What are your top 5 trading rules???

[tech/a]

I'm not saying buy on a pullback, I'm just saying that buying a high price isn't necessarily a high price. If we buy into what we think is a higher point in the wave because we think it's going to break out or something, then it's not really a high point in the wave (at least by our judgement). If we buy into a higher point of the wave looking for momentum to carry it a little higher, we're probably making a trade that better relates to a faster chart, and on that faster chart we would have been buying into or after the retracement (we may miss these on the slower chart, so the trade isn't really efficient on the slower chart).

Well you've lost me completely I have no idea what your on about.

Your definition of buying low has become a clouded tangle of timeframes.
Cant you post a few charts explaining what your saying or are you just as confused as I am?

Id be really interested in seeing whatever this is traded in realtime.
Not every second but simply Ive entered here because (Chart/s).
Now I'm exiting here because (Chart/s)

 

[wayneL]

Just to throw some fuel on the fire:

http://www.aim-users.com/

(Not my cup of tea, but interesting)

 

[MRC & Co]

1) Know what's happening in the big picture, i.e. global macro environment. Where are funds sloshing their money, what correlations are there and why are they correlated that way, at that point in time.
2) Narrow that down and pick a specific instrument, now you know where it sits in the big picture.
3) Always have a daily view, a 60 minute view and a 5 minute view (if your trading intraday).
4) Know your levels very well, combine them with trend structure and price patterns and you can now define entries/exits and trade with a bias, use volume for tops/bottoms or to confirm a trend (depending on bar range and close).
5) Be contrarian, buy/sell pullbacks or pick tops/bottoms.

 

[tech/a]

Question.

If timing is so important wether conventional "Go with the trend" OR "Contrarian-- pick tops and bottoms" would you expect to out perform.
Lets say the ASX 300

(1) Buying highs to out perform--
(2) Buying Bottoms.

Say buy the highest high for 1,2,3 or whatever years and sell after increase in price of 100,200,300 or whatever %
OR
Say buy the lowest low for X periods1,2,3 or whatever years and and sell after increase in price of 100,200,300 or whatever %

Say over 10 yrs. Or suggest time frames for entry exit and test priods.


Which would you expect to out perform?
Would any out perform the market (Regarding the statement--Entry is important)?

I havent tested any but will if others dont beat me to it!
Happy if you do---save me some work.

 

[Mr J]

Well you've lost me completely I have no idea what your on about.

Your definition of buying low has become a clouded tangle of timeframes.
Cant you post a few charts explaining what your saying or are you just as confused as I am?

Id be really interested in seeing whatever this is traded in realtime.
Not every second but simply Ive entered here because (Chart/s).
Now I'm exiting here because (Chart/s)

Okay take the picture below. On the left is a chart of a slower timeframe, and on the right is a faster timeframe. My aim here is to show that a trade could look completely different depending on the timeframe we look at, and therefore a high price is not necessarily a high price, because it is all relative.

We take an entry at point 1, seeing momentum, and hoping to capture a small profit. We do that at point 2. The faster chart shows what may have happened if we had made the same trade on a faster chart, entering somewhere near 3 and exiting at 4. On the slower chart, it looks like we're buying near the top of a wave - a relatively high price. However, once the trade objective is considered (ride momentum to catch a small profit), I think a faster chart would have been more appropriate to make the trade, and we may have come up with something similar to the left.

Might still be unclear, so maybe I'll explain my perspective. I view a chart as a piece of a greater picture, and containing many pictures of its own. Every chart is part of a larger move not shown on the chart, and every chart contains many smaller moves than may only be apparent on a faster chart. The "one chart's retracement is another's trend" factor. Since I look at charts that way, I always consider what a trade might look like on another chart, and it often completely changes how the trade looks. What most may consider a high price could very well be a cheap price on another chart, so when someone looks to "buy high and sell higher", my argument is that they are probably buying low and selling high, whether they know it or not. If they're buying high on the faster charts, then it's not an optimal trade.

