Dear Db

[DbPhoenix 42]

For questions that I may not see in the Trading threads (since I don't read every entry in every post -- sorry).

Edited January 2, 2016 by DbPhoenix

[Schaefer 10]

Dear DB,

 

First, find a range: Okay, now that we're in uncharted all time high territories again, how do we go about trading the extremes? I mean, we know where the lows are, but what about the highs? Thank you.

 

Schaefer

 

That's what the SLA is for, which we switched to just before 1000 (see my notes). It seems like it's been weeks since the SLA was relevant for intraday (but actually only about a week). If you've forgotten how it works, tonite would be a good opportunity to brush up.

 

Thank you, DB, will do.

 

Schaefer

Edited March 3, 2015 by Schaefer

[Ged 10]

db, you mentioned that back testing can be done in an evening or two. It doesn't seem to work that way for me and I'm wondering if perhaps I'm going about it in the wrong way. You mentioned 50 to 100 samples as being the range you would consider.

 

Did you mean in an evening or two one item could be back tested 50 to 100 times, retrace entries on a BO for example?

 

Are you anticipating a static chart 'back test' or a replay back test of each item? I have assumed the latter, but I don't see how that could be done in an evening or two, complete with notes, etc.

 

All that you can share to elucidate best procedures will be greatly appreciated. (My apologies if you answered this before.)

 

Thanks much,

Ged

 

I didn't say backtesting could be done in an evening or two. Backtesting could take months. What I said was that one could test entry and exit triggers in a couple of days.

 

As to your other questions about backtesting, I have no idea what you're doing so I can't give a pertinent answer. If you're trading the SLA/AMT, no backtesting is necessary. Or forwardtesting, for that matter. It's all been done. However, if you can't or don't want to follow the rules but would rather come up with something of your own, then you'll have to start at the beginning with observation.

Edited December 9, 2016 by DbPhoenix

[Ged 10]

db said, "What I said was that one could test entry and exit triggers in a couple of days."

 

I mis-spoke, as the above is what I am referring to: entry and exit triggers. I must be approaching the test of these wrongly, as it seems to take me quite a while. I was doing replays and taking notes. Is viewing static charts sufficient for these?

Ged

 

Yes, static charts are sufficient to answer the question. If you know what a range is, what a breakout is, and what a retracement is, you can determine how far away from the trough or crest of the retracement you have to enter in order to determine the probability that the trade will be successful, i.e., how to avoid being too early and how to avoid being too late. These entries may change every day. They may change several times during the same session. But that's part of the challenge of reading price movement. It is also a large part of the reward.

 

All of which sounds very kumbayah. But if you're afraid to take the trade at all, no amount of counting ticks is going to make you unafraid. Your time would be better spent determining what it is you're afraid of and why. If you're not afraid and the trade looks good, just take it. If it works out, great. If it doesn't, scratch it and move on to the next.

Edited January 2, 2016 by DbPhoenix

[Ged 10]

[db said, "Yes, static charts are sufficient to answer the question..."]

 

Thank you db. Appreciate you taking the time.

Ged

[V_2008 10]

Dear, Db

 

How is it possible to use half way points in trading? Is it something like potential support or resistance? Or is it a level which gives us signs of increasing, deacreasing strength of opposite side? Or both? See picture.

As Wyckoff wrote: "Rallies and Reactions as Indicators. When, a stock rises 10 points, a normal reaction would be one-half, or 5 points. This does not mean that a stock must

react 5 points after an advance of 10. It means that the extent of its reaction, after a rally or an advance is checked, should be regarded as one of the indications of its technical strength or weakness. Strength is indicated by a smaller reaction than one-half.

Weakness is indicated by a reaction greater than one-half. That is to say, when a stock declines 10 points, a normal rally would be approximately one-half, or about 5 points. A smaller rally would indicate technical weakness and a greater rally than one-half would indicate technical strength."

 

So, I was using it like this: If I see reaction more then 50%, then I have probability that next rally will not make a new high or up move will be sluggish, as selling wave(-s) has increased. And I will wait, when this up wave will end, before I will try to sell.

 

Some days ago I have discovered that this 1/2 points can represent some sort of support/ resistance. See second picture.

 

If you're referring to Wyckoff's course, you are correct: it is a measure of strength or weakness. As far as trading off it, this would be a factor only if you weren't in the market already on one side or the other. If, for example, you were long and price turned against you, you'd already be out by the time price retraced half its advance. Whether you'd have gone short after you exited would be part of your plan. If price were to bounce at half, that might give you some indication that your short might be in trouble. Without knowing your plan, all I can do is provide generalities.

