stevo

Member

Post Number: 254
Registered: 01-2003

Rating: N/A
Votes: 0


Sunday, January 23, 2005 - 10:08 pm:

Perler
My secrets??? >>> lots of testing and backtesting, trial and error, and consistent trading, plus some money to trade with. Get a good portfolio back tester and good quality data and just start testing. This approach works for me.

I really don't understand why relative few traders do any really intensive backtesting. I see posts on this site of all sorts of systems. People seem to follow these methods without any serious attempt to see if the method actually works - many don't. Beats me!

I would think that there are a few longer term traders on this site that would have similar, or even better, profit factor stats. Longer term trend following trading styles should achieve these sort of profit factor stats, especially in the Aussie market over the last 2 years.

My average trade length is around 190 days, but winners last over 5 times longer than losers. This is longer term trend following.

Profit factor on closed trades is only just above 3. I have a couple of good size winners that have been running for well over a year, plus some other nice open trades.

I know that I can do better than these results, I was a little tentative in 2003. I was too conservative when it comes to risk. Longer term systems take some time to reveal themselves.

It would be good to see some more profit factor stats from some more traders.

stevo

davkell

Member

Post Number: 28
Registered: 07-2004

Rating: N/A
Votes: 0


Sunday, January 23, 2005 - 10:44 pm:

Hey Alann;

I've just calculated my PF on my initial backtesting for my system and it is a healthy 6.35. Now this is not real trading, is only based on 18months of backtesting, but is a promising number for determining whether to trade for real, don't you think?

"Trade Your Way To Financial Freedom"

alann

Member

Post Number: 54
Registered: 05-2004

Rating:
Votes: 2


Tuesday, January 25, 2005 - 04:43 am:

One thing I should have brought out in my initial POST on X-Raying your trading system is that to have a statistically acceptable Profit Factor your calculations should be based upon at least if not more than 30 closed out positions that stretch over a period of time that encompasses the complete market environment that your trading system is designed to work in. If this data is not available then all you can do is to calculate the PF based upon the closed out positions you have knowing that in the long run the profit factor will be somewhat different. It follows the law of averages in that with time the Profit Factor will settle down at its true average level. For example, my trading system Profit Factor is 3.3 based upon more than 600 closed out positions over the last five years. As a result it doesn't hardly vary at all as new closed out positions are added.

Also, if you want to know what the Profit Factor is including open positions that is alright, but the more accurate calculation is to use the closed out trades as this reflects your complete trading system including entry, exit, and your discipline, etc.

Overall, I would say that everyone has a good trading system and you all are to be complimented for the efforts you have put into them.

In the following POST's we will pick apart our systems alittle by disecting the Profit Factor to learn more about our systems and what we might be able to do to improve them even further. Stay tuned!

(Message edited by Alann on January 25, 2005)

Alann,

I cannot predict the future I can only follow the trend. Why? Because the trend is my friend.

stevo

Member

Post Number: 255
Registered: 01-2003

Rating:
Votes: 2


Tuesday, January 25, 2005 - 09:28 am:

Alann

I find it good to look at both closed and open positions when looking at profit factor because I am a longer term trader. Over a 2 year period some of the biggest winning trades are still open trades. Whilst these open gains are not yet realised it is difficult to ignore them. That is why I posted both open and closed profit factor above. A shorter term trader would not have the same sort of variance on this stat as a longer term trader might have.

For a longer term trader 2 years is really not a lot of time. It is not only the number of trades but the period of time the trader has been running the system. I would like to get at 5 years under my belt to see how the systems perform in different market conditions and get a more realistic longer term profit factor. Backtesting can provide a lot of confidence in this area though.

In terms of number of trades it is also not quite as clear cut for some traders. I have made 94 trades on 69 different stocks. However I might buy/sell the same stock more than once over a week for the same buy/sell trigger. What looks like a separate trading event might just be a repeat buy on the same trading trade trigger to get the position size required.

