Ok, fine. But then the fund manager must be counted in the same statistic that I mentioned above: "90% of all day traders lose money" - so now we say "90% of all traders lose money".
The 50/50 proposition has no basis.
A message to some day traders.
- Thread starter Amahrix
- Start date
Ok, fine. But then the fund manager must be counted in the same statistic that I mentioned above: "90% of all day traders lose money" - so now we say "90% of all traders lose money".
The 50/50 proposition has no basis.
The point is that, regardless whether they're day traders or hedge fund managers, if they fall within the 10% who are successful, then they have an edge. The edge can only be found in the data. If it's found in the data and it's repeatable, then it's not random.
No one knows the stats, we don't know how many dead bodies are in the graveyard (the failures) to count and compare against non-failures. That is not the point. The point is to highlight that by sheer luck, a small group in the large sample set can have long winning streaks doing what they're doing and it is possible to be sheer luck.
Traditional day trading using TA to look into past and into the future assumes
1) A deterministic view.
2) Assumes data carries a precise intended message of the future.
Reality is
1) Daily charts, intra-day charts are randomly constructed. Randomness meaning incomplete information.
2) Prediction doesn't predict where it matters/where it has the biggest impact, that is the tails. Data does not embed information about the future.
Wittgenstein’s ruler: Unless you have confidence in the ruler’s reliability, if you use a ruler to measure a table you may also be using the table to measure the ruler. The less you trust the ruler’s reliability, the more information you are getting about the ruler and the less about the table.
The ONLY test of intelligence that counts is to see if the person detects Wittgenstein's ruler situations: 1) When you use a ruler to measure the table, you are also using the table to measure the ruler. 2) The more unexpected the measurement, the more you apply W's ruler.
This means that you're actually getting more information about the tool (Technical Analysis in this case, and it's ineffectiveness) than about the chart you're studying.
But fools keep blaming everything but the ruler in their hands.................................................................... then you have a couple dozen that "succeed" and then the fools believe it is possible, combine that with those that are succeeding and thinking they're god and they start writing ebooks about how they do it, then when others apply their strategy they lose........because it doesn't work!
False. You're fooled by randomness. Also, you're basing this off of subjective probability. Even IF it were true, your conclusion is fallacious.
If this were true, then their results would be random as well. And they're not. Wittgenstein's ruler doesn't apply here because the ruler and the table are fixed, defined objects. Market data is dynamic. But in either case, the "ruler" can be anything from the size of a pin to a yard stick. The translation for the ruler and the table regardless which measures the other, will always be the same. Data is different. Measurement is interpreted as redundancy or repeatability. It was a poor analogy to use in his book.
You cannot prove conclusively that the data is random. You assume it's random because of the statistical results traders are achieving. But that still does not account for the 10%. And I understand we don't have accurate statistics - I mentioned that too. But the outcome appears to be the same - 10%, give or take a standard deviation, are successful. That says something about the data they're analyzing and how they're analyzing it. If it's repeatable, then it's not random.
Randomness, as it relates to the market, is incomplete information.
I am conclusively concluding that it is random because I am epistemically humble enough to tell you directly to your face that I do not have complete information of the marketplace. I am not God.
Take a look at Apples chart Friday before signal emerged in it in the form of China news.
Was the data that was constructed from 9:30am-1:13pm randomly constructed or did it carry a precise intended message of the future?
Generally, in almost all but a handful of trading days out of the year, the daily and intra day chart is constructed of purely random noise.
If you zoom out to yearly charts, you begin to see some signal, if you zoom out to 5 year charts, you begin to see even more signal in the data. But you cannot tell me you see signal when it really is noise, on a intra-day or daily chart. Just to note; even then you cannot accurately predict the future, because the data doesn't carry important information such as the tail events that have the most impact. This is the problem of induction proposed by Hume.
Imagine this: Read every single headline of the world daily for one year, then read the headlines every 3 months. Will you have a better sense of the world if you look at every headline daily, or look at the news every 3 months or so? Daily news is filled with noise, looking at the data (news) infrequently and over longer time periods will give you a better sense of the world because the noise will flush away, and the real impacting news will stick around.
Last edited:
Here's a recent statement by Jim Simons (Renaissance Technologies):
"Initially, the company was a fundamental investor. “Then it occurred to me that maybe price series were not random, completely,” said Simons who, with his wife Marilyn, is a major donor to the basic sciences. “There should be some ways to exploit the non-randomness. And we studied the question. I hired scientists, and started to come up with systems that seemed to beat fundamental trading, and gradually after many years, we are 100 percent systematic.”"
https://www.institutionalinvestor.com/article/b1dk61tspy6hxc/How-Jim-Simons-Scored-on-His-Math-SAT
If Jim Simmons had "incomplete" information, how was he able to exploit non-randomness??
You're implying that because you're not privy to certain information, that everything you do have is random. Randomness is a lack of predictability. That's very different than lacking information.
"Initially, the company was a fundamental investor. “Then it occurred to me that maybe price series were not random, completely,” said Simons who, with his wife Marilyn, is a major donor to the basic sciences. “There should be some ways to exploit the non-randomness. And we studied the question. I hired scientists, and started to come up with systems that seemed to beat fundamental trading, and gradually after many years, we are 100 percent systematic.”"
https://www.institutionalinvestor.com/article/b1dk61tspy6hxc/How-Jim-Simons-Scored-on-His-Math-SAT
If Jim Simmons had "incomplete" information, how was he able to exploit non-randomness??
You're implying that because you're not privy to certain information, that everything you do have is random. Randomness is a lack of predictability. That's very different than lacking information.
Renaissance profits from portfolio-level statistical arbitrage.
https://www.quora.com/What-are-the-...l-analyses-to-predict-non-equilibrium-changes
Don't strawman me, I never said "everything I do have is random". It is more random than you think. ESPECIALLY, when the data is 15 minutes old.
Answer my question on Apple chart prior and you'll have your answer if you're being honest with yourself.
Also, randomness is incomplete information, which then makes predictability more difficult. Nothing is random to an All-Knowing God because He encompasses all information and therefore can predict(know) the future with certainty.
The article reinforces the point. Statistical arbitrage relies on mean reversion to predict certain outcomes. That's structured data. Structured anything is not random. It has structure. Predictions can be made on repeatable structures.
Predictions don't predict where it counts. The tails!
This goes back to Renaissance's emphasis on survival and hedges against risk of ruin as the number 1 rule for successful trading that almost no retail trader understands. After understanding this, they can safely make any prediction they want to make. Renaissance is not foolish to rely on this prediction hence the immense effort in hedging everything before playing the game.
I am also saying that prediction is easier when you have signal. There is no signal in 30 minute data on intra-day, but signal emerges the farther you zoom out.
Answer my apple chart question.