 

If you're not in the market at all when price penetrates or bounces off one of these levels, entering can be messy since the halfway level is often where traders seek equilibrium, and they can bounce around there for quite some time while you get eaten up. Keep in mind also that it isn't the penetration -- if it does so -- but whether or not price recovers from it. In fact, if price penetrates a level and rejects that penetration, that strengthens importance of the level that it penetrated. IOW, traders really don't want to go there.

 

As far as support and resistance go, yes, insofar as these are levels beyond which traders don't want to go. It doesn't matter whether or not you know the reasons for their rejections of certain levels, only that you acknowledge them and respect them. If you can go beyond that into using them, good for you.

 

Thanks, Db! Need some time to think over.

Edited March 4, 2015 by V_2008

[youngin 10]

DB- Is there a thread where you have talked more about trading the 60m chart?

 

The SLA/AMT pdf uses 60m bars.

Edited January 2, 2016 by DbPhoenix

[youngin 10]

What effect, if any, do you think Apple leaving for the Dow will have on the Naz?

 

No idea. It wouldn't have any effect on my trading plan.

 

You've talked about scaling out, but do you have any thoughts on pyramiding?

 

The most logical places to pyramid are the retracements.

 

Thanks Db.

Edited March 6, 2015 by youngin

[BabyT 10]

DbPhoenix.

 

Now I am practising SLA with single stock futures in my country,

by trade 1 contract with real money.

 

1. Could you please recommend some criterions that I should get before I add another one contract.

 

..............

 

My problem in SLA now is lacking discipline in entry and exit.

 

I often open position too quick and fear to hold position so I get out quickly too.

 

I think too much about profit and loss.

 

But I still try to correct it.

 

And any day that I can follow my plan. My day is often positive.

 

...............

Thank you very much.

 

Apologize with my poor English.

 

Well, first, nail down your understanding of the approach, then nail down your application of it. If you can't do both, worrying about size is a waste of time and effort.

 

Once you can do both, when to increase size will become obvious.

 

Edited January 2, 2016 by DbPhoenix

[youngin 10]

DB-

 

Why do you think there isn't more material on AMT?

 

Do you think hedge funders are using this type of methodology or are they using something completely different? I've read that Paul Tudor Jones is a big proponent of "reading the tape" and that for trading macro, it is absolutely essential. He also says that reading price action is a lost art and that much of the younger generation struggles with it because it requires them to just close their eyes and trust the price action instead of trying to come up with fifty different reasons why this or that stock should go up or down. Is Wyckoffian logic the 'secret' to these guys making billions?

 

Wyckoff was close, particularly with regard to turning points and the search for "equilibrium". But he never really put it together. Steidlmayer was the first as far as I know to assemble all of this into a logical construct in Market Logic. Unfortunately, like so many of the type, he got all tangled up in software (Market Profile) and made it next to impossible to implement any of it without spending tons of time and money, not only on software but books. Lots of books.

 

As to professional traders, you may be interested in this.

 

Which leads me into my next question... At what point have you run into fill issues day trading and trading off the 60m in the NQ? At some point size has to become an issue, I'm just curious to know this point. I'd like to know how large I can scale this strategy.

 

Nail it all down with one contract. Your worries about being filled with multiple contracts are way down the road. Note, however, that getting filled is a function of the entry order. If you use a stop limit entry, you will very likely not get filled. On the other hand, if you use a market order, no problem, though your entry may not be anywhere near where you wanted it to be.

 

As I've done repeatedly before, I'd like to thank you once again for opening my eyes to how markets truly operate.

 

youngin

Edited March 7, 2015 by DbPhoenix

[DaKine 10]

Hi db!

 

Can you explain the supply lines on this chart please? I dont believe I have seen them drawn in this fashion before. Why there as opposed to the swing high 4455?

 

Thanks!

 

 

You're reading it in hindsight, right to left. If read in real time, left to right, AMT is used to trade what preceded the breakout. The trend doesn't "officially" begin until you're out of the range.

 

One could, of course, enter the trade at the swing high you mention, but, again, this is hindsight. In real time, all this was taking place at a median, with lots of congestion. Most would wind up making multiple attempts to enter and getting chopped up in the process. The clean entry is provided by the breakout, though it may chap a few to "give up" the points that couldawouldashoulda been added by shorting at the very top.

