S/R and the Mirror of Erised

[ljyoung]

Hope you don't mind me saying, but it would think it's safer to start all trendlines from the same point of origin. In this case, the highest point near the top of the chart above 130.2...

I most definitely do not mind you saying what you did. Let me run the thought process I used to generate those lines by you. As I mentioned in a very early post on the P(V) thread, I believe that the Victor Sperandeo protocol for drawing trend lines is the best and it is outlined in both of his books (just google his name and you'll get the titles of the books - good stuff if I might say). Briefly put: beginning at a swing high, draw the line from that point through the point which is the highest high before the next swing low without passing through any other lines. You will excuse the rather liberal use of the term "swing" but it is meant in a more time-constrained manner than its usual sense. What I did then was to shift the starting point for the trend line whenever a support area was broken, thereby generating a new line while remaining rigorously faithful to Sperandeo's rule. How does that old song go "Call me anal-retentive, call me ...."?

Db clearly agrees with your point and while I am not trying to form a heretical sect and well appreciate that he (db) has talked about how he draws these lines, I thought it was interesting to see how price played out when in the vicinity of these lines. It is very true that any interaction between price and the line may be more apparent than real but trend lines do appear to be useful for at the least, alerting one to a possible change in the mood of price.

ljey

 

[firewalker99]

I most definitely do not mind you saying what you did. Let me run the thought process I used to generate those lines by you. As I mentioned in a very early post on the P(V) thread, I believe that the Victor Sperandeo protocol for drawing trend lines is the best and it is outlined in both of his books (just google his name and you'll get the titles of the books - good stuff if I might say). Briefly put: beginning at a swing high, draw the line from that point through the point which is the highest high before the next swing low without passing through any other lines. You will excuse the rather liberal use of the term "swing" but it is meant in a more time-constrained manner than its usual sense. What I did then was to shift the starting point for the trend line whenever a support area was broken, thereby generating a new line while remaining rigorously faithful to Sperandeo's rule. How does that old song go "Call me anal-retentive, call me ...."?

Db clearly agrees with your point and while I am not trying to form a heretical sect and well appreciate that he (db) has talked about how he draws these lines, I thought it was interesting to see how price played out when in the vicinity of these lines. It is very true that any interaction between price and the line may be more apparent than real but trend lines do appear to be useful for at the least, alerting one to a possible change in the mood of price.

ljey

Yes, I've read the Trader Vic books too In a downtrend, I believe the right procedure is to draw a line from the highest high to the lowest minor high that preceeds the lowest low of all. As you say indeed, without passing through any other lines, extending that line further into the future. I must say, I've already had trouble with this myself, as sometimes "external" events (eg news) can cause this line to fail for a brief time period.

So in that aspect, I think TL3 is correct if you consider only the second half of the chart or as you're saying shifting the starting point. It's all in perspective of what you're looking at of course.

Personally (and that's the opinion of a newbie so take this with the necessary amount of salt), I think S/R lines offer a better view of price action than trendlines. I would use trendlines to indicate where price is going (up, down, nowhere), not to give me possible entry points, but there are probably many out there who do use it for that kind of purpose. I don't know how it works out, any thoughts on the matter?

Btw, care to put up a chart to see where TL1 goes?

 

[ljyoung]

I think S/R lines offer a better view of price action than trendlines. I would use trendlines to indicate where price is going (up, down, nowhere), not to give me possible entry points, but there are probably many out there who do use it for that kind of purpose. I don't know how it works out, any thoughts on the matter?

Btw, care to put up a chart to see where TL1 goes?

I agree with you. It's the S/R that matters and T/L's are signposts not defined entry/exit points.
As for what happened with TL1, check it out. (I just changed my charting software so am using Prophet-Basic and hence no chart note capability).

ljey

 

[firewalker99]

I agree with you. It's the S/R that matters and T/L's are signposts not defined entry/exit points.
As for what happened with TL1, check it out. (I just changed my charting software so am using Prophet-Basic and hence no chart note capability).

ljey

Hi, could you tell me on what makes you draw TL1? After TL3 is broken, price seems to find support somewhere between 129.7 and 129.75. It's not making a lower low, instead buyers are able to push price way back to around 129.87, which is higher than the LSL. I've attached a personal view of the chart, blue lines are S/R, red ones are TL.

For me, but again this is only what I'm reading in it (and I have only limited experience on the matter), this constitutes a break of the trend. Price discontinues moving down. If you then consider my blue lines, which can be drawn before price even is ranging between. The upper line basically connects two points: the level where price paused it's run down the hill from 1040-1100 and the last attempt to go higher at 1115. So when price again touches this level at 1230 you would see resistance confirmed as the line would already be on your chart in realtime.

The TL1 you've drawn, would mean the downtrend is still in progress, while at 0200 that high does not preceed a new low. So I would see no reason in particular to assume price will still move further down, on the contrary the validity of that line is already denied by price breaking out above the resistance level 129.87. At this point, I'd be looking to draw a trendline upwards. At around 0245 you have the retest, R becomes S and price is on it's way up.

