Price, (Volume), Support, Resistance, Demand, Supply . . .

[dbphoenix]

May I suggest you try a different approach temporarily before analysis by paralysis completely takes over?

The market is real so why not try to engage with it rather than struggle from a perspective of confusion and perplexity?

IMO until you can learn to 'let go' than no number of high probability signals would be of any use to you.

When the DAX opens next (I chose this because you mentioned it previously)just passively observe the price action alone and nothing else. The aim is to become the market, to internalise and flow with the other thousands of traders doing the same as you. This may sound touch feely BS but just try it. After a while you'll see where all the S & R levels are and more importantly the character of this market will become something you know and with much practise will be to some extent become predictable. Then when that stage is reached it will be time to get the charts back up and depending on how your unique intuitive reading of the market has formed then you in turn will be able to develop your own strategies relevent to your interpretation.

I hope this is of some help...

Jeez, rols, you're beginning to sound like me. And while I don't want to talk about FW as if he weren't here, and while at least some of this belongs in his journal, what's wrapped around his ankles and causing him to stumble and fall is much the same as what's wrapped around the ankles of everybody who stumbles and falls: looking for a set of instructions.

This search for instructions as to where EXACTLY to draw the line is in large part what makes Pivots and Fib and Gann and MAs and so forth so seductive. One doesn't have to think about just where it is that price(traders) really react. All the trader has to do is draw the calculated lines. This search for exactitude also motivates the search for the EXACT stop and exact TYPE of stop that the trader should use, along with the EXACT trigger and the EXACT target. But if it were all that simple, one could package it into a kit and sell it (wait a minute . . . ).

Many people can't get this. Maybe most people can't get it. They simply cannot trade without indicators, they can't trade without patterns, they can't trade without candlesticks, etc. And if they make money doing whatever they're doing, who's to say they're not right to do it. However, a lot of people also struggle with all of that and can't make money at it. They find instead that focusing on price is best for them. Unfortunately, by the time they reach that point, they have to unlearn an extraordinary amount of what for them is generally -- or entirely -- useless information (I've read that . . . People say that . . . I've been told that . . . ). This state of affairs makes learning to trade by price vastly more difficult than it would have been had the trader learned how to do it outright in the first place. But there's no going back, this side of amnesia, so wanting to is simply wishful thinking.

This isn't the first time I've heard all of this, of course. And the Go With the Force, Luke stuff only goes so far, true as it may be. But the individual who's willing to backtrack and learn a new or at least different way of looking at charts and price action may -- not will -- find that when he's looking at his umpteenth chart, the light suddenly goes on and he understands all those back and forth pressures which are propelling price one way or the other. All the babble about pace and momentum and trend and chop and all the rest of it will make sense.

But there's no shortcut. One may have to look at hundreds of charts. Maybe thousands. And he may never get it. Which is why people continue to spend so much money on 4x Made Easy and Weekend Seminar (lunch included) and Profits R Us.

Db

 

[ljyoung]

What would then be a good definition of a high probability signal? If I were to look for one, I need to know actually what that is.

Fw,

One of the best definitions of probability I have come across is the one given by Richard Feynman, a bongo-playing, theoretical physicist who said something to the effect, "... and so how probable is it that as you drive down the freeway, you're going to see a license plate with the numbers 1234567 or 6487190 or 5463829 or whatever? The answer is that it's incredibly improbable, but it's no more improbable (or probable) than if, prior to your setting out for your excursion, you set an initial condition of 'I wonder if I'll see a license plate with the numbers 1234567 on it'". Feynman's thought is exactly what db is talking about in his response. So one cannot expect that ones set of initial conditions will affect the outcome of a formulated observation.

A way around this apparent conundrum is to empirically generate a set of observables and determine how many of them have to come to pass for you to have a profitable trade. Eclecticism is apposite. To have the whole wad resting on whether you've correctly interpreted the latest signal from, let's say, Elliott Wave Theory, is to court disaster (speaking from personal experience on that one).