Whilst profit factor is an important statistic for traders to look at what about all the other basic stats that give a trader an idea of what to expect, such as profit index, CAR, % winners, expectation and drawdown? Not to mention the various graphical representations that traders can use to evaluate their trading. And let's not ignore dollars in the bank and tradeability of the system!

What else do you use to evaluate your trading?

regards,
stevo

alann

Member

Post Number: 55
Registered: 05-2004

Rating: N/A
Votes: 0


Wednesday, January 26, 2005 - 02:26 am:

Stevo

I agree with everything you said and I was not picking on you for including open positions in calculating the PF as I do a similar thing. I calculate the PF of just my open positions, as well calculate the PF of my closed positions. There is nothing wrong with including your open positions along with your closed positions in calculating your PF. My point about calculating the PF using just the closed positions is that it is the more accurate measure of the trading system. However, having said that I also realize that calculating the PF of just the closed positions is also not very accurate if it does not include sufficient length of time and a large enough sample of closed out positions.

I also agree that it is not only the number of trades but the period of time the trader has been running the system. That is what I meant when I said the the closed out positions need to stretch over a period of time that encompasses the complete market environment that your trading system is designed to work in, ie bull, bear, and trading ranges. But the number of trades is also important in determining how good a trading system is. When the PF is calculated on just a few trades it is an estimation of how good a trading system is. When the PF is calculated on a large number of trades it becomes a more accurate measure of the trading system. The greater the sample the greater the accuracy. Of course there is a point where having more closed trades to use in the calculation does not increase the accuracy that much more. The fact that you have 94 trades is a good statistical sample in determining your trading system. You obviously have a professional trading system on your hands and I am sure that it will continue to do very well as time goes by. I say professional because not all professional traders have a trading system with a PF of 3 or greater and yours certainly does. Congratulations!

As for other measures used to determine the value of a trading system there certainly are many ways of looking at your system just as there are many technical indicators to use in looking at the current condition or a stock, but that does not mean we use them all. In trading I only use a few technical indicators and in determining the value of a trading system I only use one, the profit factor. Some of the measures you point out are actually derivatives (fall out) of the PF and some are actually part of the PF as you will see when we break the PF down into its individual parts. So hang on, we'll get there...

Alann,

I cannot predict the future I can only follow the trend. Why? Because the trend is my friend.

alann

Member

Post Number: 57
Registered: 05-2004

Rating:
Votes: 1


Friday, January 28, 2005 - 01:43 am:

X-Ray Your Trading System -- Part 2

Just as the X-Ray machine allows doctors to look at the health of a patient by looking beneath the skin so does the Profit Factor allow the trader to look below the surface of his or her trading system in determining its health. When we turn up the X-Ray power of the Profit Factor a little higher, we are able to look a little deeper into the health of our trading system and we are able to see that the Profit Factor is made up of two separate parts. For now, we will look at the first of these two parts.

One of the parts that make up the Profit Factor is the number of winning trades divided by the number of losing trades (winners/losers). When we look at the stock market using a statistically large enough sample of all the stocks that make up the stock market, we see that 50% of all days are winners and 50% of all days are losers. Apply the same analysis to weeks and we see that 50% of all weeks are winners and 50% of all weeks are loser. The same applies to months. 50% of all months are winners and 50% of all months are losers as well. When we look at the closed out trades in our own trading accounts we should expect that at least 50% of our closed out trades to be winners and the other 50% losers. Now some trading systems will generate more winning closed out trades than losers, while other trading systems will generate more losing trades than winners. Although it is possible for a trading system that generates more losers than winners and still have a Profit Factor greater than 1.5 a healthy trading system will generate at least 50% winners if not more. I know of some trading systems that generate 60% to 75% winners. The reason why it healthy for a trading system to generate at least 50% winners is because the number of profitable trades in a row increases while the number of losing trades in a row decreases.

For example, a trading system that generates 75% winners and only 25% losers has a probability of generating three winners in a row 42% of the time. This means that out of 100 trades there will be 42 times when this system will generate three winning trades in a row. It will also generate three losses in a row less than 2% of the time. Or, for every 100 trades this system will generate three losses in a row less than two times.