You suggested that I buy into or just after retracements, and that is true. It may not be true on the chart I'm using, but it would be true on a faster chart. I have charts of various timeframes open just for this reason, because regardless of the chart I'm trading, I will also use the information on faster charts to make an entry.

Hopefully someone understandings by ramblings .

 

[tech/a]

I understand.
I have multiple timeframes open for shorter term trading myself.
All your doing is finding a pattern in a timeframe which suits your mindset/criteria. Thats fine but just trade the timeframe in which the patterns/criteria belong.

From my experience.
You can refine a retracement entry on a daily using a 10 min chart which may show a new high being made for 150 periods.(150/10 min periods).
But what your actually trading is the retracement in the daily and that is what should ultimately control the trade.

10/60 and Daily.

Still it doesnt mean these entries OR exits are any more reliable/ profitable than any other.
All they do is supply Beginning/During/ and End points.
What you do with them will determine success or failure.

Again experience has taught me this.

 

[Mr J]

All your doing is finding a pattern in a timeframe which suits your mindset/criteria. Thats fine but just trade the timeframe in which the patterns/criteria belong.

I don't jump around different timeframes, and the ones I use I do trade differently, so the chart on which the trade is made is the one that controls it. A few months I looked at letting the trends of a faster chart control the trade, but it was significantly less profitable. My fast trades do what the fast chart tells me, the medium what the medium tells me etc.

 

[tech/a]

I don't jump around different timeframes, and the ones I use I do trade differently, so the chart on which the trade is made is the one that controls it. A few months I looked at letting the trends of a faster chart control the trade, but it was significantly less profitable. My fast trades do what the fast chart tells me, the medium what the medium tells me etc.

And so then we are back to square one.

That said (Written above) You have no idea wether you have entered a trade at a low or a high for that matter until it can be seen in hindsite.

If you look at all the charts I posted above they can be in ANY timeframe.
You simply DONT know if its a well timed buy/or its is "THE" low.

Timing cannot be claimed until after the fact.

 

[Mr J]

You always seem to reply just before I check!

That said (Written above) You have no idea wether you have entered a trade at a low or a high for that matter until it can be seen in hindsite

Fine, what we perceive to be a high or a low, a good or bad price.

Timing cannot be claimed until after the fact.

In the way that we claimed we timed it correctly or incorrectly? That's arguable, because we can obviously only trade the information we have at the time. If we make a trade in situation that we know over the long run is +ev, then it can be argued that the trade was a good one, and that timing was correct, regardless out of the outcome.

It is how we trade the situation that matters, not whether the result proved us right, because the result can't prove us right. We're playing a game of probabilities, and all that matters is that the probabilities are in our favour. From one perspective, if one has an edge, one can never be wrong (ignoring mistake trades) even if the trade loses.

 

[fapturbo]

Okay take the picture below. On the left is a chart of a slower timeframe, and on the right is a faster timeframe. My aim here is to show that a trade could look completely different depending on the timeframe we look at, and therefore a high price is not necessarily a high price, because it is all relative.

We take an entry at point 1, seeing momentum, and hoping to capture a small profit. We do that at point 2. The faster chart shows what may have happened if we had made the same trade on a faster chart, entering somewhere near 3 and exiting at 4. On the slower chart, it looks like we're buying near the top of a wave - a relatively high price. However, once the trade objective is considered (ride momentum to catch a small profit), I think a faster chart would have been more appropriate to make the trade, and we may have come up with something similar to the left.

Might still be unclear, so maybe I'll explain my perspective. I view a chart as a piece of a greater picture, and containing many pictures of its own. Every chart is part of a larger move not shown on the chart, and every chart contains many smaller moves than may only be apparent on a faster chart. The "one chart's retracement is another's trend" factor. Since I look at charts that way, I always consider what a trade might look like on another chart, and it often completely changes how the trade looks. What most may consider a high price could very well be a cheap price on another chart, so when someone looks to "buy high and sell higher", my argument is that they are probably buying low and selling high, whether they know it or not. If they're buying high on the faster charts, then it's not an optimal trade.