As this thread is about S/R I'm wondering if dbphoenix perhaps rather have us discuss this someplace else, as it's more to do about trendlines than s/r, but as far as I'm concerned, not less interesting for that matter...

 

[ljyoung]

Hi fw,
As I said above, I use Sperandeo's protocol when drawing a trendline and adhere to it rigorously. It serves me well in that it eliminates a subjective variable with respect to my having to decide when a price bar is not "acceptable", which would then allow me to march through it instead of just touching it on my way to wherever. It's an example of empiricism in technical analysis with the only justification for using it being that "it works" (for me). Is it right? I don't know, but if I stick to using it and nothing else, then if it stops working at least I will know it wasn't because I diddled with the protocol. Again a T/L for me is a signpost and seeing as I don't know where I'm going I may find something else a little further down the road which makes me believe that what I thought that trendline violation was telling me, was not correct.

My analysis of the SPY chart is somewhat different from yours (not better than yours - just different) and the way I look at it is that between 11AM and 230PM price hasn't gone anywhere. Except for that dip down from 1130 to 1155, it's been stuck between 129.70 and 129.90. I agree that it has been making higher lows but on the high end of the range the action has been a little erratic. It finally breaks out at about 230, retests several times, sneaks past the mini-power spike and makes a new high for the afternoon session.

Did price "interact" with TL1 between 200 and 230? There's enough S/R in that region to say that once again the interaction is more apparent than real. The problem with trendlines in a more general sense is that if you just leave them on the chart, weeks or even years later price often runs into them again. Does that mean anything? Maybe and maybe not. The fact that some people (and/or their software systems) attach a lot of significance to T/L's may explain some of the "interaction" but it very rare that it supercedes S/R. The best way I have seen a T/L's utility demonstrated is with a "Trader Vic 1-2-3" play. How tradeable a play like this is depends on you and your trading time frame, risk tolerance, etc.

A final point which I'm sure you've heard before but which always bears repeating - beware of the potential for abuse when using the "retrospectroscope". This mind-boggling apparatus will allow you to see things you didn't even know were there, with pinpoint accuracy and amazing predictability. A poor man's chart replay is a piece of paper "a couple of hours wide" which you place on the righthand portion of your chart to put a governor on this probe.

ljey

 

[dbphoenix]

Sperandeo's protocol notwithstanding, the job of a trendline is to show trend. If price is nowhere near it, it's simply a line dangling in space, and the "1-2-3" becomes moot since price may have changed trend long before reaching an incorrectly-drawn trendline.

If one is in doubt regarding whether or not the TL is drawn correctly, plotting a simple MA -- which is a moving trendline -- or a regression line will give the trader a better idea of where price says the trend lies.

 

[ljyoung]

Sperandeo's protocol notwithstanding, the job of a trendline is to show trend. If price is nowhere near it, it's simply a line dangling in space, and the "1-2-3" becomes moot since price may have changed trend long before reaching an incorrectly-drawn trendline.

If one is in doubt regarding whether or not the TL is drawn correctly, plotting a simple MA -- which is a moving trendline -- or a regression line will give the trader a better idea of where price says the trend lies.

Agreed, db. TLs are, as you have pointed out elsewhere, entities through which our creativity (our inner warm and fuzzy self) can find expression and if we are not careful to the detriment of our wallets. Given their formulaic basis, MAs and linear regression lines are more closely associated with price and its direction than is a TL. If we stick to a Dave Landry view of TLs: "the blue line is pointing up (or down)" it is probably the safest (not necessarily the most profitable) use that can be made of them. When working in very short time frames, the "1-2-3s" and "crossovers in the ether" can be extremely hazardous to your financial health and should be avoided by those who cannot manage their risk appropriately.

In my mind, anything you can use to hone your edge is acceptable. If you test it and it works then go ahead and use it, always bearing in mind the Mamisean point-counterpoint (there is probably a better way to say it but basically too many "market tells" can suggest an entry point just prior to the end of the move).

 

[dbphoenix]

It's also worth noting that Sperandeo's TL crosses through several bars in fig. 7.10. So, as usual, it pays not to be too literal.

 

[ljyoung]

It's also worth noting that Sperandeo's TL crosses through several bars in fig. 7.10. So, as usual, it pays not to be too literal.

If we are having fun, then it is similarly worth noting that Figures 7.1 and 7.3, show no evidence of the slopiness apparent in the, no doubt, hastily drawn (or more probably poorly transcribed)trendline of Figure 7.10 and what's more are given as paradigmatic examples of his method (point 2, in the second paragraph of page 71).

All praise and honor to thee, Victor Sperandeo. For the supreme test of devotion I suggest you read "Crashmaker" a two volume, 1572 page novel but really an opus magnus of his thoughts on everything from fiat currencies to Wall Street fat cats to "How to cause a stockmarket crash in several easy steps".

