| Author |
Message |
     mum
Member
Username: mum
Post Number: 115
Registered: 08-2005Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 02:06 pm: |  
|
I would be interested in peoples % of picks that go up and make money. The Turtles evidently had 45%. I asked around other traders and they said about 45% and looking back at my motley selection , 45%. Makes you wonder if technical analysis does work. Less than half of our picks work, that doesn't seem to me a great result.
Does chart reading work ? I was told the market is random as a defence so that would make TA impossible hey ?
I would like to get my ratio up to 60% or is this impossible. !!!
|
ken
Member
Username: ken
Post Number: 412
Registered: 04-2003Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 02:21 pm: |
|
Hello Mum,
The longer your timeframe, the lower will be your % of wins. The % of wins also depends on your exits.
Through automated backtesting I have some one-day systems with 73% wins, 7 day average systems with 58-61%, but when I put long term exits on them they can change to as low as 35%.
Don't forget that there are 3 possibilities for each trade, up, down and sideways. 45% up is not so bad under this view.
Ken
Trade with the trend, not against it.
The trend is the direction of the 22ema line (Elder)
|
david_louisson
Member
Username: david_louisson
Post Number: 253
Registered: 02-2004Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 02:31 pm: |
|
Hi Mum
It's possible to come out ahead with a 45% win rate if one's average win size is sufficiently high, compared to one's average loss size.
For example, if I make 4 wins averaging $400 each, and 6 losses of $200 each, then my net result is
( 4 x 400 – 6 x 200 ) = +$400, even though my win rate is only 4 trades in 10, or 40%.
The average loss size is determined by the distance of the protective stop from the entry. If the stock price trends steeply upward, then the amount of the win could potentially be several times this value. That's how the turtles, and others, have made their millions.
If one's analysis is purely technical (price & volume only), and price movements are random, as Burton Malkiel states in his best seller 'A Random Walk down Wall Street', then it is mathematically impossible to profit, and any gains are the result of lucky guesswork. My own trading experience tends to confirm this view.
If the market is "trending", then there is a bias to price movement, i.e. it is not random. Under such circumstances, one can potentially profit by trading in the direction of the trend, cutting losses short, and letting profits run as long as possible (creating the situation described above, where the average win > average loss).
If trends persist long enough for you, on average, to overcome costs, then you will profit, otherwise you will lose. Trading is both as simple, and as difficult, as that.
There are contributors to this forum whose accounts apparently show consistent annual profit. Some of these have been kind enough to freely share their winning methods on this forum.
Best wishes
David
|
mum
Member
Username: mum
Post Number: 116
Registered: 08-2005Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 02:46 pm: |
|
Ken, thanks for that. Yes I see what you mean, if you take out breakevens its sounding better.
Thats what I hate about weekly charting systems. You usually have to first go through a period of drawdowns .Then if the whole market corrects your way down. I think I feel happier using daily chart system.
so for 7 day period or more it is possible to get the ratio up-great
|
david_louisson
Member
Username: david_louisson
Post Number: 254
Registered: 02-2004Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 02:46 pm: |
|
Some more thoughts:
Setting a tighter stoploss will decrease your win rate, as you will be stopped out more frequently, but your average loss size will be smaller.
Conversely, you could probably achieve a 90% win rate by setting incredibly loose initial stops, and taking profits very quickly. However, it would only take one (relatively) big loss to erase all of the small profits.
There is no easy way of increasing both potential win size, and potential win rate, simultaneously, as one is inversely dependent on the other. How easy life would be if one could have one's cake and eat it too!
