Come into the Trader's Den

[trader.trends]

Hi TT,

As a thumb rule you can expect to reverse after 5 days.

Hi Columbus

Your comment set me on a new path of enquiry. One was why do we have thumb rules which are no more than 'hearsay', 'intuition', 'tradition' etc when a rule can easily be verified. Data is not in dearth. And the NF trading began only in June 2000 which means we have to verify only about ten years data hence the sample world itself is small. So here is my finding on the 5 day thumb rule:

In all its history NF has had 5 and more trending on 78 occasions. Would we have made money if at the end of the fifth trending day we had bet that the market would reverse the next day? In practical terms it would mean that we take a short position at EOD if the mkt has gone up for 5 consecutive days and long position if it had declined for 5 consecutive days and hold it till EOD of sixth day.

You would have made money on 43 occasions and lost 35 times as the mkt continued its trend for more than five days. Percentage wise you would have made it 55% and lost it 45%. Not really a good thing to bet on. I have not gone into how much you would have made and lost just the number of times you would have made or lost.

The enquiry threw up some more things:
6 day trend has happened 14 times, 7 day trend 12 times, 8 day trend 6 times, 9 day trend 1 time, 10 day trend 1 time and 12 day trend 2 times!!

Less than 5 day trend was again no better than chance:

4 or more days trend occured 156 times where it terminated at 4 days 78 times and continued beyond 4 days 78 times. 50:50
3 or more days trend occured 318 times where it terminated at 3 days 162 times and continued beyond 3 days 156 times. Again 50:50.
2 or more days trend happened 609 times where it terminated at 2 days 291 times and continued 318 times. 47:53

Each and every unbiased enquiry into the market movement leads us inexorably to the single conclusion: market movements are random. market movements are random and market movements are random.

I have mentioned that there were two occasions with a 12 day trend. I don't know how many people can understand this: Such trends only prove the random nature of the market. If the sample were sufficiently bigger, you can find a 20 day trend too and still the mkt movement would be random.

Most people lose money because they do not understand the randomness. When a million people play the markets it is inevitable that a few would succeed with or without a method or a system. Randomness would require that. To the nature of the mkt when we add our own nature which sees what it wants to see, which is indisciplined, which has emotions like greed fear etc, we set ourselves up for disaster most times.

 

[SwingKing]

Hi Columbus

Your comment set me on a new path of enquiry. One was why do we have thumb rules which are no more than 'hearsay', 'intuition', 'tradition' etc when a rule can easily be verified. Data is not in dearth. And the NF trading began only in June 2000 which means we have to verify only about ten years data hence the sample world itself is small. So here is my finding on the 5 day thumb rule:

In all its history NF has had 5 and more trending on 78 occasions. Would we have made money if at the end of the fifth trending day we had bet that the market would reverse the next day? In practical terms it would mean that we take a short position at EOD if the mkt has gone up for 5 consecutive days and long position if it had declined for 5 consecutive days and hold it till EOD of sixth day.

You would have made money on 43 occasions and lost 35 times as the mkt continued its trend for more than five days. Percentage wise you would have made it 55% and lost it 45%. Not really a good thing to bet on. I have not gone into how much you would have made and lost just the number of times you would have made or lost.

The enquiry threw up some more things:
6 day trend has happened 14 times, 7 day trend 12 times, 8 day trend 6 times, 9 day trend 1 time, 10 day trend 1 time and 12 day trend 2 times!!

Less than 5 day trend was again no better than chance:

4 or more days trend occured 156 times where it terminated at 4 days 78 times and continued beyond 4 days 78 times. 50:50
3 or more days trend occured 318 times where it terminated at 3 days 162 times and continued beyond 3 days 156 times. Again 50:50.
2 or more days trend happened 609 times where it terminated at 2 days 291 times and continued 318 times. 47:53

Each and every unbiased enquiry into the market movement leads us inexorably to the single conclusion: market movements are random. market movements are random and market movements are random.

I have mentioned that there were two occasions with a 12 day trend. I don't know how many people can understand this: Such trends only prove the random nature of the market. If the sample were sufficiently bigger, you can find a 20 day trend too and still the mkt movement would be random.