 

[dbphoenix]

I don't know that I'd call it "fun", just a caution against being too literal.

 

[ljyoung]

As an addendum, the ADR is a reliable target for the NQ, which generally comes within 10% of it on a daily basis, and is the primary reason why I don't trade anything else. If I don't get stopped out for some reason, I have no reason to stop myself out. Instead, I just wait for the ADR.

None of which is a plug for the NQ. Just an explanation of why the ADR is posted on these charts.

Db
In the calculation of an ADR for a stock or ETF, would you include any opening gap (from the immediately prior intraday close) or not? This consideration would only be necessary if at the end of the day, the gap remained unfilled.
TIA
ljey

 

[dbphoenix]

I use a 10d avg of the high to low.

 

[firewalker99]

Sperandeo's protocol notwithstanding, the job of a trendline is to show trend. If price is nowhere near it, it's simply a line dangling in space, and the "1-2-3" becomes moot since price may have changed trend long before reaching an incorrectly-drawn trendline.

If one is in doubt regarding whether or not the TL is drawn correctly, plotting a simple MA -- which is a moving trendline -- or a regression line will give the trader a better idea of where price says the trend lies.

As no one has asked, I'll assume this is clear for everybody else. Unfortunately not for me... so could you explain please what the different colored lines indicate and what meaning you attribute to them?

 

[dbphoenix]

As no one has asked, I'll assume this is clear for everybody else. Unfortunately not for me... so could you explain please what the different colored lines indicate and what meaning you attribute to them?

The green, blue, and red lines are regression lines. The green is drawn using the green dots, etc. The pink lines are supply lines.

 

[firewalker99]

analysis of S/R in real time?

I'd like to invite anyone who wants to give his or her best shot at analyzing S/R in real time to also have a look at the charts I put up in my journal, where I've find it next to impossible to find something useful S/R to even remotely trade in...

NQ charts: http://www.trade2win.com/boards/showpost.php?p=273477&postcount=447
DAX charts: http://www.trade2win.com/boards/showpost.php?p=273460&postcount=444

 

[dbphoenix]

I'd like to invite anyone who wants to give his or her best shot at analyzing S/R in real time to also have a look at the charts I put up in my journal, where I've find it next to impossible to find something useful S/R to even remotely trade in...

NQ charts: http://www.trade2win.com/boards/showpost.php?p=273477&postcount=447
DAX charts: http://www.trade2win.com/boards/showpost.php?p=273460&postcount=444

Unfortunately, "real time" is at this point no longer possible. But anyone interested in providing a hindsight analysis is welcome to do so.

 

[firewalker99]

Unfortunately, "real time" is at this point no longer possible. But anyone interested in providing a hindsight analysis is welcome to do so.

Pseudo-realtime then, if he or she puts a paper on the right hand side and shifts it bar to bar, he can experience a similar "feeling". Sure it's not the same. Nothing is the same as in real time, but it's the best I can do do as I as of yet can't predict whether tomorrow will provide another ranging day.

And anybody who has a replay function can also choose to play those days. The dates are noted in the filenames of the attachments.

 

[dbphoenix]

Occasionally, pre-calculated "pivot points" will offer potential S/R when previous lines or zones of S/R are somewhat less than firm, if they are to be found at all. Yesterday provided a good example (note to the anti-hindsight contingent: yes, this is a hindsight posting, but the PDH and the pivot points are available to everyone at the end of the previous trading day).

Here, there were pre-plotted S/R at 1585 and 1590. However, there was nothing above 1590 until one got past 1600. That's quite a distance. But the pre-calculated "pivot point" happened to coincide with S/R at 1585 and the pre-calculated R1 happened to coincide with the PDH. So why not plot the pre-calculated R2 and see what happens?

It overshot a little, but it did seem to hold. And even if taking that particular short were too risky, taking the break of the opening low or taking the lower high would also be worth a few points.

The afternoon test of the OH appears to be just that rather than a test of R2. But, either way, that is also worth a few points.

This is not to suggest that pre-calculated pivot points provide consistent S/R, but they are another available tool, even if they aren't used that often (the Allen wrench of S/R tools).

 

[dbphoenix]

Making high probability trades:

Use a variety of bar intervals and timeframes to confirm and time trades.

Trade in the direction of the primary trend.

Have a predetermined exit strategy.

Plan trades before the market opens.

Have a reason for every trade.

Predetermine the amount of risk you’re willing to assume.

Stay focused.

Maintain discipline.

Be patient and wait for the best opportunities.

Incorporate the inevitability of loss into your system.

Don’t chase trades. Any entry subsequent to the best entry has a lower probability of success.

Adjust your risk according to the probability of success of the trade.

Trade more heavily when the odds are in your favor.

(I just saved you $40 . . . )

 

[rainman2]

How do you define primary trend?

If the trend is down on let's say the hourly but is up on the 5min and I trade the 5 min, should I consider any long entries counter-trend, or should trend be defined by the time frame you trade from, or does it become a position size issue?