However, what about a trading system that only generates 40% winners. A system that generates only 40% winners has a probability of only generating three wins in a row 6% of the time. So for 100 trades this system can only expect to generate three winning trades in a row 6 times compared to the 42 times for a system that generates 75% winners. On the other hand, the system that generates only 40% winners also has a probability of 21% of generating three losses in a row. Therefore, for every 100 trades this system would generate three losses in a row 21 times. Compare this to only two times for a system that generates 75% winners.

Therefore, although a system that only generates 40 winners out of 100 trades can have a winning profit factor (PF=1.5 for example) a system like this is really working hard to be a winning system.

So, what are some of the things that could be keeping a trading system from generating more wins than losses?

#1. It could be your entry. If your entry is not correct, you could be going long when you should be going short the majority of the time. Or, you may be going short the majority of the time when you should be going long.

#2. If could be your exit. You could be holding positions too long hoping to squeeze out that extra little bit of profit. Then when that extra little profit doesn’t come along and your trade starts going against you, you keep holding on thinking that it will turn back around and give you the profit you were looking for, only to hold on too long till the gains erode and the pain get unbearable and you close your trade for a loss.

#3. You could be holding on too long. For example, out of the more than 600 closed out trades I have in my most mature trading system my average holding time is 8 days. When my trades to hit my profit target in 8 days I start analyzing my stocks very heavily for weakness and if I see just the slightest amount of weakness I get out before the trade turns into a loss.

There are many reasons why a trading system could be unhealthy even though it still generates an overall profit for your account by having a Profit Factor greater than 1. Examine your trading system and determine its win versus loss ratio. When dividing your total wins by total loss you should be looking for a ratio of one or greater (wins/loss >=1).

In a later POST I will post a couple of tables that will give you the probability of the number of wins and losses in a row that you can expect based upon your systems win loss ratio.

Alann,

I cannot predict the future I can only follow the trend. Why? Because the trend is my friend.

stevo

Member

Post Number: 258
Registered: 01-2003

Rating: N/A
Votes: 0


Saturday, January 29, 2005 - 08:22 am:

Alann
Thanks for the post - it has got me thinking :-)

. The results you have had over 600 trades / 5 years are exceptional.

We better make sure that we are talking about the same profit factor.

In the Amibroker Profit Factor is defined as the profit of winners divided by loss of losers.

This definition doesn't look at "the number of winning trades divided by the number of losing trades (winners/losers). " It only looks at the dollars.

Are you using the same Profit Factor as Amibroker uses?

In terms of number of winners to losers I think that this can be a real red herring for longer term traders, although it can be important to shorter term traders.

This is because the shorter term trader often does not have the big winners that a longer term trader is looking for. So the short term trader will, on average, lose as much as he/she wins per trade, but relies on 60% or more winners to make money.

Often whilst the longer term trader rides a trend for over a year a shorter term trader makes multiple trades on the same trend to get a higher win rate, but probably no more, and often less dollars due to transaction fees slippage and increased risk of errors.

If one chases % winners then one tends to trade shorter term, although there are always (many) exceptions to the rule.

As profit factor doesn't consider the number of winners but only concentrates the profits and losses it shows how effective the system is in this area.

Whilst I agree that profit factor is a great measure of system / trader performance it doesn't tell us what the road was like that the trader travelled. The ups and downs of the system can only come from looking at equity curves and draw down.

Although I suppose if profit factor was plotted over time it would give a very good idea of the path the trader took. That would be nice.

regards

stevo

david_louisson

Member

Post Number: 31
Registered: 02-2004

Rating:
Votes: 1


Saturday, January 29, 2005 - 10:53 am:

Hi folks

Excellent thought-provoking thread, IMHO.