Most people lose money because they do not understand the randomness. When a million people play the markets it is inevitable that a few would succeed with or without a method or a system. Randomness would require that. To the nature of the mkt when we add our own nature which sees what it wants to see, which is indisciplined, which has emotions like greed fear etc, we set ourselves up for disaster most times.

TT,

Very well researched indeed.

I agree with you that markets are random (more than 60-70% of time), but markets are non - random rest of the times. It is this window where most of the traders make money. Atleast this applies to me. Else if markets were purely random, system designs with accuracy of 75 - 80% spanning data of 30 years over 1500 + trades would not be possible. If such systems are possible, then this signifies that during periods when those systems work, markets remain non random.

I think more than randomness or non randomness of markets, profit potential is attributable to human psychology. I just cannot highlight how Important this little aspect is. A lot has been said and written about psychology, but I feel there is no book/video/course which can teach one to think correctly. This is something one can only learn after spending hours watching the market and of course after losing significant amount of money. If you research about some of the greatest traders, all have two things in common; Hours they spend studying price action and amount of money they lost before they could make some.

This essentially tells us that randomness exists more in human behavior than in markets and this is the main reason why 'n' number of people lose money in the markets. Any system or for that matter any methodology (MA system, breakout system, volatility based system) can work if one is tuned to think correctly. For me, in the end, the difference between a profitable trader and the rest is of underlying psychology. I can say this, because I believe that I can make any well designed system work, just by thinking correctly. This should not be perceived as my arrogance towards markets, but it is to highlight what exactly is required to win in markets. People who are essentially winning have this aspect of their trading plan perfectly tuned with the movements of prices.

Markets are random, but human rationale can be customized to be non random. That is where my profit comes from.

Tc

 

[DanPickUp]

Hi AW10

Welcome back and I hope you had nice holidays

We already missed you.

Dan

 

[AW10]

where is AW bhai now a days , he even not logged in after 24-july.

best regards

annu

Just returned from my short holidays of 17 days.

Waise bhee August is one of the slowest month of the market. I just read in another place that 8 out of last 10 yrs, the monthly range of August is less then the average monthly range of the market for that year. many senior traders at hedgefunds to on holidays in August leaving the trading desk in charge of junior traders to maintain status quo..

and if u look at the movement of global mkt, including ours, then it is very much in line with the observation..

so it is one of the best month to take time off and get ready for high action months of sept/oct ..

happy trading

 

[AW10]

Marc Faber, as you rightly said, always talk about bearish markets. He is thirsty for blood bath. Mostly when one watches his interviews, it develops in us a word of pessimistic feeling about the market, than the cautious approach.

marc faber is one of the guy whom i will listened to, anyday.
(note - it is not marc faber but jim rogers.. hence next few lines are for Jim)

he is bullish on commodities, on china, and about 2/3 months back,he is also bullish on Japan equity mkt. He is also bullish on agriculture..

He is just not a trader but also teaches economics in some university.. and he trades long term fundamentals.. He sold his house in US to cut exposure to USD and moved to singapore currently where he is grooming his 4/5 years old daughters in Mandarin language to face the future..

happy Trading

 

[SwingKing]

Be it Marc Faber or Warren Buffet, I'll listen to what I say. I am no Marc Faber or Warren Buffet, but someday, even if I become one, I want to be with my own opinion and my own vision. Who knows, I may well become one of them. To realize that "Guru Status", one needs to back his own views and literally shun other's opinion.

No offence to Mr. Faber or Mr. Buffet.

Tc

 

[SwingKing]

Welcome back AW10. Hope you had a Rocknrolla holiday.

 

[AW10]

Hi Columbus

Your comment set me on a new path of enquiry. One was why do we have thumb rules which are no more than 'hearsay', 'intuition', 'tradition' etc when a rule can easily be verified. Data is not in dearth. And the NF trading began only in June 2000 which means we have to verify only about ten years data hence the sample world itself is small. So here is my finding on the 5 day thumb rule:

In all its history NF has had 5 and more trending on 78 occasions. Would we have made money if at the end of the fifth trending day we had bet that the market would reverse the next day? In practical terms it would mean that we take a short position at EOD if the mkt has gone up for 5 consecutive days and long position if it had declined for 5 consecutive days and hold it till EOD of sixth day.

You would have made money on 43 occasions and lost 35 times as the mkt continued its trend for more than five days. Percentage wise you would have made it 55% and lost it 45%. Not really a good thing to bet on. I have not gone into how much you would have made and lost just the number of times you would have made or lost.

The enquiry threw up some more things:
6 day trend has happened 14 times, 7 day trend 12 times, 8 day trend 6 times, 9 day trend 1 time, 10 day trend 1 time and 12 day trend 2 times!!

Less than 5 day trend was again no better than chance:

4 or more days trend occured 156 times where it terminated at 4 days 78 times and continued beyond 4 days 78 times. 50:50
3 or more days trend occured 318 times where it terminated at 3 days 162 times and continued beyond 3 days 156 times. Again 50:50.
2 or more days trend happened 609 times where it terminated at 2 days 291 times and continued 318 times. 47:53

Each and every unbiased enquiry into the market movement leads us inexorably to the single conclusion: market movements are random. market movements are random and market movements are random.

I have mentioned that there were two occasions with a 12 day trend. I don't know how many people can understand this: Such trends only prove the random nature of the market. If the sample were sufficiently bigger, you can find a 20 day trend too and still the mkt movement would be random.

Most people lose money because they do not understand the randomness. When a million people play the markets it is inevitable that a few would succeed with or without a method or a system. Randomness would require that. To the nature of the mkt when we add our own nature which sees what it wants to see, which is indisciplined, which has emotions like greed fear etc, we set ourselves up for disaster most times.

Thanks TT for sharing such a wonderful research with us here.

here is piece of my research (bit old from mid 2009 or so) on similar lines

Up #	Total	Cumm
0	749	42.3
1	356	62.4
2	228	75.3
3	156	84.1
4	100	89.8
5	66	93.5
6	41	95.8
7	24	97.2
8	18	98.2
9	11	98.8
10	7	99.2
11	4	99.4
12	3	99.6
13	3	99.8
14	2	99.9
15	2	100.0

Notes -
1) it covers about 1700+ days of data that includes the bull mkt of 2007 and bears of 2008..
2) when count = 0, those days were down days.
3) last column cummulates the % of times when mkt crossed that sequence.
so if we see 93% for 5, it indicates that only 7% of days, when mkt could continue with more then 5 days of bullish trend.

To me this is this is strong statistical odd in my favour to indicate to me that mkt is entering 2 standard deviation type of situation of showing bullishness.
Now how i convert this edge into tradeable system is different topic.
If i can manage the loss part of my trades, and let my profit run then and odd of 93% is not a simple edge in mkt. To me it is one of the highest odds that u can get.

similarly if u see > 8 updays in sequence, then there is hardly 2% chance that mkt will show us 9th positive day.. hence looking for shorting on 8th days is giving us 98% odd in our favour.

Generally, such a long seq of updays is indication of strong bullish momentum, so we may not see big bearish trend.. but if that happens on wave 5 stage, then the crash could be lottery ticket cause first drop of bear mkt of correction is the steepest and fastest

bottomline, as long as we can manage the risk, the market rewards the contrarians. in my observation some the biggest wins of top traders are thru their contrarian trades..

happy trading.

 

[AW10]

Be it Marc Faber or Warren Buffet, I'll listen to what I say (Anyday and Anytime). I am no Marc Faber or Warren Buffet, but someday, even if I become one, I want to be with my own opinion and my own vision. Who knows, I may well become one of them. To realize that "Guru Status", one needs to back his own views and literally shun other's opinion.

No offence to Mr. Faber or Mr. Buffet.

Tc

raunak, i do listen to you as well.. and will continue to do so..
but, yes you rightly pointed out that we should listen to everybody but take action based on our views. At times, probably we might disagree with them and that is ok

 

[AW10]

ya....that may be true !!

but it is not the answer to the question.....

traders are not born but made....

i entered the query because lot of seniors may be there....

and a method which worked in 7 out of 10 trades yesterday (26th)...i hope

is worth discussing

out of 3 losses one was simply that, that stock was in the ovrbought zone

so come on.....let us discuss

Ravi, in my view, success of trader is in 2 points
1) having a system with positive expectancy
where expectancy = (% win * avrg win size) - (% losers * avrg loss size).

2) and traders ability to execute the system correctly. Lots of psychology related issues come here.

IMO, You can very well trade a coin flip system with 50% win rate, as long as u can keep losses within limit and let yr winners run so that the system has +ive expectancy.

happy trading.