Rules for Drawing Trendlines
- Thread starter dklepak
- Start date
L
lovethetrade
Rules for Drawing Trendlines
Rule No.1 : Don't draw them
Rule No.1 : Don't draw them
L
lovethetrade
Because they hinder you from finding out how the game is really played.
Some thoughts:
"Trend" cannot be defined with any sort of certainty -- any "trend" you see is someone's invention.
"Trend lines" -- similarly, cannot be defined, a priori.
What this means in practice is that trends are drawn (defined; depicted) post hoc. Always.
What *that* means in practice, is that changes in that trend are unpredictable from the trend itself.
What can you do about the intrinsic post hoc nature of trends?
Embrace it:
Draw the lines of the trends you see, in the buying and selling of the market, as you see those trends occur.
End your trend lines at the newest data. See, as time passes, how well your trend 'definition' fits with what follows.
Imagine a trading pit (or local farmers' market) in front of you, and listen for the voices of those calling prices and volumes: put BUY orders and SELL orders into words and phrases and larger agendas and morning memoranda and midday briefs and end-of-day alerts and event warnings and news alerts and holiday/end-of-month/end-of-quarter/tax year inventory adjustment/reallocation requirements. ("Phew!") Do you hear the clamor? The cacophony of opposing market moves? The give-and-take that makes up a moving market?
Now, go back and look at your graph:
Do you see the weekly trend? The turnover by sector?
Do you see the humps of every 15 or 20 or 50 minutes in the four-hour?
Do you see how they died at the Euroclose? Or how they started with the US-open?
(And they rode it up or down to a consistent amount, then back off, and did it a 2nd and 3rd and 4th time, until lunch? And they started up 90 minutes later, and then jacked it with 20minutes to close?) etc etc etc...
Think in terms of an Agenda:
Don't think in terms of trend lines -- instead practice reading the chart in front of you -- explaining what's going on. Price and volume, volume and price -- it's all the same, whether it's tomatoes in the Saturday open-outcry market down the street, or global electronic securities. Volume and price, price and volume. So look at the graph, draw some lines, and explain them to yourself. Don't try to hit every high spot or low spot. With a bit of practice, you'll find that there are more robust trends that, when you ignore spikes above a capping ("resistance") trendline, or pokes below a "support" trendline, you will likely see a concomitant spike in volume as well -- and you imagine what was going on -- some particular party had a market agenda working ("Bob, have your team cut our IBM this morning -- all 2M shares. Work it within -1%, or to 11:30amET, but lose it all.") And then someone else comes in with their own selling agenda -- maybe more, maybe less -- and you can see the battle is on, fighting to get out of their IBM positions at the quickest pace, or the highest price. Price and volume; volume and price.
When "Agenda-writing" becomes Second Nature...
Then you'll realize, too, the futility of drawing trend lines without also knowing that anybody can come to the market, in the next second, with a big ol' agenda at odds with, or in scary concurrence with, the existing agenda(s), and change the entire picture.
You'll realize that strong agenda days (when things look amazingly blatant) are gifts to smaller traders (who mostly provide liquidity to help the market on its way), and that weak agenda days (when "size" is not there: lunch; 3-day weekend; morning crisis+afternoon doldrum; FOMC days) are a *great* time to back away from the keyboard: Price flops around, and volume is nil, and anyone with an agenda impacts the market scarily...
So,
1) forget about any sort of magical trendlines -- it's post hoc addressing of symptoms, and not-at-all addressing of the underlying cause(s).
2) focus on explaining what is in front of you, in terms of tapping into the trading agenda(s) of those traders of size.
3) recognize that trading agendas -- like rocks dropped in a pond -- have additive and subtractive effects on the existing market, and that
4) a new trading agenda may be dropped in the pond at any moment, changing everything up once again. So,
5) Your job is to read the old agendas, and listen for the gaps or holes or sensitivities in the current market, and engage or disengage as experience suggests.
"Trend" cannot be defined with any sort of certainty -- any "trend" you see is someone's invention.
"Trend lines" -- similarly, cannot be defined, a priori.
What this means in practice is that trends are drawn (defined; depicted) post hoc. Always.
What *that* means in practice, is that changes in that trend are unpredictable from the trend itself.
What can you do about the intrinsic post hoc nature of trends?
Embrace it:
Draw the lines of the trends you see, in the buying and selling of the market, as you see those trends occur.
End your trend lines at the newest data. See, as time passes, how well your trend 'definition' fits with what follows.
Imagine a trading pit (or local farmers' market) in front of you, and listen for the voices of those calling prices and volumes: put BUY orders and SELL orders into words and phrases and larger agendas and morning memoranda and midday briefs and end-of-day alerts and event warnings and news alerts and holiday/end-of-month/end-of-quarter/tax year inventory adjustment/reallocation requirements. ("Phew!") Do you hear the clamor? The cacophony of opposing market moves? The give-and-take that makes up a moving market?
Now, go back and look at your graph:
Do you see the weekly trend? The turnover by sector?
Do you see the humps of every 15 or 20 or 50 minutes in the four-hour?
Do you see how they died at the Euroclose? Or how they started with the US-open?
(And they rode it up or down to a consistent amount, then back off, and did it a 2nd and 3rd and 4th time, until lunch? And they started up 90 minutes later, and then jacked it with 20minutes to close?) etc etc etc...
