The Truth of Trading

[MightyMouse 12]

The difference is that I think that he has an edge that is the original edge.

 

His adaptation results in continuous improvements to the edge.

 

Without the initial edge the adaptation might just result in his flapping in the wind.

 

 

I feel that I'm saying that a sailor has a edge from his sail and an edge multiplier by way of adjusting the sail to the wind strength and minor directional changes. And you might be saying that his edge is only the second part (what I define as an edge multiplier).

 

I think, and have never tried to verify, that no matter what the pattern, in the long run if you trade it perfectly, you'll likely break even if you trade it every time it appears. Break even really means losing since you have to pay the house a small amount to play. After a thousand trades, you are down +- 5000. On the other hand, if a trader has learned when to trade that pattern, when not to trade that pattern, and when to take the opposite side of that pattern, then he has an edge. I am minimizing the importance of the pattern.

 

I read MD's Zone and it was a great book to get me to break even. In a nutshell, me getting to BE meant ceasing to move my stops and targets out of fear and greed. It didn't take long to do, and I laugh now at how wimpy my behavior was. However, I do not think that a trader can use what he suggests as an entry and exit strategy for an extended period of time. And, I think he is off on his definition of an edge which seems to be a commonly used definition on these threads. I do not see how a trader would stumble trading if he thought he had an edge like a casino. You will never see the day when you are playing blackjack and you put your chips in the circle and the dealer second guesses whether he wants to take your bet or not. He knows in the long run that he is going to kick your ass.

 

My belief is that when you enter the market, you do not know what the odds of your trade really are, and what type of payout is available to you. So, you need to be prepared to be wrong, right or prepared to be more right or more wrong than you thought. The better you get at this, the more you will be able to take and keep.

[natedredd10 10]

My belief is that when you enter the market, you do not know what the odds of your trade really are, and what type of payout is available to you. So, you need to be prepared to be wrong, right or prepared to be more right or more wrong than you thought. The better you get at this, the more you will be able to take and keep.

 

First, "edge" is the expected value of a trade. Its really a shame how the term "edge" has taken off in trading as opposed to expected value. Edge in gambling is a very vague term, tends to be used as a yes or no thing for a bettor. He has an edge or he doesn't...Expected value is a more precise idea, I imagine "edge" took off though because you need quite a large sample size of actual trades for an EV calculation to have any meaning.

I don't agree with the first statement and the second you are just describing probabilistic thinking. I almost feel like there is some dissonance in your thinking because if you really can not know the odds and payouts available on a trade then it would be purely random as far as your results on pushing a trade when you are right or pulling a trade when you are wrong. Really, you almost saying trading is random with that but i think its because you are using "know" for odds and payouts still in an absolute sense as opposed to a probabilistic sense.

You can really only know the odds and payout of a bet in an absolute sense if someone is setting the line/payout and the bet is binary, win or lose. This is a totally different game than the markets because finding value in that kind of betting setup comes from betting against the line being set wrong as opposed to the outcome, which skews the payout wrong. If the house sets the line on a binary bet at -200/+300 but you believe the fav should be -600 and bet on the fav, both you and the house have the same belief on what side is going to win. You just disagree on the probability win rate.

On the other hand if your delt AA at a $2/5 no limit holdem table at a casino, while you can't say my expected value using only this single hand is X...you know without question you have an "edge" on this trade and if you have played enough hands you would have some idea of what the EV of the hand is in its most basic sense. Some idea of what the odds and payout will be.

[natedredd10 10]

In the above post ignore the rollover for "skew"...It would be a good idea to delete these types of things on here for terms that have exact mathematical definitions.

Its almost intellectually offense to me that a newbie could read that post and think I'm referring to some bogus concept as (VWAP - PVP)/SD if I didn't post this.

Of course it wouldn't make sense to have a rollover for skew that shows

 

If anything the roller for something with an exact mathematical definition should point to wikki like this

Skewness - Wikipedia, the free encyclopedia

 

To me in learning to trade there is right, wrong and way worse than wrong..Its good to find "wrong" because its not so far from "right"..Way worse than wrong though leads you down these nonsense paths into a maze that is hard to get out of and even if you do just wastes a lot of time.