You suggested that I buy into or just after retracements, and that is true. It may not be true on the chart I'm using, but it would be true on a faster chart. I have charts of various timeframes open just for this reason, because regardless of the chart I'm trading, I will also use the information on faster charts to make an entry.

Hopefully someone understandings by ramblings .

Buy Stop Above a Turning Point in the Market Down.

SL Below Previous Turning Point Up in the Market.

Move SL up when each new Turning Point Up occurs... keep doing this until finally get Stopped out.

Buy Higher Sell Higher .....

 

[tech/a]

In the way that we claimed we timed it correctly or incorrectly? That's arguable, because we can obviously only trade the information we have at the time. If we make a trade in situation that we know over the long run is +ev, then it can be argued that the trade was a good one, and that timing was correct, regardless out of the outcome.(Uhhh? thought you said the outcome was +)

If a trade ends up a loss or profitable it can be argued only in hindsite.
At the time of placing any trade or exiting any trade we just don't know how good it really is. Even with stop orders, the stops will do their job but in hindsite how perfect that job was will be clear. If the trade was stopped within tolerance or exited with good R/R then its an "Acceptable" job.

What we do know at the placement of a trade is risk. (Mind you most don't even know that!).

It is how we trade the situation that matters, not whether the result proved us right, because the result can't prove us right. We're playing a game of probabilities, and all that matters is that the probabilities are in our favour.

Which you don't know until you have enough past data to tell you whether your Idea/method has a higher probability of success.
This doesn't necessarily equate to higher win rate either!

From one perspective, if one has an edge, one can never be wrong (ignoring mistake trades) even if the trade loses.

If trading a plan with a blue print from rigorous testing. If you don't have a data set of results you don't actually know.
T/Trader has a massive edge over the market and its correct only 30% of the time.

Mind you if you know how to keep Positive expectancy on your side of the agenda and absolutely control risk than you can trade without a Tested methodology.However I would argue that those who practice this do so through "Method". A very deliberate method of cutting losses and maximising profits.
If you go back to Reply (1) there in lies the truth.
Reply (2) gives one a way to equate "The truth".

Once people prove to themselves that the true edge IS "The Truth" they can move forward and trade most anything in most any time frame.
In my view.

Just elaborate on your "Perspective".

 

[Sunder]

1. Back test your trading system - preferably in blocks of 12 months. If any of the 12 months seem out of sort with any others, perhaps your system only works in particular markets and be aware the market can change quickly. A low volatility stock that almost never trips false alarms the last two years, can suddenly move 5% a day for three months during turbulent times, costing you brokerage and giving you losing trades. If there are some prerequisite that your system relies on (i.e. low/high volatility), then keep an eye out on those prerequisites.

2. Trade small to start off with. (That's when starting trading, not your initial position) Never put all your eggs in one basket, even for a "sure thing". Remember the two sayings "The bargain of the century comes once a week" and Mark Twain's quote that ranks as one of my favourites: "It's not what you don't know that gets you into trouble. It's what you know for sure that just ain't so." Small also helps with discipline. If you can't close out a losing stock when it's lost $1000, what hope do you have of terminating a $10,000 loss when you "know for sure" it's going to turn?

3. Have someone keep you accountable. Explain to them your rules, make sure they understand it. If a trade signal is triggered, you call them, you tell them that you are executing a trade. If you fail to call them, they have to call you. "I notice BHP dropped below its 20 day EMA yesterday. Did you close out your position? No? Why not?". Obviously this has to be a reciprocal favour, or else someone who loves you/owes you. Mine is my fiance, although sometimes she's a bit sloppy with the calling...

4. I'm still learning. Maybe when some of the rules debated now are settled, they'll make position 4.

5. I'm still, still learning. Ditto with position 5.

 

[tech/a]

1. Back test your trading system - preferably in blocks of 12 months. If any of the 12 months seem out of sort with any others, perhaps your system only works in particular markets and be aware the market can change quickly. A low volatility stock that almost never trips false alarms the last two years, can suddenly move 5% a day for three months during turbulent times, costing you brokerage and giving you losing trades. If there are some prerequisite that your system relies on (i.e. low/high volatility), then keep an eye out on those prerequisites.

