Search | User Listing Vidsite | Calendars | Quotes
Home Page ->  Custom TradeTight Routines -> Tools In Development -> View ThreadLogon (or Register, or Join TradeTight)

You are logged in as a limited-access Guest.To join TradeTight, first read the info in the Organization & Content room, then click the link above. 

Proverbs 16:9 (ESV) ... The heart of a man plans his way, but the Lord establishes his steps.


Profit-Grab Strategy Exits
Jump to page : 1
Now viewing page 1 [50 msgs/pg]
Jump to room :
JimDean
Posted 9/26/2017 4:42 PM (#8983)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
I could use some opinions about an Exit method that I'm putting together for release fairly soon:

Background: I've always believed that well-designed exit logic is more important to longterm trading success, than the quality of entries (tho Entries are important too!). I've probably spent more unpaid-development hours working on custom Exits than on anything else, for the past dozen or more years. And it's coming time now, with the anticipated release of TradeTrak (described in the MTV Section), to start providing these Exit methods in DLL's that folks can buy.

This post is an illustration of just one of many Exit methods, refined over the years. Please understand that I'm not totally disclosing the math/logic behind it ... but if you look hard enough at the charts, and maybe play around a bit with some canned N indicators, you'll figure out what's going on, for the most part.

Please note that this exit-method would typically be used with one or more other methods that use Stop-Loss thinking (ie preventing pullbacks from becoming disasters). For some trades, this logic would never kick in ... or just partially scale out of the trade. This is what I call a "Profit-Grab" (as opposed to a "Loss-Stop").

The various arrows show actual price-points and bars that the exit-logic would tell the broker to take out a portion of the shares of the trade. I could choose to implement method C, method G, or a combo of the two (&/or some other variants). If I use the combo, each partial-exit's size would be smaller, since there potentially would be more of them. Also note that since no entries are shown, the additional qualifying rule that an adequate price-move would have had to occurred since entry *could* invalidate some of the arrows. The rule requires 2x Initial Stop-Gap as profit since Entry. If the initial stopgap was say 1.5-2.5xATR, then the required profit would have to be at least 3-5ATR before any of those arrows would be allowed to fire.

It's a MOO order, placed the night before based wholly on info from the earlier bar (does not depend on early-morning price action) ... therefore it's easy to implement. This kind of stop is most useful on "trend" type trading, but also can be used for "trading-range/swing" type trades. NOT for RTM strategies.

Btw, I did not "cherry pick" these charts ... I just grabbed five alphabetically-sequential symbols from the Nas100. Also ... the arrows identify ALL the potential exits ... even the ones that might occur earlier or later than you might ideally wish. The problem with many presentations of entry and exit logic are that they show examples of the "ideal" occurrences, but not "all" the occurrences. Don't forget the 3-5ATR profit-since-entry requirement, as you study them.

Thanks for your thoughts!


(ProfGrabs-EXPE.png)



(ProfGrabs-FAST.png)



(ProfGrabs-FB.png)



(ProfGrabs-FISV.png)



(ProfGrabs-FOX.png)




Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits


Attachments
Attachments ProfGrabs-EXPE.png (50KB - 0 downloads)
Attachments ProfGrabs-FAST.png (47KB - 0 downloads)
Attachments ProfGrabs-FB.png (49KB - 0 downloads)
Attachments ProfGrabs-FISV.png (53KB - 0 downloads)
Attachments ProfGrabs-FOX.png (52KB - 0 downloads)
Top of the page Bottom of the page
JimDean
Posted 9/26/2017 5:29 PM (#8984 - in reply to #8983)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
First reply had some questions, and a comment ...

1. Would your exits include besides ATR, also % and pts. ? i.e., accommodating for forex and futures globex trades. Thanks.

2. Method G (yellow - fuzzy) sounds scary - not knowing its exit personality, as able to somewhat figure out and prefer combo and thereafter method C.