 

I'm sure I'm on the extreme end here because if this was my board I would actually block the ability to post the words "fibonacci" and "gann" but thats a bit of a tangent.

 

Educating people that skew = (VWAP - PVP)/SD is an incredible disservice.

Edited July 17, 2010 by natedredd10

[MightyMouse 12]

First, "edge" is the expected value of a trade. Its really a shame how the term "edge" has taken off in trading as opposed to expected value. Edge in gambling is a very vague term, tends to be used as a yes or no thing for a bettor. He has an edge or he doesn't...Expected value is a more precise idea, I imagine "edge" took off though because you need quite a large sample size of actual trades for an EV calculation to have any meaning.

I don't agree with the first statement and the second you are just describing probabilistic thinking. I almost feel like there is some dissonance in your thinking because if you really can not know the odds and payouts available on a trade then it would be purely random as far as your results on pushing a trade when you are right or pulling a trade when you are wrong. Really, you almost saying trading is random with that but i think its because you are using "know" for odds and payouts still in an absolute sense as opposed to a probabilistic sense.

You can really only know the odds and payout of a bet in an absolute sense if someone is setting the line/payout and the bet is binary, win or lose. This is a totally different game than the markets because finding value in that kind of betting setup comes from betting against the line being set wrong as opposed to the outcome, which skews the payout wrong. If the house sets the line on a binary bet at -200/+300 but you believe the fav should be -600 and bet on the fav, both you and the house have the same belief on what side is going to win. You just disagree on the probability win rate.

On the other hand if your delt AA at a $2/5 no limit holdem table at a casino, while you can't say my expected value using only this single hand is X...you know without question you have an "edge" on this trade and if you have played enough hands you would have some idea of what the EV of the hand is in its most basic sense. Some idea of what the odds and payout will be.

 

A trader will have a problem with my first statement only if he thinks he needs to know the odds and payout of a trade before he enters a trade. The question is then can you trade if you really do not know the odds and payout and risk? My answer is yes.

[natedredd10 10]

A trader will have a problem with my first statement only if he thinks he needs to know the odds and payout of a trade before he enters a trade. The question is then can you trade if you really do not know the odds and payout and risk? My answer is yes.

 

The fact a trader can pull the trigger based on nonsense means nothing.

Good luck if you think you can trade without knowing the odds, risk and payout...

Utterly laughable.. Its hard for me to view trading other than the odds, risk and payout.

[MightyMouse 12]

The fact a trader can pull the trigger based on nonsense means nothing.

Good luck if you think you can trade without knowing the odds, risk and payout...

Utterly laughable.. Its hard for me to view trading other than the odds, risk and payout.

 

I agree that it is hard and discomforting and it requires a paradigm shift. Of course you can think you know the odds and take trades and make money and be none the wiser.

So if you back test your hypothesis and determine that it will provide you with a 2:1 payout 42% of the time and it works in real time, then you are a winner whether you know the odds or not. Similarly, a Gann trader can populate his chart with Gann Square of 9 data and take trades based on it and swear the numbers have meaning and make money using it.

 

Ultimately, if you have figured out how to take money from the market, that is all that matters.

[BlowFish 10]

 

Based on the assumption the markets are random 'most of the time' to the unaware, .....

 

That's a whole can of worms opened right there! I'm going to say nothing more than that is quite an assumption.

Edited July 20, 2010 by BlowFish

[BlowFish 10]

 

Educating people that skew = (VWAP - PVP)/SD is an incredible disservice.

 

Having said that Pearsons contribution in the field have been pretty worthwhile imho. I am not sure anyone is 'education people' that his is the same as the skew. It is simply a quick and dirty. Determining 3 central moments of a weighted distribution with a continuous algorithm is a very tricky problem. Very. If you are happy to use an iterative approach it is fairly trivial of course and that option is always open to you. So you have two choices, use one of Pearsons approximations or use an algorithm that does not scale with large data sets.