If you have a trading system that holds for longer periods a set of tests every 12 mths wont reflect a set of results for which the system is designed.
The data period is far to short.

2. Trade small to start off with. (That's when starting trading, not your initial position) Never put all your eggs in one basket, even for a "sure thing". Remember the two sayings "The bargain of the century comes once a week" and Mark Twain's quote that ranks as one of my favourites: "It's not what you don't know that gets you into trouble. It's what you know for sure that just ain't so." Small also helps with discipline. If you can't close out a losing stock when it's lost $1000, what hope do you have of terminating a $10,000 loss when you "know for sure" it's going to turn?

There are times when all eggs in one basket should be a priority!
True re losses. Many get to a point where taking the loss is just to painful.
Don't get there in the first place!

3. Have someone keep you accountable. Explain to them your rules, make sure they understand it. If a trade signal is triggered, you call them, you tell them that you are executing a trade. If you fail to call them, they have to call you. "I notice BHP dropped below its 20 day EMA yesterday. Did you close out your position? No? Why not?". Obviously this has to be a reciprocal favour, or else someone who loves you/owes you. Mine is my fiance, although sometimes she's a bit sloppy with the calling...

You really need to have SELF discipline!

 

[motorway]

Flip this upside down and you get my thoughts

1. Timing is nothing. Trade management is everything.
2. Don't believe in yourself, if you are wrong then forget about it. Next Trade.
3. Cut losing trades off fast.
4. Never average down. Averaging up is OK.
5. I agree here, you won't always get the selection right.

Ah some sence.

For Clarity are you two stating,

That in the system results that tech has posted

Anyone could have gone short at each signal ( instead of long )
and made as much or even more return

and you have data that proves that ?

Timing is nothing. Trade management is everything

That where you start an entry has no effect on results ?
That every entry is as good a short as it is a long ?
Simply manage the trade ?


motorway

 

[tech/a]

For Clarity are you two stating,

That in the system results that tech has posted

Anyone could have gone short at each signal ( instead of long )
and made as much or even more return

and you have data that proves that ?

I have data which shows that Random entry gives a result which isnt a great deal worse than the actual system entry.

That where you start an entry has no effect on results ?
That every entry is as good a short as it is a long ?
Simply manage the trade ?


motorway

Perfect timing is not possible.(Well I havent seen it in anything but rare occasions).
My point is that given a set of conditions for entry say long entry that result wont vary a great deal from one set of long conditions to another.
Using a short entry set of conditions for a long however does.
Infact most entries in profit makingsystems I have tested dont perform any better/much better than random.

Ive not yet tested one which gives a result more than 10% better than random. Ie remove the entry conditon and just set random entries.
TT is a few %.
This indicates to me that its not the entry making the profit!

 

[MRC & Co]

Question.

If timing is so important wether conventional "Go with the trend" OR "Contrarian-- pick tops and bottoms" would you expect to out perform.
Lets say the ASX 300

(1) Buying highs to out perform--
(2) Buying Bottoms.

Say buy the highest high for 1,2,3 or whatever years and sell after increase in price of 100,200,300 or whatever %
OR
Say buy the lowest low for X periods1,2,3 or whatever years and and sell after increase in price of 100,200,300 or whatever %

This is my entire problem with the method tech, it is purely mechanical, I wouldn't care which of those would outperform because I would never try to trade that way. Over 10 years, I don't see any edge in specific timing, but it's not my niche.