Think in terms of an Agenda:
Don't think in terms of trend lines -- instead practice reading the chart in front of you -- explaining what's going on. Price and volume, volume and price -- it's all the same, whether it's tomatoes in the Saturday open-outcry market down the street, or global electronic securities. Volume and price, price and volume. So look at the graph, draw some lines, and explain them to yourself. Don't try to hit every high spot or low spot. With a bit of practice, you'll find that there are more robust trends that, when you ignore spikes above a capping ("resistance") trendline, or pokes below a "support" trendline, you will likely see a concomitant spike in volume as well -- and you imagine what was going on -- some particular party had a market agenda working ("Bob, have your team cut our IBM this morning -- all 2M shares. Work it within -1%, or to 11:30amET, but lose it all.") And then someone else comes in with their own selling agenda -- maybe more, maybe less -- and you can see the battle is on, fighting to get out of their IBM positions at the quickest pace, or the highest price. Price and volume; volume and price.
When "Agenda-writing" becomes Second Nature...
Then you'll realize, too, the futility of drawing trend lines without also knowing that anybody can come to the market, in the next second, with a big ol' agenda at odds with, or in scary concurrence with, the existing agenda(s), and change the entire picture.
You'll realize that strong agenda days (when things look amazingly blatant) are gifts to smaller traders (who mostly provide liquidity to help the market on its way), and that weak agenda days (when "size" is not there: lunch; 3-day weekend; morning crisis+afternoon doldrum; FOMC days) are a *great* time to back away from the keyboard: Price flops around, and volume is nil, and anyone with an agenda impacts the market scarily...
So,
1) forget about any sort of magical trendlines -- it's post hoc addressing of symptoms, and not-at-all addressing of the underlying cause(s).
2) focus on explaining what is in front of you, in terms of tapping into the trading agenda(s) of those traders of size.
3) recognize that trading agendas -- like rocks dropped in a pond -- have additive and subtractive effects on the existing market, and that
4) a new trading agenda may be dropped in the pond at any moment, changing everything up once again. So,
5) Your job is to read the old agendas, and listen for the gaps or holes or sensitivities in the current market, and engage or disengage as experience suggests.
Last edited:
I started with trendlines in 1978, no home computers, it was graph paper, pencil, ruler and copy of Wall Street for stock data.
The reason you are getting so many responses to how to draw trend lines is simple, many don't use them or know how because a trendline to me is considered into itself a trading system and once you know how to program, you can test your ideas. Some like myself have rigid rules as I can easier program them and others much less so and they will draw through bars which I never do. It comes down to much back testing for you to gain your own experiences. We all have different levels of experience here at forum, different account sizes, retail to professional, individual to Hedge funds, the game is played by many.
Let's start with strength, strength is determined by bars, if you are seeking a strength of 25, this means a Pivot low, to the left of it has 25 or more higher lows, very strong in my opinion of any timeframe, price has not been lower in 25 bars of the past. Then we have to consider "slope", this can be tricky cause as you cram more bars onto a chart, the slope of the trendline changes, so it is best to maintain some type of standard that works for you as far as number of bars on the screen, then you start testing out different slopes, some charting software will give slope. The reason of knowing slope, some slopes don't offer respectable percentages as far as losing, too great. sharp slopes I never take. Once you determine what degrees of trendlines you are happy, then comes "C" or 2nd pivot you will be drawing to, again you can develop a strength of these, how many to the left this pivot low has not been violated, mine are always shorter than my anchor, so let's say strength of 5, means "C" not been broken on at least 5 bars and angle of trendline has limited slope to not exceed what you tested to be favorable slope, then you can have how many bars to right to define "C" so next time price touches either market or limit order. Backtesting to see how much for stops etc...
There is nothing perfect and everything is a bet, nothing foretells the future, all you can do is enough back testing to give yourself ideas of what can happen and ALWAYS know drawdowns will exceed whatever you backtested and goals can as well.
Good luck.
The reason you are getting so many responses to how to draw trend lines is simple, many don't use them or know how because a trendline to me is considered into itself a trading system and once you know how to program, you can test your ideas. Some like myself have rigid rules as I can easier program them and others much less so and they will draw through bars which I never do. It comes down to much back testing for you to gain your own experiences. We all have different levels of experience here at forum, different account sizes, retail to professional, individual to Hedge funds, the game is played by many.
Let's start with strength, strength is determined by bars, if you are seeking a strength of 25, this means a Pivot low, to the left of it has 25 or more higher lows, very strong in my opinion of any timeframe, price has not been lower in 25 bars of the past. Then we have to consider "slope", this can be tricky cause as you cram more bars onto a chart, the slope of the trendline changes, so it is best to maintain some type of standard that works for you as far as number of bars on the screen, then you start testing out different slopes, some charting software will give slope. The reason of knowing slope, some slopes don't offer respectable percentages as far as losing, too great. sharp slopes I never take. Once you determine what degrees of trendlines you are happy, then comes "C" or 2nd pivot you will be drawing to, again you can develop a strength of these, how many to the left this pivot low has not been violated, mine are always shorter than my anchor, so let's say strength of 5, means "C" not been broken on at least 5 bars and angle of trendline has limited slope to not exceed what you tested to be favorable slope, then you can have how many bars to right to define "C" so next time price touches either market or limit order. Backtesting to see how much for stops etc...
There is nothing perfect and everything is a bet, nothing foretells the future, all you can do is enough back testing to give yourself ideas of what can happen and ALWAYS know drawdowns will exceed whatever you backtested and goals can as well.
Good luck.
Buy that shrimp future now as shrimp prices could be on the rise in the USA if this clip is accurate: