Why are value investor types so rigidly opposed to TA?

[kimchi]

[ QUOTE ]
how about now you deal with benoit mandelbrot...one of the smartest people of the 20th century, friend to and admired by albert einstein for his deep insights...founder of much of the mathematics behind much of chaos theory (fractal geometry)...and creator of the multifractal model of price changes.

it was with that last model that he fooled TAs. how do you explain that his simulations were so good that veteran technical analysts were pointing him towards all o fthe price action you speak of? how, if traders are the ones driving the market in all cases, can their actions be so amazingly closely be described by an entirely random process?


[/ QUOTE ]

Barron,

I've googled for some info on benoit mandelbrot and his 'TA trick' but to no avail. If you have a link about this, could you post it? It would be interesting to see.

I remember reading about Mandelbrot & his fractals in a trading book I read some time ago (can't remember which). Ironically perhaps, he was quoted in a book about technical analysis. He sees financial markets like the coastline of a country. No matter whether you view it from space (or view a yearly price chart) or under a microscope (an intra-day chart) they have the same intricate patterns making it difficult to dicypher the distance/time frame without a scale. I've also read briefly about Elliot Wave theory and Fibononacci retracements on price charts which seem to apply complex (and subjective imo) maths to the financial markets.

As for this ongoing squabble between FA and TA...I think it's fruitless. I think you guys should get together and use both disciplines. If an FA guy has an undervalued stock which may increase in value, but may take a few years, then why not use TA to decide when to buy? If you're waiting for a big move then why tie up your money in the meantime?

A good example of this would be William O'Neill. He advocated his CANSLIM method of screening companies using fundamental analysis. It's pretty basic, but the screen when properly done only yields a few companies (sometimes none). Then he uses some chart formations (cup & handle) followed by a breakout (confirmation) as an entry.

I think it's a very elegant use of both FA & TA. He uses FA to find set-ups and TA to pinpoint entries. He obviously leaves out some more important things such as proper decent risk management. He's wasn't a value investor, but would often buy companies trading at or above recent highs.

[DcifrThs]

[ QUOTE ]
[ QUOTE ]
how about now you deal with benoit mandelbrot...one of the smartest people of the 20th century, friend to and admired by albert einstein for his deep insights...founder of much of the mathematics behind much of chaos theory (fractal geometry)...and creator of the multifractal model of price changes.

it was with that last model that he fooled TAs. how do you explain that his simulations were so good that veteran technical analysts were pointing him towards all o fthe price action you speak of? how, if traders are the ones driving the market in all cases, can their actions be so amazingly closely be described by an entirely random process?


[/ QUOTE ]

Barron,

I've googled for some info on benoit mandelbrot and his 'TA trick' but to no avail. If you have a link about this, could you post it? It would be interesting to see.

I remember reading about Mandelbrot & his fractals in a trading book I read some time ago (can't remember which). Ironically perhaps, he was quoted in a book about technical analysis. He sees financial markets like the coastline of a country. No matter whether you view it from space (or view a yearly price chart) or under a microscope (an intra-day chart) they have the same intricate patterns making it difficult to dicypher the distance/time frame without a scale. I've also read briefly about Elliot Wave theory and Fibononacci retracements on price charts which seem to apply complex (and subjective imo) maths to the financial markets.

As for this ongoing squabble between FA and TA...I think it's fruitless. I think you guys should get together and use both disciplines. If an FA guy has an undervalued stock which may increase in value, but may take a few years, then why not use TA to decide when to buy? If you're waiting for a big move then why tie up your money in the meantime?

A good example of this would be William O'Neill. He advocated his CANSLIM method of screening companies using fundamental analysis. It's pretty basic, but the screen when properly done only yields a few companies (sometimes none). Then he uses some chart formations (cup & handle) followed by a breakout (confirmation) as an entry.

I think it's a very elegant use of both FA & TA. He uses FA to find set-ups and TA to pinpoint entries. He obviously leaves out some more important things such as proper decent risk management. He's wasn't a value investor, but would often buy companies trading at or above recent highs.

[/ QUOTE ]

kimchi,

nice post as usual.

the example i speak of with mandelbrot comes from hit book "the misbehavior of markets" linked here:

Linky

it isn't on the web as far as i know.

further, to get an understanding of the mathematics behind his theory, read the following book:

Linky

both of those books are crucial imo if you want to understand how price changes are truly distributed. his contribution in this area is 100% unparralleled.

i also like the idea of using both TA & FA. it seems that both the clear value in FA (spotting opportunities) and TA (determining short term psychological entry points) can lead to an optimal strategy.

it seems that many shops (read: top hedge funds) use either one or the other and in some cases both. my old fund didn't even THINK about TA. the founder sided 100% w/ B.B.M. and the entire quant system was built on logical and fundamental indicators of the drivers of asset prices in addition to momentum/flow factors.

i think i'm rambling now so i'll end:

to sum...read the mandelbrot books, they're worth it. TA & FA in conjunction can lead to a great outcome...aka i agree

Barron

[Mark1808]

[ QUOTE ]
[ QUOTE ]
how about now you deal with benoit mandelbrot...one of the smartest people of the 20th century, friend to and admired by albert einstein for his deep insights...founder of much of the mathematics behind much of chaos theory (fractal geometry)...and creator of the multifractal model of price changes.

it was with that last model that he fooled TAs. how do you explain that his simulations were so good that veteran technical analysts were pointing him towards all o fthe price action you speak of? how, if traders are the ones driving the market in all cases, can their actions be so amazingly closely be described by an entirely random process?


[/ QUOTE ]

Barron,

I've googled for some info on benoit mandelbrot and his 'TA trick' but to no avail. If you have a link about this, could you post it? It would be interesting to see.

I remember reading about Mandelbrot & his fractals in a trading book I read some time ago (can't remember which). Ironically perhaps, he was quoted in a book about technical analysis. He sees financial markets like the coastline of a country. No matter whether you view it from space (or view a yearly price chart) or under a microscope (an intra-day chart) they have the same intricate patterns making it difficult to dicypher the distance/time frame without a scale. I've also read briefly about Elliot Wave theory and Fibononacci retracements on price charts which seem to apply complex (and subjective imo) maths to the financial markets.

As for this ongoing squabble between FA and TA...I think it's fruitless. I think you guys should get together and use both disciplines. If an FA guy has an undervalued stock which may increase in value, but may take a few years, then why not use TA to decide when to buy? If you're waiting for a big move then why tie up your money in the meantime?

A good example of this would be William O'Neill. He advocated his CANSLIM method of screening companies using fundamental analysis. It's pretty basic, but the screen when properly done only yields a few companies (sometimes none). Then he uses some chart formations (cup & handle) followed by a breakout (confirmation) as an entry.

I think it's a very elegant use of both FA & TA. He uses FA to find set-ups and TA to pinpoint entries. He obviously leaves out some more important things such as proper decent risk management. He's wasn't a value investor, but would often buy companies trading at or above recent highs.

[/ QUOTE ]

If short term market movements are random (as they have proven to be) it wouldn't matter whether you used TA to time your entry or not, it just wouldn't add any value.

[stephenNUTS]

[ QUOTE ]
to sum...read the mandelbrot books, they're worth it. TA & FA in conjunction can lead to a great outcome...aka i agree

Barron

[/ QUOTE ]

[img]/images/graemlins/cool.gif[/img]

[DcifrThs]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
how about now you deal with benoit mandelbrot...one of the smartest people of the 20th century, friend to and admired by albert einstein for his deep insights...founder of much of the mathematics behind much of chaos theory (fractal geometry)...and creator of the multifractal model of price changes.

it was with that last model that he fooled TAs. how do you explain that his simulations were so good that veteran technical analysts were pointing him towards all o fthe price action you speak of? how, if traders are the ones driving the market in all cases, can their actions be so amazingly closely be described by an entirely random process?


[/ QUOTE ]

Barron,

I've googled for some info on benoit mandelbrot and his 'TA trick' but to no avail. If you have a link about this, could you post it? It would be interesting to see.

I remember reading about Mandelbrot & his fractals in a trading book I read some time ago (can't remember which). Ironically perhaps, he was quoted in a book about technical analysis. He sees financial markets like the coastline of a country. No matter whether you view it from space (or view a yearly price chart) or under a microscope (an intra-day chart) they have the same intricate patterns making it difficult to dicypher the distance/time frame without a scale. I've also read briefly about Elliot Wave theory and Fibononacci retracements on price charts which seem to apply complex (and subjective imo) maths to the financial markets.

As for this ongoing squabble between FA and TA...I think it's fruitless. I think you guys should get together and use both disciplines. If an FA guy has an undervalued stock which may increase in value, but may take a few years, then why not use TA to decide when to buy? If you're waiting for a big move then why tie up your money in the meantime?

A good example of this would be William O'Neill. He advocated his CANSLIM method of screening companies using fundamental analysis. It's pretty basic, but the screen when properly done only yields a few companies (sometimes none). Then he uses some chart formations (cup & handle) followed by a breakout (confirmation) as an entry.

I think it's a very elegant use of both FA & TA. He uses FA to find set-ups and TA to pinpoint entries. He obviously leaves out some more important things such as proper decent risk management. He's wasn't a value investor, but would often buy companies trading at or above recent highs.

[/ QUOTE ]

If short term market movements are random (as they have proven to be) it wouldn't matter whether you used TA to time your entry or not, it just wouldn't add any value.

[/ QUOTE ]
if they (S-T market movements) weren't random and based on psychological factors that drive human decisions, then TA combined w/ FA can provide higher overall returns than either in isolation.

Barron

[Mark1808]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
how about now you deal with benoit mandelbrot...one of the smartest people of the 20th century, friend to and admired by albert einstein for his deep insights...founder of much of the mathematics behind much of chaos theory (fractal geometry)...and creator of the multifractal model of price changes.

it was with that last model that he fooled TAs. how do you explain that his simulations were so good that veteran technical analysts were pointing him towards all o fthe price action you speak of? how, if traders are the ones driving the market in all cases, can their actions be so amazingly closely be described by an entirely random process?


[/ QUOTE ]

Barron,

I've googled for some info on benoit mandelbrot and his 'TA trick' but to no avail. If you have a link about this, could you post it? It would be interesting to see.

I remember reading about Mandelbrot & his fractals in a trading book I read some time ago (can't remember which). Ironically perhaps, he was quoted in a book about technical analysis. He sees financial markets like the coastline of a country. No matter whether you view it from space (or view a yearly price chart) or under a microscope (an intra-day chart) they have the same intricate patterns making it difficult to dicypher the distance/time frame without a scale. I've also read briefly about Elliot Wave theory and Fibononacci retracements on price charts which seem to apply complex (and subjective imo) maths to the financial markets.

As for this ongoing squabble between FA and TA...I think it's fruitless. I think you guys should get together and use both disciplines. If an FA guy has an undervalued stock which may increase in value, but may take a few years, then why not use TA to decide when to buy? If you're waiting for a big move then why tie up your money in the meantime?

A good example of this would be William O'Neill. He advocated his CANSLIM method of screening companies using fundamental analysis. It's pretty basic, but the screen when properly done only yields a few companies (sometimes none). Then he uses some chart formations (cup & handle) followed by a breakout (confirmation) as an entry.

I think it's a very elegant use of both FA & TA. He uses FA to find set-ups and TA to pinpoint entries. He obviously leaves out some more important things such as proper decent risk management. He's wasn't a value investor, but would often buy companies trading at or above recent highs.

[/ QUOTE ]

If short term market movements are random (as they have proven to be) it wouldn't matter whether you used TA to time your entry or not, it just wouldn't add any value.

[/ QUOTE ]
if they (S-T market movements) weren't random and based on psychological factors that drive human decisions, then TA combined w/ FA can provide higher overall returns than either in isolation.

Barron

[/ QUOTE ]

I believe the market is driven by psycholigical facors that is why FA can work long term, I just don't believe these psychological factors can be predicted with any sory of reliability near term. It is amazing that when a guy like Warren Buffett, who turned $100,000 in to $50 billion buying stocks, says TA is worthless why that wouldn't carry any weight with people.