I expect that how ‘tight’ one sets one’s initial stop loss could have a significant bearing. A tighter stop means that one would (at least in theory) be stopped out of some eventual winners prematurely, thereby registering a loss instead of a win. The up-side to this approach being that all losses are kept small. In other words, tight stops should decrease the ratio of winning trades to losing trades, but increase the average win size compared to the average loss size. Mathematically:

Numerator of PF = total winnings = number of winning trades x average win size
Denominator of PF = total losses = number of losing trades x average loss size

I agree that one needs to take costs into account, i.e. to ensure a valid result, it is the NET outcome of each constituent trade that should be considered.

The ratio of winning trades to losing trades has a direct bearing on cash flow. If one is trading infrequently (e.g. longer term, low diversification, strict set-up criteria aimed at catching only the ‘high probability’ trades), then I expect it becomes more significant, i.e. even a small losing streak could mean poor cash flow.

One further point: PF measures the effectiveness of entries and exits, but it is ultimately the allocation of capital to each trade (i.e. position sizing) that determines one’s overall return, and drawdown. Position sizing and PF are completely independent variables.

Overall return is proportional to PF, and also proportional to average position size.
Drawdown is inversely proportional to PF, but directly proportional to average position size.

Alann, I also enjoyed your point about the three successive wins or losses: everything else being equal, the sequences of WIN-LOSS-WIN-LOSS-WIN-LOSS will yield a similar account balance to WIN-WIN-WIN-LOSS-LOSS-LOSS or even LOSS-LOSS-LOSS-WIN-WIN-WIN, but (if your position sizes are too high) the latter sequence could wipe you out before you even get started. Again demonstrates the importance of tight capital management.

To illustrate some of the above points, there’s a trader at our TA group who has almost doubled his starting account in his first month, trading Aussie stock CFDs with CMC. He sets very tight stops, and advances the stop to the break-even point at the earliest possible opportunity. His entry system is sufficiently good that his ratio of winners to losers is also high (I will ask him for some exact figures). He uses the leverage available to operate perhaps 6-8 simultaneous diversified positions, the majority of which he holds for less than a few days, and whose total capital requirement would otherwise be around 5-10 times his account size. I suspect that this level of position sizing puts him in the ‘high risk’ category of traders. It will be interesting to see if subsequent months are as successful as his first.

Happy New Year to all!
David

smallworld

Member

Post Number: 131
Registered: 01-2004

Rating: N/A
Votes: 0


Saturday, January 29, 2005 - 11:40 pm:

Expectancy rings a bell here?
Expectancy ($) = %Winners * AvgProfit - %Losers * AvgLoss
or Expectancy (per unit risk) = Expectancy / Avg max risk per trade;

WIN/LOSS has a special place for me in system development (based on Lucas and Le Beau.... quoted by Chande which I read......) in that it can be used to check if the entry is better than random. test the entry signal with exit on the close of the ra-th day, without any stops, and no deductions for slippage and commissions.

Just a thought or two.

alann

Member

Post Number: 58
Registered: 05-2004

Rating:
Votes: 1


Tuesday, February 01, 2005 - 01:33 am:

Stevo

Yes it is the same Profit Factor as used with AMIBroker. The only difference is that I express the profit in percent instead of dollars but that does not change the end result. (See the last paragraph in the very first POST of this thread.)

What AMIBroker does not bring to light is that the Profit Factor can be broken down into its individual mathmatical terms.

For example:

X = a x b (a times b)

if a = 2 and b = 1.5 then X = 3

Relating this to the Profit Factor X = PF and a = winners/losers (winners divided by losers).

So the Profit Factor is as follows:

PF = (winners/losers) x (b)

So if when you divide your winners by your losers and you get, for example 2, then:

PF = 2 x b

I will define what the variable "b" is equal to in X-Ray Your Trading System -- Part 3.

The reason why it is important to break the Profit Factor down into its individual terms will come to light in my following POSTs. But for now, it is important to calculate the ratio of winners to losers because the result has a direct influence on the number of winning trades and losing trades that you can expect to have in a row using your current trading system. The higher this ratio is the more winners you will have in a row and the less losers you will have in a row. So if the ratio of winners/losers is low it informs the trader that he should try to improve his or her trading system or at least to understand why his or her trading system generates so many losses in a row.