David
|
mum
Member
Username: mum
Post Number: 117
Registered: 08-2005Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 03:31 pm: |
|
yes but setting incredibly loose stop losses would mean you have less money in the trade so the profit would be small. Tight stop losses i think would work better as the trade is more likely to keep on going up especially if you bought in on high volume breakout. I went to a talk by L Bedford the other day and she likes to enter market at end of day so she can see if late selling comes in. I suppose that would protect you from a fake outs to an extent
|
djb
Member
Username: djb
Post Number: 23
Registered: 07-2003Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 03:48 pm: |
|
I have developed a couple of long term mechanical weekly chart systems that average over 50% win rate.Both systems use different entry and exit methods and yet have similar results.Extensive backtesting and Monte Carlo testing give win rates of 50-60% with win/loss ratios of about 8 to 1.This is over ten years of ASX data.
One system looks for volatility breakouts and the other looks for strongly trending stocks.The win/loss ratio is also quite dependent on the postition sizing that I use.In fact small adjustments to position sizing tends to have bigger effects than small adjustments to system parameters.I find it easier to build a workable weekly system than a daily system, but maybe thats me.
As far as your loss being less with a tighter stop, this does'nt make sense to me.Your capital at risk should be invariant to your stoploss placement.With fixed fractional position sizing, you risk x% of capital regardless of where your stop is.
Dave
|
mum
Member
Username: mum
Post Number: 118
Registered: 08-2005Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 04:01 pm: |
|
I don't mean that your loss would be less with a tighter stop but that the trade would be in profit sooner(more money in trade) and you could get out at break even more often, hope that makes sense , I'm just surmising here. I suppose you would have to trade more often
|
ohkoolnutz
Member
Username: ohkoolnutz
Post Number: 278
Registered: 10-2005Rating: N/A
Votes: 0
|
| | Friday, September 01, 2006 - 08:23 pm: |
|
I have an alternate way of looking at this.
The usual person judges a winner to be where the buy price is lower than the sell price. The opposite is true for a loser. A break-even scenario may be put into either category.
Focus is on the entry price and exit price with no regard to the price movement ('up and down' or 'down and up'). Our inability to control the price means we should focus on capital at risk and where and when we got stopped out. The only thing we can control is our maximum loss at any point in time. This holds true for our initial stop loss and then any possible trailing sell stop. The trailing sell stop may be below or above the stop loss and takes precedence over the stop loss only when time allowed it to move above the stop loss.
It therefore makes sense to change the way we count winners and losers. I have done this change and it works wonders for my mental state. Before I buy a stock I calculate the stop loss. I try to choose stocks where the maximum drawdown is 1.5%. That is 1.5% risk on the position and not the entire portfolio! The portfolio risk is 0.1% on such trade!
I then buy stop limit the stock and put the stock into a watchlist where I put the stop loss (!!!) as my entry price. Choosing stocks with maximum risk of 1.5% allows me to mentally write off this dollar value out of my capital and feel at ease.
This achieves that every stock I buy immediately is in the black. I use parabolic SAR as trailing sell stop. Using this method a loser is a stock that is stopped out at the stop loss or below. A winner is a stock that is stopped out above the stop loss but not necessarily above my buy price.
I can give an example. In this case the stop loss and trailing sell stop are equal from the get go which is the sweetest setup you could want. Google has been going sideways for months. I wanted to buy with maximum protection. I watched and waited for weeks.
August 23: GOOG moves down to SAR. The price action is between 378.27 and 372.66. The risk on entry after this day is 1.48%. It's barely under 1.5% and a bit of slippage can ruin this play. I decide to wait.
August 24: GOOG touches SAR. The day is tighter with 376.40 and 372.26. Risk is now 1.1%. I put in my buy stop limit at 376.76.
August 25: GOOG tightens again with 375.32 and 372.50. I calculate the risk using 375.32 and 372.26 (prior day's low) and it is 0.8%. I move my buy stop limit to 375.66
August 26: I get filled and go into the black immediately. Money at risk on this position is 0.8% because my stop loss is at that level and the trailing sell stop (SAR) is at that level as well. Portfolio at risk is 0.05%.
Practically nothing has changed. I am more certain how well protected my capital is. My watchlist tells me I have 15 positions. ALL of them are in the black. The maximum drawdown if I got stopped out on all positions is currently 6%. The value increases as your stocks move higher. If it reaches 10% it may be a sign of vulnerability and some stops should be tightened to protect the bank.