Here is my example of which I traded recently and just one of many of what I mean by my rules (then again, these aren't really rules, just tried to give a few basic ideas):

Risk appetite was on a high, sentiment was bullish when S&P was at 930-955s in that range at the top, currently percieved risk assets (all to do with corrlations) were for the most part, sitting on their recent highs. Whilst risk averse assets were on their lows (USTs, USD index). All required a technical pullback. Understand global macro environment. Geithner and the US were desperte to hold up the USD, as any serious break down in it, could cause a run of sorts on both USTs and the USD, effectively destroying their QEP attempts. Meeting with the Chinese to support USD and some spectacularly successful record bond auctions in the US were enough to turn the tide. S&P false broke 950 and this looked firm resistance. Considering oil was on it's highs and now being viewed as a potential to hinder any economic recovery (further push in stocks), whilst bonds were also on their lows, resultant in high yields, particularly a steep curve, passing into the mortgage market hindering rates (believe 30 year mortgage rates are effectively priced off the 10 years), and we had a perfect storm of global macro factors to push the S&P down. We knew the level on the S&P of 950 was near range top and a perfect area to get short in anticipation, considering the trend structure was also running out of momentum. That is timing to me.

We got down to 870s looked as though we would push below the range (probably to the 50% retracement on S&P at 810ish) and on a Monday, Asia positioned short for this event. Overnight, Europe and US put in a huge push higher. Coming into earnings season with the first being what we know as a company that ALWAYS beats the street in GS, we had some bullish appetite. The S&P came back up to 900s and you could clearly see Asia panic out on Tuesday in the order flow. Guys dumping at market and HUGE size coming online and showing their hand, they wanted to be clipped and wanted out. Nobody there to clip them? Why not? Everybody who was anybody or otherwise referred to as 'they' were now pricing for a rally. Tuesday, get long. Of course, this could have just ended up in chop, but I think it was clear by the action that there was a decent probability of a rally. Just prior to this, the S&P had false broke an important level just above the 870s, on no effort and tiny range bars, with a move back through 900s and the rest of the factors probably a good reason to time an entry. Not buying the exact low in this case, but still timing nonetheless.

Combination of knowing the overall environment, knowing where everything sits in the picture, knowing what the current psychology is behind the moves, knowing what it would take to shift that sentiment and understanding levels and how markets move to narrow it down and time specific entries.

Of course, intraday has less fundamentals and more simply flow, lower probability IMO, but more opportunities.

 

[Garpal Gumnut]

OK listen up.

Follow these rules and you will always make a profit.

1. Buy a monkey,a dart and 2 copies of the AFR.

2. Pin the Stock prices pages from the copies of the AFR to a cork board so that all stocks are visible.

3. Train the monkey to throw the dart at the pages. ( This is the time intensive part.)

4. Go long on any stock the monkey hits directly with the dart, if it falls between two, go short on both in a bear market, and long on both in a bull.

5. Ensure the monkey is adequately pissed, the first monkey I had preferred Jamesons, my most recent one will only drink Sazerac Rye, its more expensive, but he's performing better than the first.

gg

 

[Timmy]

This is my entire problem with the method tech, it is purely mechanical, I wouldn't care which of those would outperform because I would never try to trade that way. Over 10 years, I don't see any edge in specific timing, but it's not my niche.

Here is my example of which I traded recently and just one of many of what I mean by my rules (then again, these aren't really rules, just tried to give a few basic ideas):

Risk appetite was on a high, sentiment was bullish when S&P was at 930-955s in that range at the top, currently percieved risk assets (all to do with corrlations) were for the most part, sitting on their recent highs. Whilst risk averse assets were on their lows (USTs, USD index). All required a technical pullback. Understand global macro environment. Geithner and the US were desperte to hold up the USD, as any serious break down in it, could cause a run of sorts on both USTs and the USD, effectively destroying their QEP attempts. Meeting with the Chinese to support USD and some spectacularly successful record bond auctions in the US were enough to turn the tide. S&P false broke 950 and this looked firm resistance. Considering oil was on it's highs and now being viewed as a potential to hinder any economic recovery (further push in stocks), whilst bonds were also on their lows, resultant in high yields, particularly a steep curve, passing into the mortgage market hindering rates (believe 30 year mortgage rates are effectively priced off the 10 years), and we had a perfect storm of global macro factors to push the S&P down. We knew the level on the S&P of 950 was near range top and a perfect area to get short in anticipation, considering the trend structure was also running out of momentum. That is timing to me.