[DcifrThs]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
how about now you deal with benoit mandelbrot...one of the smartest people of the 20th century, friend to and admired by albert einstein for his deep insights...founder of much of the mathematics behind much of chaos theory (fractal geometry)...and creator of the multifractal model of price changes.

it was with that last model that he fooled TAs. how do you explain that his simulations were so good that veteran technical analysts were pointing him towards all o fthe price action you speak of? how, if traders are the ones driving the market in all cases, can their actions be so amazingly closely be described by an entirely random process?


[/ QUOTE ]

Barron,

I've googled for some info on benoit mandelbrot and his 'TA trick' but to no avail. If you have a link about this, could you post it? It would be interesting to see.

I remember reading about Mandelbrot & his fractals in a trading book I read some time ago (can't remember which). Ironically perhaps, he was quoted in a book about technical analysis. He sees financial markets like the coastline of a country. No matter whether you view it from space (or view a yearly price chart) or under a microscope (an intra-day chart) they have the same intricate patterns making it difficult to dicypher the distance/time frame without a scale. I've also read briefly about Elliot Wave theory and Fibononacci retracements on price charts which seem to apply complex (and subjective imo) maths to the financial markets.

As for this ongoing squabble between FA and TA...I think it's fruitless. I think you guys should get together and use both disciplines. If an FA guy has an undervalued stock which may increase in value, but may take a few years, then why not use TA to decide when to buy? If you're waiting for a big move then why tie up your money in the meantime?

A good example of this would be William O'Neill. He advocated his CANSLIM method of screening companies using fundamental analysis. It's pretty basic, but the screen when properly done only yields a few companies (sometimes none). Then he uses some chart formations (cup & handle) followed by a breakout (confirmation) as an entry.

I think it's a very elegant use of both FA & TA. He uses FA to find set-ups and TA to pinpoint entries. He obviously leaves out some more important things such as proper decent risk management. He's wasn't a value investor, but would often buy companies trading at or above recent highs.

[/ QUOTE ]

If short term market movements are random (as they have proven to be) it wouldn't matter whether you used TA to time your entry or not, it just wouldn't add any value.

[/ QUOTE ]
if they (S-T market movements) weren't random and based on psychological factors that drive human decisions, then TA combined w/ FA can provide higher overall returns than either in isolation.

Barron

[/ QUOTE ]

I believe the market is driven by psycholigical facors that is why FA can work long term, I just don't believe these psychological factors can be predicted with any sory of reliability near term. It is amazing that when a guy like Warren Buffett, who turned $100,000 in to $50 billion buying stocks, says TA is worthless why that wouldn't carry any weight with people.

[/ QUOTE ]

but that is only 1 person. and it isn't actual "evidence" because it doesn't 100% deal with the issue.

it is obvious that i don't think that much of TA, but i don't think too little about it either.

i recognize that the indicators they use may be proxies for psychological factors. those factors are not likely to change and if they are actually contributors to excess returns to correct use of TA then that might explain why TA continues to exhibit, in some cases, solid returns.

Barron

[kimchi]

[ QUOTE ]
I believe the market is driven by psycholigical facors that is why FA can work long term, I just don't believe these psychological factors can be predicted with any sory of reliability near term. It is amazing that when a guy like Warren Buffett, who turned $100,000 in to $50 billion buying stocks, says TA is worthless why that wouldn't carry any weight with people.


[/ QUOTE ]

If you believe the market is driven by psychological factors, then surely this means those extremes in optimism and pessimism are tradable ST/MT with a disciplined approach?

Also, don't confuse reliablility of "predictions" with profitability. I trade a profitable system that loses money on 70% of the trades.

[Mark1808]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
how about now you deal with benoit mandelbrot...one of the smartest people of the 20th century, friend to and admired by albert einstein for his deep insights...founder of much of the mathematics behind much of chaos theory (fractal geometry)...and creator of the multifractal model of price changes.

it was with that last model that he fooled TAs. how do you explain that his simulations were so good that veteran technical analysts were pointing him towards all o fthe price action you speak of? how, if traders are the ones driving the market in all cases, can their actions be so amazingly closely be described by an entirely random process?