Picture now the Pythons, sitting on chairs discussing who was the poorest and I'd like to be Michael Palin with my newest story just placed on the table. As a medical student, when the teaching clinician would preface a remark with "In my experience, ....", [vide supra] it was the signal for myself and others of my ilk, to bolt for the doors - if not physically then most certainly mentally. In retrospect it would have been better to let the old windbag have his say [insert a Charles Schultz, wah wah wah - wah wah wah wah] and let the weight of evidence decide if his thought had merit. Sometimes it takes a while to accumulate the necessary evidence - sometimes it can happen rather quickly.

ljey

Primum non nocere - Florence Nightingale

 

[dbphoenix]

So one cannot expect that ones set of initial conditions will affect the outcome of a formulated observation.

Which is what Douglas is saying when he states that regardless of how high the probability derived from one's testing may be for the success of a particular setup, the outcome of any given trade is unknowable. In other words, you don't have to know or even care whether a given trade is going to be successful. You know, or should know, from your testing that over time, i.e., over a series of trades, a given percentage of them +/- are going to be successful. If the setup isn't clearly defined, much less tested, then none of this applies, and the probability of success for any given trade is pretty much left to chance.

Db

 

[firewalker99]

Which is what Douglas is saying when he states that regardless of how high the probability derived from one's testing may be for the success of a particular setup, the outcome of any given trade is unknowable. In other words, you don't have to know or even care whether a given trade is going to be successful. You know, or should know, from your testing that over time, i.e., over a series of trades, a given percentage of them +/- are going to be successful. If the setup isn't clearly defined, much less tested, then none of this applies, and the probability of success for any given trade is pretty much left to chance.

Db

So you're actually saying that the percentage doesn't matter. What matters if you cleary define and outline the rules for a setup that define an entry, so you have a solid basis for comparing the result of trades. If the setup isn't cleary defined or tested, you're not analyzing the probability of a system because you're analyzing random trades. Say one would have a setup which over time show a success ratio of 20%, then would you not be inclined to think that doing exactly the opposite of each setup (including reversing target, exit, risk/reward,...) you would have a success ratio of 80%?

So does it really matter whether your system has a 40% or 70% chance of being successful? Why would you search for a high probability setup than? If you know your system has a low 30% success rate, but each succesful trade makes you win 10 points and each failure makes you lose 4, the net result would still be positive.

 

[firewalker99]

Fw,

One of the best definitions of probability I have come across is the one given by Richard Feynman, a bongo-playing, theoretical physicist who said something to the effect, "... and so how probable is it that as you drive down the freeway, you're going to see a license plate with the numbers 1234567 or 6487190 or 5463829 or whatever? The answer is that it's incredibly improbable, but it's no more improbable (or probable) than if, prior to your setting out for your excursion, you set an initial condition of 'I wonder if I'll see a license plate with the numbers 1234567 on it'". Feynman's thought is exactly what db is talking about in his response. So one cannot expect that ones set of initial conditions will affect the outcome of a formulated observation.

I know people have a tendency to believe it's near improbable to see what we believe to be a specifically chosen string of characters, numbers, patterns as opposed to a completely random one. But you're right, chances af catching a license plate 1234567 or the same as catching one 4591327. I think a great deal lies in perception.

I'm off on a tangent here...people play the lottery because they see people winning every time although they forget that winning it has actually a smaller chance than being hit by lightning. But how many times have you heard of someone being hit by lightning? Exactly, almost none. And how many times have you heard somebody win the lottery? Weekly I suppose...

 

[dbphoenix]

So you're actually saying that the percentage doesn't matter. What matters if you cleary define and outline the rules for a setup that define an entry, so you have a solid basis for comparing the result of trades. If the setup isn't cleary defined or tested, you're not analyzing the probability of a system because you're analyzing random trades. Say one would have a setup which over time show a success ratio of 20%, then would you not be inclined to think that doing exactly the opposite of each setup (including reversing target, exit, risk/reward,...) you would have a success ratio of 80%?

No, I'm not saying that percentage doesn't matter. Not saying that at all.

As for the rest of your question, try it. Let me know how it all works out.

So does it really matter whether your system has a 40% or 70% chance of being successful? Why would you search for a high probability setup than? If you know your system has a low 30% success rate, but each succesful trade makes you win 10 points and each failure makes you lose 4, the net result would still be positive.