Hope this helps. Hang in there, and stay tuned...

Alann,

I cannot predict the future I can only follow the trend. Why? Because the trend is my friend.

captain_chaza

Member

Post Number: 1058
Registered: 02-2003

Rating:
Votes: 2


Thursday, February 03, 2005 - 05:00 pm:

Ahoy Sea-cadet Alan with 2n's

Firstly, may I congratulate you on a very good thread
However, surely the bottom "$" line and the "x" # of days is the paramount factor
(IE To make best use of our limited resources and compound in good times)

The problem as I see it with your premise is that you assume the the world is flat and the winds will always blow from the East or from another direction

Not only that but they will blow at a consistent rate throughout any 5 year voyage

"No two 5year terms are ever quite the same"
Compounding at the beginning or nearing the end of any voyage will show very different results

You also assume that you will be always on deck to manage your craft
And that
All you have to do is set in place a foolproof PLAN in a chaotic enviroment and your last 5year figures will look after themselves

This to me is a very academic view of life on land and even more so when living upon the sea on any exchange

The bottom line divided by the # of trading days is the true challenge
Freeing up capital to do it again or just playing it safe

I know I am no good on land
I said "Nobody would ever pay for bottled water in Aust"
"that was a really bad bet!"
"Sliced bread was not nearly as good as Unsliced and it would never sell!"
"Same for sliced cheese"

Yes! I lost all 3 bets

Anyway
Keep the good work up
You are either on to something or nothing at all
Such is Genius!

capn

"While we stop and think, we often miss our opportunity." Publilius Syrus, 1st century B.C.

"I believe the future is only the past again, entered through another gate."
Sir Arthur Wing Pinero 1893

"There are two times in a man's life when he should not speculate: When he can't afford it, and when he can." Mark Twain, 1897

captain_chaza

Member

Post Number: 1060
Registered: 02-2003

Rating:
Votes: 2


Friday, February 04, 2005 - 07:08 pm:

Ahoy Officer Alan with 2n's

I was only joking re the rank of Sea-cadet
Nobody ever (except Admiral Twiggy) takes me seriously

What I mean about the 5 year practice drill can be easily seen if you do all the 5 year practice drills over the last 100 years

This data cost me $5.00 /year in the early 90's

It was the best money I ever spent on data but unfortunately my data supplier went belly up after the arrival of the 2YK
Bugger!! (Research Technology)

If you click on each 5 year term on the CHARLIE DOW and his 1st mate JONES VOYAGE over the last century (20 charts) you will see that no 5 year voyages are really the same

Some can be horrible/horrific experiences while others can be breathtaking!
Surely in rising sea and improving wind/weather conditions there will be more winners and conversely in shit seas and unfavourable weather conditions there will be fewer wins/gains

I pass this private information on to you and "Incredible Charts " BUT warn you ALL to respect the laws on land of copyright (C)

I hope you Enjoy these practice drills, If you have not already participated in the past in each 5 year ocean trial !!!

Just go to my home page

home.iprimus.com.au/ckirzner
Then go to
"The Greatest Sport of All"
Then do your own 5year practice drillls and ocean trials

Bon Voyage and Gods speed
Captain Chaza
capn

Apologies for not finishing the site but things have been really hectic over the last 8 years

I am still trying to work out
How brave is too brave !!!
and
How safe is too safe !!

As my 1st Officer Spids used to say
"It depends on whether it is your money or theirs?"

PS Once I work it out I'll probably go directly to my grave without passing GO!

captain_chaza

Member

Post Number: 1061
Registered: 02-2003

Rating:
Votes: 2


Friday, February 04, 2005 - 10:05 pm:

Come to think of it, 1st Officer Spids,

You have been sailing around with their money for more years than I would like to remember
When do you think is the right time to change the ownership of those winnings/dollars and call it your own?