[i trade 10]

Speaking from a purely discretionary stand point...I think an important aspect of trading "an edge" (or you can say it's part of your edge) that's hasn't been mentioned is varying the leverage used depending on the underlying context of the market in which your "edge" is setting up in. Most retail traders either {a} use max leverage all the time on their entries which requires you to be amazingly accurate with every entry or {b} they don't know when to up the leverage and increase their position size when underlying market conditions warrant it.

 

Everyone loves the poker analogy : picking up pocket 10's or picking up pocket aces, both hands you would want to play but then again 2 hands that you would probably play differently pre-flop not only according to the quality of the starting hand itself but also according to the context of the situation - are you 1st to act or last to act, has any1 limped, has any1 raised, how many chips do you have...etc , etc.

[TheDude 10]

The fact a trader can pull the trigger based on nonsense means nothing.

Good luck if you think you can trade without knowing the odds, risk and payout...

Utterly laughable.. Its hard for me to view trading other than the odds, risk and payout.

 

You only think you do.

 

If I were a gambling man (I'm not), I'd say you base these figures on past outcomes of a series of trades.

 

However, the past is no guarantee of future events. That kind of puts the kyber on the idea of backtesting and the scientific approach to trading I guess. Ho hum.

 

If you take the trade at the singular level, then MM is right.

If you take the trade in the context of a series of trades you may be right.

 

Thats where the skill of the (good) discretionary trader stomps all over the mechanical trader. The discretionary trader can see the probability unfold in real time. He is working and focusing and calculating all the time. He knows whether to let the trade run for more, or whether to cut it before a loss is incurred. Sometime he gets it wrong, but thats the market - it can do unexpected things suddenly. The mechanical trader is a lazy trader. He takes his trade and hopes it fits the average - probably while listening to music or something.

Edited July 20, 2010 by TheDude

[Kiwi 16]

However, the past is no guarantee of future events. That kind of puts the kyber on the idea of backtesting and the scientific approach to trading I guess. Ho hum.

 

That's really something that is said to protect people.

 

All trading, be it 100% systematic or 100% discretionary is based on the assumption that the past is a good "guarantee" of future events but guarantee is obviously too strong. The adaptive systems or discretionary trader (never forget that a lot of those failed after the 1990s bull market finally topped) realizes that markets change, sometimes in cycles or waves, sometimes in ways that have never happened before. At that point your predictors fail and you have to adapt.

 

Darwin didn't say "survival of the fittest" he said "survival of the adaptable." Carrying on with evolution many people think that natural selection creates perfect creatures - but in fact it is a satisficing mechanism and it simply creates a creature now sufficiently adapted to the current conditions.

 

The lessons we might take from evolution are:

- the past is a good predictor of future events (but sooner or later it won't be so think adaptation not perfect for ever)

- the traders job is to satisfactorily fit his niche picking out bucks often enough to survive and procreate but the job spec doesn't need you to be perfect or to be best.

 

Good enough is good enough.

[macdfx 10]

The lessons we might take from evolution are:

- the past is a good predictor of future events (but sooner or later it won't be so think adaptation not perfect for ever)

.

 

Exactly. You never know when a six-sigma event may happen.

 

In the markets. Or in nature, like this:

 

 

Happy trading!

[TheDude 10]

Well perhaps I should put some more definition on my statement.

 

This is my view, and I understand and welcome different opinions:

 

In some ways the past can be a useful insight into future events. By that, I refer to human emotion and it's corresponding behaviour. That you can see in the market, but it's quite difficult to backtest as you probably cant get a good feel for traders emotional levels looking at last years charts.

 

In other ways the past is no use at all, especially when developing mechanical systems. If it were, then someone, somewhere, by now would have developed that killer system. How many traders around the world over time have developed and tested such systems based on indicators, candle shapes, TA, etc, then optimised, then forward tested, etc just to see it fall over? If such a system did exist, doesnt anyone here think it would have been discovered by now? Maybe that method did work in the past for a window of time, but by the time it takes to discover, back test, forward test etc, the cycle changes just as the trader starts to trade it.