[/ QUOTE ]

Barron,

I've googled for some info on benoit mandelbrot and his 'TA trick' but to no avail. If you have a link about this, could you post it? It would be interesting to see.

I remember reading about Mandelbrot & his fractals in a trading book I read some time ago (can't remember which). Ironically perhaps, he was quoted in a book about technical analysis. He sees financial markets like the coastline of a country. No matter whether you view it from space (or view a yearly price chart) or under a microscope (an intra-day chart) they have the same intricate patterns making it difficult to dicypher the distance/time frame without a scale. I've also read briefly about Elliot Wave theory and Fibononacci retracements on price charts which seem to apply complex (and subjective imo) maths to the financial markets.

As for this ongoing squabble between FA and TA...I think it's fruitless. I think you guys should get together and use both disciplines. If an FA guy has an undervalued stock which may increase in value, but may take a few years, then why not use TA to decide when to buy? If you're waiting for a big move then why tie up your money in the meantime?

A good example of this would be William O'Neill. He advocated his CANSLIM method of screening companies using fundamental analysis. It's pretty basic, but the screen when properly done only yields a few companies (sometimes none). Then he uses some chart formations (cup & handle) followed by a breakout (confirmation) as an entry.

I think it's a very elegant use of both FA & TA. He uses FA to find set-ups and TA to pinpoint entries. He obviously leaves out some more important things such as proper decent risk management. He's wasn't a value investor, but would often buy companies trading at or above recent highs.

[/ QUOTE ]

If short term market movements are random (as they have proven to be) it wouldn't matter whether you used TA to time your entry or not, it just wouldn't add any value.

[/ QUOTE ]
if they (S-T market movements) weren't random and based on psychological factors that drive human decisions, then TA combined w/ FA can provide higher overall returns than either in isolation.

Barron

[/ QUOTE ]

I believe the market is driven by psycholigical facors that is why FA can work long term, I just don't believe these psychological factors can be predicted with any sory of reliability near term. It is amazing that when a guy like Warren Buffett, who turned $100,000 in to $50 billion buying stocks, says TA is worthless why that wouldn't carry any weight with people.

[/ QUOTE ]

but that is only 1 person. and it isn't actual "evidence" because it doesn't 100% deal with the issue.

it is obvious that i don't think that much of TA, but i don't think too little about it either.

i recognize that the indicators they use may be proxies for psychological factors. those factors are not likely to change and if they are actually contributors to excess returns to correct use of TA then that might explain why TA continues to exhibit, in some cases, solid returns.

Barron

[/ QUOTE ]

There is a group of people, all schooled under Benjamin Graham, who have compiled stellar records and don't use TA. If you take a million TA experts, a normal distribution curve around the market average should occur. Those that perform above the market feel they have a superior technique. Unless I see a strategy in writing that can be tested and verified with a track record I just don't believe it. Calling 63 coin flips out of a 100 does not make someone able to predict coin flips.

[DcifrThs]

[ QUOTE ]
[ QUOTE ]
I believe the market is driven by psycholigical facors that is why FA can work long term, I just don't believe these psychological factors can be predicted with any sory of reliability near term. It is amazing that when a guy like Warren Buffett, who turned $100,000 in to $50 billion buying stocks, says TA is worthless why that wouldn't carry any weight with people.


[/ QUOTE ]

If you believe the market is driven by psychological factors, then surely this means those extremes in optimism and pessimism are tradable ST/MT with a disciplined approach?

Also, don't confuse reliablility of "predictions" with profitability. I trade a profitable system that loses money on 70% of the trades.

[/ QUOTE ]

yea, confusing profitability w/ correct % is not good.

but, that said, it is the rarer profitable system that loses money more than it makes money.

my old fund made the most money when its hit ratio was >1.5:1 (i.e. 60/40 winner/loser).

this probably has to do with risk constraints though b/c if they let their highest conviction trades absorb more of their risk budget the overall EV may increase, but the volatility would likely increase by more in relative terms.

you've mentioned before that you don't have the ability to institute good risk controls so the differential between "hit ratio" and "profitability" might be accounted for in that...

just some thoughts,
Barron