Maybe then and only then will we see the true calibre of your seamanship skills at the helm and at your best

Give a man an inch or a Sucker an even break I say and guess what?
They'll take it!

Only Landlubbers buy the worst house in the worst condition, in the worst street and in the worst suburb!!
If they bought the worst ship in her worst condition to challenge the roughest sea and weather conditions ?
I am sure they would NEVER EVER go a Sailing again!

I have always tried to teach my sons that PRICE is not everything
"It's all about
CONDITION! CONDITION! CONDITION!"
and that
"If you ever see a fork in the road?
TAKE IT!"

capn

Thinking a forks "Claudius//Hillarius"
Has an Index ever gone belly up?
Made a call,
Milked the public with unbelievable offerings, fluffy toys and loyalty points?
Lots to think about if one had a couple of long terms left in their lifetime

Unfortunately for me at my age there is only the short term!
Such is Life!

alann

Member

Post Number: 70
Registered: 05-2004

Rating: N/A
Votes: 0


Saturday, February 05, 2005 - 08:38 am:

Gentlemen,

It is my assumption that no one is truly interested in this thread and therefore I had decided not to continue it. If my assumptions are in fact incorrect please let me know. Otherwise this will be my last POST in this thread.

Alann,

I cannot predict the future I can only follow the trend. Why? Because the trend is my friend.

hilarius

Member

Post Number: 468
Registered: 04-2004

Rating: N/A
Votes: 0


Saturday, February 05, 2005 - 08:52 am:

Why end the thread?

I am still seeking to learn how to increase reward while reducing risk with the aid of after the event analysis

At present my position is that pre-event research is paramount and after the event analysis reveals little of added value

Stops give me the necessary exits to try again with better pre-event research

So what is the source of that Holy Grail ... minimised risk ... maximised reward?

Or is it the case that larger risks are compulsory for larger rewards?

Hilarius

PS If it can be shown that after the event analysis creates greater insight into pre-event improvements in research and more frequent use of those improvements then I can start to see some merit in post-event analysis

As long as it is not so complex and time consuming that it detracts from improving pre-event research

I come in peace to share my thoughts and to shine my candle light on possible long term opportunities

david_louisson

Member

Post Number: 32
Registered: 02-2004

Rating:
Votes: 1


Saturday, February 05, 2005 - 02:28 pm:

Alann – please continue the thread. I, for one, was enjoying the read.

Hilarius –

The following are mathematically based options for increasing return. Of course, the real-life value of theories such as these will be eroded by whatever extent the markets are driven haphazardly by emotion (as opposed to the precise probabilities of casino-type games).

1. Increase return by simply increasing position size. But this also increases risk (drawdown) in much the same proportion.

2. Increase return by improving your system of entries and exits, i.e. increasing the PF that Alann has been talking about, by whatever your chosen method (e.g. back-testing using indicators, trendlines, chart patterns, etc; discretionary; astrological, etc). A higher PF both increases return and decreases risk.

3. Increase return by diversification. Mathematically, X simultaneous events will increase the rate of return X times (assuming overall system expectancy is positive, i.e. PF > 1), WITHOUT ALTERING RISK, provided that the events are completely INDEPENDENT. The key word is independent. Hence the beneficial effect will be eroded to the extent to which the events are correlated (as stocks, markets, etc tend to move in harmony with each other).

4. There may be others, perhaps......?

If you do find the 'Holy Grail', please let us all know!!!

David

hilarius

Member

Post Number: 472
Registered: 04-2004

Rating:
Votes: 1


Saturday, February 05, 2005 - 02:52 pm:

Hi David

Great to hear from you and thanks for the excellent points

By the tone of your response it appears you have overcome your earlier crisis of faith with trading the markets?