 

Magic? Nope. Victor N points out the same thing happens in horse racing. The gambler studies the form, realises a pattern has emerged, just as the others who study form identify the same or very similar patterns. The bookie then has to change the odds as everyone bets the same way and so the system no longer works.

[bakrob99 11]

I have traded both systematically with 100% automation and on a 100% discretionary basis. After many years of watching the ES and trading it .. I have discovered that the actual process of trading during the day while the tape is running and the indicators (Tick, Trin etc.) etc are making alerts... has somehow now worked inside my brain to give me a pretty good idea of what the market condition is like. This is completely missing on backtest / review of historical data. Looking back, you can see what happened to the price bars... but you can't "feel" them.

 

Now I find that seeing the time and sales and watching how traders are lifting the offer (or not) gives me the occasional (dare I say) "gut feel" for the move. Not for the entry technique. Of course, I have to rely on my entry criteria and be patient and watch for a setup ... but one develops a feel after time spent. And one knows that the time will come. Don't chase.

 

The truth about trading for me is that when I first started I would make the normal mistakes of entering after a move had started and exiting on a pullback or worse... Newbie mistakes. I went thru a period of being " total mechanical or systematic" to get around this market trickery.

 

Now, as the trading day develops, I absorb, perhaps subconsciously, market price action and often can tell when a market is done... and about to sell off hard... or when it is about to run parabolically and positions should be held, and added to, when possible. I look back on entries a week later and wonder then how I knew at the time...

 

My Truth about trading is: practice and screen time is required before one can confidently make consistent money.

Edited July 21, 2010 by bakrob99

[SultanofSwing 10]

I remember when I first started, most of my strategies revolved around waiting for a major news event and predictions of how that event would affect the given market. Sadly, this strategy worked less than half the time, because it's based on "feelings." When these "feelings" can be predicted, that can be planned against, thus tipping the scale back to unpredictability. This same outcome can be applied to really any strategy; expectation versus the amount of people using the same strategy.

 

Whether or not there is a group of an elite few who actually control the strings of the market is unknown and rather irrelevent.

 

My truth? "Be like water my friend!"

[i trade 10]

I have discovered that the actual process of trading during the day while the tape is running and the indicators (Tick, Trin etc.) etc are making alerts... has somehow now worked inside my brain to give me a pretty good idea of what the market condition is like. This is completely missing on backtest / review of historical data. Looking back, you can see what happened to the price bars... but you can't "feel" them.

 

Good point....the condition or context of the market, not only on a short term basis but maybe more importantly on the background or longer term, is often overlooked when testing strategies or an "edge".

 

My truth be told.....successful traders realize that setups for both winning and losing trades look identical when looked at only in the context of short term market activity.....its the longer term background that can distinguish between them.

[natedredd10 10]

That's a whole can of worms opened right there! I'm going to say nothing more than that is quite an assumption.

 

To me traders have an almost moral obligation to destroy this idea when it comes up.

Either markets are forward pricing economic activity or they are not. If you believe markets are not pricing forward economic activity then I don't see why you would bother since you can not gamble on a random walk at profit.

If you isolate my economic activity today and think the reason I didn't buy an HD projector tv because i flipped a fair coin and gasoline/pizza won out, producing a random walk in my economic behavior then i don't know what to tell you.

Its not hard to figure out that Paul Samuelson is a fraud and that the psuedo science of the economics profession has a feedback loop because of self interest that amplifies nonsense like a religion.

[natedredd10 10]

Having said that Pearsons contribution in the field have been pretty worthwhile imho. I am not sure anyone is 'education people' that his is the same as the skew. It is simply a quick and dirty. Determining 3 central moments of a weighted distribution with a continuous algorithm is a very tricky problem. Very. If you are happy to use an iterative approach it is fairly trivial of course and that option is always open to you. So you have two choices, use one of Pearsons approximations or use an algorithm that does not scale with large data sets.

 

The problem is there is zero evidence that volume has anything to do with the return distribution of markets. My point though was that there is no reason to highlight "zero" in my last sentence and inform people its something other than the exact mathematical definition no matter how many post were generated if I posted a system I called "zero"...