So out of every 10 trades, you gain two points, then deduct commissions, slippage, trading costs, and so on. I can't imagine any more rewarding way of spending my time.

 

[firewalker99]

So out of every 10 trades, you gain two points, then deduct commissions, slippage, trading costs, and so on. I can't imagine any more rewarding way of spending my time.

Ok, perhaps a bad example because the reward is only 2,5 times the risk. Anyhow, I think you know what I meant. You can gain profit is your setups only are -let's say 30%- successful but have a high enough return to compensate for the many, but small losses.

 

[dbphoenix]

Ok, perhaps a bad example because the reward is only 2,5 times the risk. Anyhow, I think you know what I meant. You can gain profit is your setups only are -let's say 30%- successful but have a high enough return to compensate for the many, but small losses.

Yes, I know what you meant. And, as I said, you can make money that way. But you can only make real-life money -- not textbook money -- this way by trading size. Are you willing to trade size with only a 30% win rate given where you are in your trading?

Db

 

[firewalker99]

Yes, I know what you meant. And, as I said, you can make money that way. But you can only make real-life money -- not textbook money -- this way by trading size. Are you willing to trade size with only a 30% win rate given where you are in your trading?

Db

You're right of course, and I was only hypothesizing, not implying that I would rely on a system with a 30% success ratio. In any case, I'm getting the feeling that finding a system that has a high rate for success will probably remain an illustrous goal for me. I'm not giving up, but things are looking bleak.

 

[ljyoung]

Fw,

re: post 792

FWIW what I see in your chart is:

1. For sure HH's and HL's (see post 797). A rough rule of thumb is, in the absence of a really impressive upthrust, to look for 2 LH's and 2 LL's before thinking of going short and even then you get whacked not infrequently.
2. If 1. is so, then you're in an uptrend and so what you're doing is"top-picking" = death by a 1000 cuts.
3. You do not wait for confirmation before entering - in trade 1 you shorted after a gap up; in trade 2 the next bar was up impressively; in trade 3 volume is falling with the big black bar and continues to fall.
5. If you are going to use volume you need some sort of a volume MA. I use 30 EMA but find your own.
6. If you are going to use VSA, it must be put in context. Using single bars in isolation (or semi-isolation) is NOT context.

On the other hand, if you step back and look at the overall chart action, you are right in the sense that we are getting higher prices on lower volume but you need to be in a different time frame if you're going to trade on that. Interestingly the far upper right portion shows the first LH/LL combo suggesting that things might be getting ready to roll over on the 3 minute bar (what about the 1,5,15,30,60 minute bars?).
As Mr. Kessler's friend in the book "Wall Street Meat" remarked, "... you want to come to the party early, but not too early". Although all things begin on the 1 minute bar (the tick count actually but that for me is really too much noise), the S/N level sucks. It's not much better on the 3 minute bar. So if you want to get an early start you need confirmation and you need to be aware of what the market is doing right now in more than one time frame.

The last thing relates to the fact that your opponents in the market,e.g., other traders, MM's and smart money, undoubtedly know all about VSA and probably every other method you might know. So in terms of raw facts there ain't much difference. The field is not tilted in your direction in these short time frames. Livermore avoided them like the plague. So maybe trade smaller on a longer time frame - the S/N is better. It will also force you to be more patient.

ljey

 

[firewalker99]

1. For sure HH's and HL's (see post 797). A rough rule of thumb is, in the absence of a really impressive upthrust, to look for 2 LH's and 2 LL's before thinking of going short and even then you get whacked not infrequently.
2. If 1. is so, then you're in an uptrend and so what you're doing is"top-picking" = death by a 1000 cuts.
3. You do not wait for confirmation before entering - in trade 1 you shorted after a gap up; in trade 2 the next bar was up impressively; in trade 3 volume is falling with the big black bar and continues to fall.
5. If you are going to use volume you need some sort of a volume MA. I use 30 EMA but find your own.
6. If you are going to use VSA, it must be put in context. Using single bars in isolation (or semi-isolation) is NOT context.

ljey

Thanks, I appreciate the feedback. Is that a rule of thumb you found out for yourself?
Agreed, shouldn't be top picking. But instead of wanting to buy at an increasingly higher price, I believed shorting offered a better opportunity. About this time frame though, I already switched from 2 to 3 minutes, but I find that a 5 min chart does include a serious difference in bar spread. That would mean taking more risk, which I'm not prepared to do right now. Yes, less contracts but I'm already at my minimum. But I guess I have no choice than giving it a shot as lot's of people have advised me now to take a step back and look at a bigger time frame.