Hilarius

I come in peace to share my thoughts and to shine my candle light on possible long term opportunities

spider

Member

Post Number: 1907
Registered: 10-2002

Rating:
Votes: 2


Saturday, February 05, 2005 - 05:50 pm:

Come to think of it, 1st Officer Spids,

You have been sailing around with their money for more years than I would like to remember
When do you think is the right time to change the ownership of those winnings/dollars and call it your own?

Ahoy captain.

Very interesting question.
I have touched on the answers a few times in the past but, as I cannot find half my stuff I don't expect others to.

In short:- EXIT strategies equal profit.

It might sound strange but, where others are looking for reasons to get out , I'm looking for reasons to stay in.

Entries are not that hard, you just jump into anything that fits your rules, most roll over but your stops protect you, the ones that survive are your cash cows.
Milk them till they go dry.
Know before you go in what an exit will look like.
When you see it, pull the trigger and move on.
Get back in if you think that it is worth it, but mostly you will move on to the next opportunity.

This only works if you don't need the daily excitement of jumping in and out of the market.

I'm in the top 5% of profitable traders with my internet broker but they don't consider me a special client as I don't trade that often! Brokers don't care if you make money or not, they only want you to trade (often).
spider.

hilarius

Member

Post Number: 473
Registered: 04-2004

Rating:
Votes: 3


Saturday, February 05, 2005 - 06:01 pm:

Spider

Please remind me again ...

What does your exit look like?

A lower high followed by a lower low followed by a lower high?

Or something well ahead of that?

Hilarius

I come in peace to share my thoughts and to shine my candle light on possible long term opportunities

david_louisson

Member

Post Number: 33
Registered: 02-2004

Rating: N/A
Votes: 0


Sunday, February 06, 2005 - 02:53 am:

Hilarius

Many thanks for your kind comments. No credit to me, however, all I am doing is regurgitating textbook material.

Still non-committal as to whether I might one day take on the markets. My state of mind has not been helped by recent ill-health.

Anyway, thanks again
David

staybaker

Member

Post Number: 48
Registered: 03-2003

Rating: N/A
Votes: 0


Thursday, March 03, 2005 - 09:40 pm:

Alann wrote:

Yes it is the same Profit Factor as used with xxx software. The only difference is that I express the profit in percent instead of dollars but that does not change the end result.

This seems to me to be wrong, if I am interpreting things correctly. For example, suppose you made two trades. For the first trade, you invested $50,000 and made 50%, or $25,000. For the second trade, you invested $10,000 and lost 50%, or $5,000. Now, using Stevo's method of comparing dollars, your profit/loss ratio will be $25,000 / $5,000 or 5.0. But expressing everything in percentage terms, you'll have a profit of 50% and a loss of 50%, giving a ratio of exactly 1.0. So the end result is VERY different, depending on the method used. It also seems clear that the dollar method makes much more sense.

I wonder if Alann is coming back? This was an interesting discussion ...

Cheers, Staybaker.

alann

Member

Post Number: 77
Registered: 05-2004

Rating: N/A
Votes: 0


Friday, March 04, 2005 - 12:38 am:

staybaker

Yes you are correct. It is more accurate to express the gain and loss in percent when calculating the Profit Factor that is why I made mention of the fact that I do calculate the PF based on percent and not dollars. What I meant by it does not change the end result was with respect to the Profit Factor I use versus the one in xxx software as being the same PF. The assumption I made with respect to the xxx software was that the PF was calculated on only one trade in which case it would not make a difference if you use dollars or percent. Now that I think about it my assumption was most likely in error. Now that I read your post I can see where I could have chosen my words better. When using the results of many trades it becomes necessary to convert from dollars to percent so as to give each trades profit or loss equal weight.

Thanks for bringing this to light!

The reason I discontinued the thread was because all the feedback was one of disagreements. There did not seem to be anyone really interested in learning from the thread. Thus I felt I was not being productive and discontinued the thread. However, if there is still some real interest in the thread I may pick back it back up. So let me hear from the masses.