 

Those who haven't read the material, much less studied it, will try to apply the SLA to ranges, though in all fairness it's possible that this effort might be a simple error. In any case, I've gone an extra mile to emphasize that the SLA is NOT to be used in ranges.

Edited March 9, 2015 by DbPhoenix

[BabyT 10]

Thank you very much. DbPhoenix

 

I will do it.

[youngin 10]

Do you day trade year round?

 

Not much after the beginning of Spring.

 

Follow up Q: Due to the warmer weather and wanting to go out and play in the yard or some other reason?

 

Trading is not my life. If there is something more interesting to do, I'd rather do the other thing.

 

What instruments do you trade on a longer interval?

 

Just the NQ.

 

Have you ever considered starting a fund? Why or why not?

 

No. Too much like a job.

 

Why do you only trade the NQ (on longer intervals)?

 

In order to choose a trading instrument, one must first decide what he wants from it, then characterize various markets until he finds one that meets his needs.

Edited March 10, 2015 by DbPhoenix

[georg7e 11]

Db, I am thinking about posting some charts. Kindly, tell me how you post your charts within the text and how you size them. Because they aren't too small and there is no need to open a separate window. I think most would agree, it is better, easier, nicer and more helpful to read text and see a chart on the same screen at the same time.

 

Thanks,

George

 

When you begin to respond, scroll down until you see Manage Attachments. Click that and upload it. Then once it's uploaded, right-click and select Copy Link Address.

 

Then place your cursor wherever you want to image to appear, select the image icon above the message window, paste the link address into it and click OK.

 

All of which is a giant pain in the ass. If you don't mind the troll problems at ET, posting there is a simple matter of copy and paste. No uploading involved.

 

As far as changing the size, I use Paint Shop Pro. Perhaps Paint has that option as well. Can't say.

Edited March 12, 2015 by DbPhoenix

[lajax 10]

Hi Db

 

 

For now I want to start my phase of testing the entries and exits, obviously I want to apply the SLA/AMT approach; however, I would like to know your opinion regarding which documentation would you recommend me to start this phase?

 

What do you mean by "documentation"?

 

I mean documents, articles or texts to take as reference, for example the sla/amt pdf, So the question is what others articles or texts I could read in order to start the new phase (entries and exits)?

 

At minimum, the Wyckoff Lite material in the stickies, esp Section 7, and you've already read the Developing a Plan pdf. That's really about it. There is such a thing as reading too much.

Edited March 12, 2015 by DbPhoenix

[piejackr 10]

Hi Db,

 

Is there any point at which one should expand an existing long term trend channel to encompass multiple tests of its bounds - especially the upper? My understanding is that the point of the trend channel is not to always be adjusting it to fit the bars perfectly, but am struggling somewhat with determining at what point, if any, an existing channel becomes too conservative.

 

Is it simply a matter of one's own comfort with the number of cautionary indicators perceived from the channel? Should an expansion only be considered if there is a change in the angle of ascent which might necessitate an entirely new trend channel to be established? Thank you in advance.

 

Doing this in real time is always a problem, but the "real time" we're talking about here is months, not days, so as long as one has the general direction right, he's okay.

 

When the bottoming process has ended and the market begins trending again, the initial trend isn't going to last long because the excitement can't be sustained. The trend will then settle into something that can last awhile. You can see here that the most stable channel is the red one (the colors themselves don't mean anything). But price then hugs the UL and never even returns to the median, eventually making it clear that it's taking a new direction. Is it necessary to pick up on this immediately? No, as long as one doesn't try to trade against the general direction. There will be plenty of opportunities for counter-trend trades when price has settled into a predictable channel, as this one did almost two years ago.

 

Edited March 20, 2015 by DbPhoenix

[DaKine 10]

What is "UL" and can we add it to the glossary?

 

Thanks.

 

Upper Limit

Edited March 21, 2015 by DbPhoenix

[Gringo 10]

What is "UL" and can we add it to the glossary?

 

Thanks.

 

Upper Limit

 

UL is fine but LL (lower limit) causes confusion with LL (lower low). Context becomes necessary to understand the comment.

 

Suggestions:

ULTC (upper limit of trend channel)

LLTC (lower limit of trend channel)

 

Gringo

[youngin 10]

What are your thoughts/ personal experiences with psychological round numbers? Obviously one has to wait to see if these round numbers provide the anticipated S/R. Apparently Livermore liked to use these, and I have noticed the NQ often respects the 00 levels as seen on the Hourly chart.

 

So what does the modern Wyckoff mastah have to say?

 

I really don't think about it much.