 

[ljyoung]

Thanks, I appreciate the feedback. Is that a rule of thumb you find out yourself?
Agreed, shouldn't be top picking. But instead of wanting to buy at an increasingly higher price, I believed shorting offered a better opportunity. About this time frame though, I already switched from 2 to 3 minutes, but I find that a 5 min chart does include a serious difference in bar spread. That would mean taking more risk, which I'm not prepared to do right now. Yes, less contracts but I'm already at my minimum. But I guess I have no choice than giving it a shot as lot's of people have advised me now to take a step back and look at a bigger time frame.

Yes a ROT by T&E but I should say it only applies to the GOOG.
Another tack you can use is to trade a stock, index or ETF which has a lower volatilty than what you have been trading. This allows you to redefine your risk (like trading MSFT instead of CME). It may take you longer to make some money but at least your capital base will be less at risk. I am moving off topic and so will STFU.

Best of luck

ljey

 

[dbphoenix]

Since we're getting into trading advice, perhaps FW's journal would be more appropriate to this particular exchange.

Db

 

[SIMA]

.

As for my trading, I've explained that several times, but to no one's satisfaction. The S/R thread provides several examples of what I use. Maybe it's too simple. In any case, my telling fw to trade the way I do would be just one more layer, and he's not that far away from being crushed by the weight of all this. His task is not to add but to subtract.

Db

o.k, had not seen that thread before, notice red pointers on the charts suggesting breakout trades, if so, do you then zoom down to say 1min or tick charts to gain entry with say market orders or limit orders

 

[ymonly]

the problem for me trading low percentage system is emotional capital spent dealing with entering most trades knowing you are going to lose.........the psychological stuff is big..........for most........

 

[dbphoenix]

Whether one plays reversals, breakouts or retracements here are the entry points what I see:

1 Reversal from support

2 Reversal from resistance

3 Breakout

4 Retracement to resistance

5 Retracement to resistance

If someone plays all or only one of them is only up to his trading strategy. Since I'm trading retracements, I would only enter 4 and 5.

Maybe there would be additional entry points if we could see the context from the previous days or weeks or whatever is here the bars represent.

Okay. The bars beginning at "10" are the beginning of the previously-posted chart, re-posted here for convenience.

 

[Szimba]

I attach both charts.
This is a very good example how S/R works!
I reposted my original chart and made some annotations which makes easier to compare both charts.
We can see, that S/R didn't came from nothing, instead it's origination could be find in the previous chart.

The context didn't change my previous view because it didn't add any new potential support or resistance.

The only difference is that now we can understand why did price find support around 48 after breaking down at red3 and the same applies for the area around 42. (where price find support before breaking stage 1 trendline)

 

[Szimba]

Ooops I forgot to attach the charts.

 

[firewalker99]

Ooops I forgot to attach the charts.

I've noticed your lines are pretty similar to the ones I've drawn (except they're longer but that's beside the point). One question however: the upper line around 51, why not around 52? Do you consider this to be a zone. I tried to interpret the chart reading from left to right only, but noticed that a S/R line drawn at 52 - so not at the outer edges of the bottoms - crosses with a new top between C-D. Let's say you drew a S at 51, what would you make of that particular price movement then?

 

[firewalker99]

Okay. The bars beginning at "10" are the beginning of the previously-posted chart, re-posted here for convenience.

I've noticed you've already posted these charts in a somewhat different context at another place on this forum. Perhaps you were getting to that to make a comparison - and I don't mean to jump the gun - but I have some questions regarding the lines on the original chart...