It would be a huge disservice to anyone who had never ran across the concept of zero.

[BlowFish 10]

Ahh OK It sounded as if you had an issue with Pearsons approximation. Again I don't think anyone is suggesting that 'volume has anything to do with the return distribution of markets' it is a very useful metric if you know what it represents and how to use it.

 

Weighting by volume makes complete sense if you think 10 trading at price p carries more weight that 1 trading at price p. It obviously makes sense to use as a benchmark for anyone wanting to judge a traders performance (the price the trader got for their lots vs the average price for the total sample). There really is no other 'fair' way to do that if you are price motivated. Of course if you are time motivated you might use TWAP.

 

Between you me and the gatepost, Im inclined to agree about random walk but didn't really want to be the torch bearer on this occasion

[clmacdougall 11]

Blowfish, I absolutely enjoyed your post on recognizing an "edge". Please carry it forward as it could be a great thread.

Thanks again!

[Kiwi 16]

Its always interesting to see the debate.

 

Even more interesting when I refrain from leaping in with "you bloody idiot ... you don't know what you're talking about." I do think it but I won't say who I thought it about this time. I'm trying to master my desire to fix your thinking.

 

A new poster, mb3000, posted this link in answer to a request of Al Brooks videos. People responding to or just reading this thread might enjoy the video on that page by Jack Schwager - Winning Methods of the Market Wizards. He's definitely on topic.

 

.

[MightyMouse 12]

Its always interesting to see the debate.

 

Even more interesting when I refrain from leaping in with "you bloody idiot ... you don't know what you're talking about." I do think it but I won't say who I thought it about this time. I'm trying to master my desire to fix your thinking.

 

A new poster, mb3000, posted this link in answer to a request of Al Brooks videos. People responding to or just reading this thread might enjoy the video on that page by Jack Schwager - Winning Methods of the Market Wizards. He's definitely on topic.

 

.

 

It's nice of Al Brooks to package and deliver his trading methodology.

[natedredd10 10]

Ahh OK It sounded as if you had an issue with Pearsons approximation. Again I don't think anyone is suggesting that 'volume has anything to do with the return distribution of markets' it is a very useful metric if you know what it represents and how to use it.

 

Weighting by volume makes complete sense if you think 10 trading at price p carries more weight that 1 trading at price p. It obviously makes sense to use as a benchmark for anyone wanting to judge a traders performance (the price the trader got for their lots vs the average price for the total sample). There really is no other 'fair' way to do that if you are price motivated. Of course if you are time motivated you might use TWAP.

 

These WAPS don't make sense at all...like i said..latest nonsense to tick down to retail thinking they have found something..Shit, this board is basically built from that BS..

Way easy to turn these algos non discreet..As if that would be some amazing thought.

TWAP just as nonsense. Not hard to figure retail guys are fucking fools.

[mktwzrd 10]

These WAPS don't make sense at all...like i said..latest nonsense to tick down to retail thinking they have found something..Shit, this board is basically built from that BS..

Way easy to turn these algos non discreet..As if that would be some amazing thought.

TWAP just as nonsense. Not hard to figure retail guys are fucking fools.

 

Calm down, natedredd10...you'll give yourself a tummy ache.

[cornhusker 10]

The problem is there is zero evidence that volume has anything to do with the return distribution of markets. My point though was that there is no reason to highlight "zero" in my last sentence and inform people its something other than the exact mathematical definition no matter how many post were generated if I posted a system I called "zero"...

It would be a huge disservice to anyone who had never ran across the concept of zero.

 

Hi Nate,

 

I've thoroughly enjoyed all your posts and will continue to follow you in the future. I appreciate the depth and knowledge embedded in your comments.

 

About your post I've quoted: Is there evidence to the contrary--that volume has no bearing on returns? I am most likely deeply misunderstanding what you mean, but isn't volume i.e. liquidity consumption at the root of all market returns? Or do you mean that volume right now has poor predictive value for price ten minutes from now? Maybe you mean neither.

 

Thanks