Edited March 24, 2015 by DbPhoenix

[Gringo 10]

What are your thoughts/ personal experiences with psychological round numbers? Obviously one has to wait to see if these round numbers provide the anticipated S/R. Apparently Livermore liked to use these, and I have noticed the NQ often respects the 00 levels as seen on the Hourly chart.

 

So what does the modern Wyckoff mastah have to say?

 

The whole thing is dependent on the ability to read and judge the supply/demand dynamic. Getting hung up on a round number doesn't really matter as much in my opinion. What you see as a round number on NQ isn't so round on QQQ or Nasdaq. The levels vary. Just look at potential support and resistance and judge the action from there. Anything that focuses on what should be usually makes us disconnect from what is. It's the what is that makes us money or not lose too much of it, not our hopes of what should be.

 

Gringo

[bmwhendrix 10]

Today for some reason, I'm thinking of The Monkeys "Last Train to Clarksville"

[eminiman414 10]

What would you consider the difference (in points) between a scratch, small win, and a loss?

 

The purpose of the scratch is to immediately get out of a trade that isn't going as expected. There may be a small gain. If there's a loss, it should be inconsequential.

 

The chief purpose of it is to enable a failing trader to regain control of his trading rather than leave it all up to the market. But by itself it's not nearly enough. One must also know exactly what he's doing. For example, I rarely see anyone doing sufficient prep. They therefore trade in a state of perpetual surprise.

Edited March 28, 2015 by DbPhoenix

[eminiman414 10]

What would you consider the difference (in points) between a scratch, small win, and a loss?

 

The purpose of the scratch is to immediately get out of a trade that isn't going as expected. There may be a small gain. If there's a loss, it should be inconsequential.

 

The chief purpose of it is to enable a failing trader to regain control of his trading rather than leave it all up to the market. But by itself it's not nearly enough. One must also know exactly what he's doing. For example, I rarely see anyone doing sufficient prep. They therefore trade in a state of perpetual surprise.

 

I sorta meant more in terms for data collection on a strategy. If the average MAE on all trades is 1.39, average MAE on winning trades is 3 ticks, and the average loss is 1.56 points, depending on what's considered a scratch vs a loss would change the "win rate" of the strategy. But if it's profitable and simple to trade does it even matter? If minus a tick is a "loss" then the win rate is about 33% but if +/- 2 points is a "scratch" the win rate climbs closer to 70%. Average win is 7.58 points. Largest loss 2.25 points largest win 40 points. So as long as this strategy is profitable does any of this even really matter?

 

Unless the trader is trying to impress somebody on a message board, a "win" or "loss" of a tick or so may be considered null. The purpose of maintaining these stats for the individual is to keep him on track. How he goes about doing so is entirely up to him.

Edited March 29, 2015 by DbPhoenix

[eminiman414 10]

Thanks for the earlier responses. Last question is what in your opinion would constitute a large enough sample size regarding testing? Should it be measured in days tested or entries? Either or what's a justifiable sample size?

 

Depends on what you're testing and how afraid you are of trading. What exactly are you trying to determine?

 

Great question. I guess it's if my method has a chance over the long haul and if it's "good enough" given the current sample size to go live with it. Results thus far are positive. I suppose though that the "testing" is really nothing more than gaining confidence in what I am doing. Again, even in writing this the answer is really up to me. If I am not fully confident in the approach keep testing it. If I am confident roll it out.

 

I've only done 30 replays with 137 entries (some scratch/re-entries). In my head it doesn't seem like enough but some trades the plan works great and others it doesn't but that's just the nature of the beast of course

 

If it's a question of your entire method, then I suggest a year, 250 trading days, or at least every other day. You need to know how it works over a complete annual cycle. If you're genuinely interested, test over a bull/bear cycle, one bull year and one bear year.

 

Glad you said that because that's what I was thinking. Thanks for the response.

Edited April 2, 2015 by eminiman414

[youngin 10]

Has daytrading or longer term (like 60m) trading made the bulk of your wealth?

 

No idea.

 

Not the answer I was expecting from someone as experienced as you.

 

For 60m trading, is a steady profit percentage (from quarter to quarter) of 85-90% good?

 

Yes, that would be good though unlikely.

 

Just saw I made a typo. I meant 75-90% ***

 

 

After backtesting my 60m strategy, the phrase "The big money is made in size, not points" keeps ringing in my head. Am I correct in remembering you're the one who said it?

 

It's difficult to make a living trading one contract.

Edited April 2, 2015 by youngin