We got down to 870s looked as though we would push below the range (probably to the 50% retracement on S&P at 810ish) and on a Monday, Asia positioned short for this event. Overnight, Europe and US put in a huge push higher. Coming into earnings season with the first being what we know as a company that ALWAYS beats the street in GS, we had some bullish appetite. The S&P came back up to 900s and you could clearly see Asia panic out on Tuesday in the order flow. Guys dumping at market and HUGE size coming online and showing their hand, they wanted to be clipped and wanted out. Nobody there to clip them? Why not? Everybody who was anybody or otherwise referred to as 'they' were now pricing for a rally. Tuesday, get long. Of course, this could have just ended up in chop, but I think it was clear by the action that there was a decent probability of a rally. Just prior to this, the S&P had false broke an important level just above the 870s, on no effort and tiny range bars, with a move back through 900s and the rest of the factors probably a good reason to time an entry. Not buying the exact low in this case, but still timing nonetheless.

Combination of knowing the overall environment, knowing where everything sits in the picture, knowing what the current psychology is behind the moves, knowing what it would take to shift that sentiment and understanding levels and how markets move to narrow it down and time specific entries.

Of course, intraday has less fundamentals and more simply flow, lower probability IMO, but more opportunities.

Wow - great stuff MRC, thank-you.

 

[Cartman]

Wow - great stuff MRC, thank-you.

:iagree:

could i read between the lines Mirc and see u becoming a "high volume" swing trader in the future?? onya m8

 

[beamstas]

Anyone could have gone short at each signal ( instead of long )
and made as much or even more return

and you have data that proves that ?

That where you start an entry has no effect on results ?
That every entry is as good a short as it is a long ?
Simply manage the trade ?


motorway

Firstly, no.
Going short at every signal would not produce the same results, why? Because going short is a whole different game to going long. Im not saying that it doesn't matter if you go short or long, im saying the entry doesn't matter.

I do have the data to back it up and can provide it if you want to question it. I also have a trading system that can produce 20% CAR even through the 2008-09 crash, using random entries.

You can analyse the market as much as you want, but at the end of the day you still have no idea what the market is going to do tomorrow. You can never be 100% certain.

Motorway, i suggest you read Scott Barlows interview in Radges book. You''ll be amazed, this bloke can trade at a 10% win rate and still be ahead at the end of the year. How does he do it? He manages the trades. He doesn't even have an entry criterea, he subscribes to get entries, and wins 30% of the time. He is an extreme example of trade management, but a good one. He doesn't care less about the entry, it's the management and exit that makes the money.

I did a little experiment using Howard Bandys P<0.05 rule of shutting down a system, from another thread on here. 50% of the time for the last 250 days the market closes higher, 50% it closes down, approximately, for each stock. P<0.05 at 5 days in a row, up or down, at 50/50, so if you have a system that goes long after 4 down days, theoretically you should have a 95% win rate. In reality you have a 30% win rate.

The market isn't predictable enough to have an entry that gives you a great edge, if it was we would all be millionaires.

I know i will be proven wrong by someone showing stats of a backtested system that has a MA Crossover entry or something ridiculous, but entries are nearly always curve fitted. People analyse charts and look for characteristics that are in trades they don't want, and make a provision to filter out the stocks that do that.

Let the winners run and cut the losers short, simple.

I give you this challenge. If you trade mechanical, remove your entry and backtest. You will find that the profit will increase/decrease very slightly. It's not the entry giving you those results, it's the management of the trade and the exit.

Go and code something up in amibroker with random entries, and you'll actually be suprised at how well the system performs.

You won't make money because X moving average crossed above Y and the volume was XYZ and the RSI was under 20, the market doesn't care about all that.

Here is another example: I coded a system for someone the other day, and a requirement was that the RSI had to be under 30 in the last 10 days. Now common sence would say this is a good thing, the stock is oversold and should come up, the results were poor.... I then removed the RSI filter and backtested, the system produced more than 3x the previous CAR.

Does it make sence that buying stocks that are even overbought gives you more profit than buying stocks that ARE oversold? No, not really, it doesn't make sence, but then again, the market deosn't care where the RSI is.

Brad