(Message edited by Alann on March 04, 2005)

Alann,

I cannot predict the future I can only see possibilities.

stevo

Member

Post Number: 275
Registered: 01-2003

Rating: N/A
Votes: 0


Friday, March 04, 2005 - 08:35 pm:

Alann
"When using the results of many trades it becomes necessary to convert from dollars to percent so as to give each trades profit or loss equal weight."

I can get funny numbers in a spreadsheet if I take a profit on a stock I received for zero dollars or very cheaply, even if I received only a handful - so I would have to ignore these.

Also, ignore my ignorance, but how exactly do you convert to percentages? Do you take the sum of % profit/loss for winners and losers as below?

PF = (sum % winners) / (sum % losers)

If I do this I get %PF of 6.3 versus $PF of 4.55 over 94 trades. I feel a little uneasy using the % approach since I can vary position size considerably from trade to trade depending on a number of factors, such as liquidity and risk.

So I am likely to lose more percentage-wise on smaller riskier trades, but make more dollars on on larger less risky positions, but not in percentage terms. Hope that makes sense! For example I could lose $10,000 on a $20,000 trade (50% loser), but make $200,000 on a $100,000 trade (100% winner). I am "happy" to lose the $10,000 if I get a shot at some 100% plus winners that bring in ten times plus the loss on the small trade. The example is a little extreme but illustrates what I am getting at.

I have been going over the posts above and really would like to get part 3. The relationships you talk about are important for (some) traders and probably hold true whether we are using % or $. I would think that expectancy might be in the numbers as well as smallworld mentions above. I like a little bit of theory.

regards

stevo

moleman

Member

Post Number: 105
Registered: 10-2003

Rating: N/A
Votes: 0


Saturday, March 05, 2005 - 09:54 am:

Hi Alann

As you can see there's a lot of interest in this thread so please continue! I think the "disagreements" have come from people's concerns with possible deficiencies they can see in the PF measure as presented so far. It doesn't seem to take the position size of different trades into account leading to wildly different results, as has been illustrated with examples. Also I can see a good measure of a trading system would need to take time into account, including the time each trade was held. eg if I told you my trading system yielded 100% you'd say Wow! Then if I told you it took 25 years for me to do that your reaction would be different... perhaps there are some assumptions on the use of PF we've yet to see?

I'm also looking forward to part 3.

cheers

moleman

shutty

Member

Post Number: 42
Registered: 12-2003

Rating: N/A
Votes: 0


Saturday, March 05, 2005 - 08:11 pm:

> I'm also looking forward to part 3.

Me too!

Cheers Shutty

deanrosario

Member

Post Number: 569
Registered: 11-2002

Rating: N/A
Votes: 0


Sunday, March 06, 2005 - 04:05 pm:

This is a very useful thread for all of us who are running a "trading business".

As with any business, I guess "creative accounting" can conjure up many definitions of "profitability" and no one is necessarily right or wrong.

I've found the following definitions very useful:

Win/Loss ratio = (average dollar quantum of wins) / (average dollar quantum of losses); will be expressed as a percentage

Win Probability = (number of wins) / (total number of trades)

Additionally, the following tool provides a nice visualisation of how profitability changes with each parameter.

http://hquotes.com/tradehard/simulator.html

Two other important parameters to me are:

a) Duration of the trade

b) Number of trades

a) Duration of the trade

Given that money has a "time value", surely time is a critical parameter when comparing trading systems.

E.g. 2 trading systems produce
- the same Win/Loss ratio and
- the same Win Probability

However, the average duration of trades with System A is 1 year; but the average duration of trades with System B is 1 hour.

Prima facie, which system would you rather trade? I say "prima facie" because, the following point needs to be considered, too.

b) Number of trades

Obviously if two trading systems have

- the same Win/Loss ratio and
- the same Win Probability and
- the same Duration

then, the final differentiator would be "number of trades".

If System A produced one "trigger" every day but System B produced one "trigger" every week then System A will deliver more dollars at the end of the year.

Great discussion.
Dean

"It's not whether you're right or wrong that's important, but how much money you make when you're right and how much you lose when you're wrong." George Soros