I used to be nervous when a stock fluctuated around my buy price. I now feel relaxed while a stock fluctuates above my stop loss. When a stock goes into the red (in my watchlist my stop loss/trailing sell stop is my entry price) I check that it got stopped out and then remove it from my watch list. My watchlist only contains stocks that are in the black which helps improve my outlook on life.
---
ohk
Lies, Damn Lies and Technical Analysis
|
david_louisson
Member
Username: david_louisson
Post Number: 255
Registered: 02-2004Rating: N/A
Votes: 0
|
| | Saturday, September 02, 2006 - 07:14 am: |
|
Mum – yes, I agree that a loose stop means that (assuming one uses the fixed fractional sizing method) your position size will be smaller. Apologies for any misunderstanding, but I was trying to answer your original post, i.e. ways to increase win rate to above 45% (ignoring everything else).
DJB – yes, agreed that, according to the fixed fractional method, by having stops twice as tight, you end up doubling the position size, to keep the loss to total capital the same. But doubling position size also doubles potential return. Hence, at least theoretically, a tighter stop increases the return to risk ratio, and therefore also the average win size relative to the average loss. The downside being that the tighter the stop, the greater the probability of being stopped out.
OHK – not sure that I understand. You say "A winner is a stock that is stopped out above the stop loss but not necessarily above my buy price." Does this mean that you count some trades that have stopped out below your buy price as winners, even though their overall effect on your bottom line is negative?
David
|
ohkoolnutz
Member
Username: ohkoolnutz
Post Number: 279
Registered: 10-2005Rating: N/A
Votes: 0
|
| | Saturday, September 02, 2006 - 09:12 am: |
|
Yes, you understood right. I can't think of a chart which illustrates this right now. This can only occur when the buy price is higher than the stop loss which is higher than the parabolic SAR. As time passes the price moves up coercing the SAR to pass the stop loss but remain under your buy price. The price then moves down and stops you out at SAR. This would be the case in a trading range or if the gap between buy price and SAR was too big and there was not enough time for SAR to climb.
Remember that we do this to solve the mental challenge that we fight every day. We have no control over price before we enter a position and after we entered a position. We only have control over our capital as we execute buy and sell decisions. It therefore makes sense to calculate your stop loss before each trade, write off the difference and judge your predetermined stop loss as the decider between winner and loser.
When you buy a lotto ticket do you consider the cover to be still in your possession? When you place a bet at poker do you consider what you contributed to the pot still to be in your possession?
Writing off the stop loss forces you to find the trades where the risk to your capital is minimal. The price you buy at is unimportant. The price you sell at is unimportant. The risk you expose yourself to is the only thing that counts.
Winner:
Loser:
---
ohk
Lies, Damn Lies and Technical Analysis
|
smallworld
Member
Username: smallworld
Post Number: 487
Registered: 01-2004Rating: N/A
Votes: 0
|
| | Saturday, September 02, 2006 - 12:19 pm: |
|
Mum,
I recently backtested the turtle system for the ASX300 between Jan 2000 and Aug 2006 and wrote an article. The system is as described by Curtis Faith, one of the original turtles.
The win rate is just 35%, but the interesting result is that the system is surprisingly profitable, returning 700% over the test period. By comparison, XAO returns just over 60%. It is even positive over the period 2000 to 2003 when XAO was under water by 12%.
Quite often people have mistaken technical analysis for tried and tested systems. The turtles have shown us that having such a system, though simple as it is, and the discipline and courage to trade it accordingly, is very profitable indeed.
For those who are interested in my article, you can PM me.
|
lafee
Member
Username: lafee
Post Number: 239
Registered: 04-2003
Rating: N/A
Votes: 0
|
| | Saturday, September 02, 2006 - 02:52 pm: | | | |