Re #1 ... this is not a "conventional Nirvana" exit, regarding its parameters ... it's more of an "expert system" kind of exit. There definitely will be some user-options, but distinctions such as ATR/%/Pts just don't matter here ... the internal calculations take care of that kind of thing. Currently, my code has eight optional user inputs, half of which are "subjective" rather than numerically specific. By "subjective", I mean a choice for an aspect of how the program works such as "small, medium, large". The other half have multiple choices but they are explicit integers from 0-2 or 2-6 or 0-6 or 1-3, rather than "infinite sliders". The purpose here is to limit the choices to ones that "make sense", and limit the permutations ... as it stands, there are approx 25k input permutations, so it's very flexible ... but due to the expert nature of the inputs and code, it's easy to use. The "out of the box" default values work just fine for many/most cases. I may well decide to cut the #inputs back to just 2 or 3, by applying a "guru" layer to the expert logic.

Re #2 ... Method G is not really much different than C ... I'm not sure why you said it "sounds" scary. The purpose of my post was to show several charts, with all the related potential exit points marked. Afaik, either/both methods produce reasonable profit-grab opportunities ... ie at/near the top extension of a run ... or just before a consolidation. By contrast, a typical "Loss-Cut" stop waits for something "bad" to happen (ie some degree of retracement from the Max Profitable Excursion) before exiting. The idea behind a Profit-Grab is to take some of your shares out at what (quite often) turns out to be an "ideal" time. Of course, if the trend continues, you can re-enter after the consolidation or retrace has run its course.

EVERYONE: maybe I didn't explain properly what I'm asking for here ... I'd like you to LOOK at the charts and see if you'd like to have partial-exits at the marked points. You really don't have to read the verbiage ... just examine the charts where the "pudding-proof" lies. Of course, other questions are welcome too! Thanks for your time!


Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
SalimHira
Posted 9/27/2017 7:32 AM (#8985 - in reply to #8983)
Subject: Profit-Grab Strategy Exits



Veteran

Posts: 142
10025
Location:
USA: MD, Columbia
Hi Jim:

Are all the Exit arrows shown profitable from their Entries ? Or, there are some arrows that may have been stopped out.

My question, would the exit work as a broker stop too, based on R:R from its entry ? If so, than how would the broker stop be determined ? Thanks.

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
JimDean
Posted 9/27/2017 8:19 AM (#8986 - in reply to #8985)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
Hmmm. I guess that these "trades" are harder to "picture" than I had anticipated.

I've added pink and lime circles to the snaps, showing *hypothetical* entry points, to help y'all picture the trades.

The MAIN thing to keep in mind is that these charts and "trades" have NOT been "cherry-picked". Their purpose is to show how the logic might provide multiple partial exits for trades, *nearly all* of which are profitable and *nearly all* of which do not wait for a pullback to occur that causes you to lose profits.

The other IMPORTANT thing, different than the "normal" paradigm, is that these represent PARTIAL exits, where only 10-50% of the position's shares are closed, to "grab" profits while the price is in a highly positive but (usually) statistically unsustainable spot.

Barring "cherry-picked" examples, or highly-curve-fit over-tuned routines, it's impossible for calc's to have a crystal ball, foreseeing that a "reasonably profitable" point actually will turn out to be a stairstep along an extended profitable trend. The idea behind profit grabs is to capture gains just *before* looks *reasonably likely* that price might stall or retrace.

The approach here is to (potentially) take *multiple* partial profit-exits ... that means if you have a fairly long run-up and a few exits occur mid-run, you're still in the game ... and can if you wish scale back in at a midstream retracement.

Regarding the question of being stopped out before reaching those noted points ... that depends on the nature of your other exit rules. I will be releasing additional information about sophisticated "loss-cut" (scaling out) and "loss-stop" tools as well ... for now, I'll just ask you to presume that *most* of the hypothetical trades you see on these randomly-selected snapshots *don't* get stopped out first ... or maybe they are "trimmed back" by a partial loss-cut, but are still in play.

THE QUESTION REMAINS:
Do y'all consider most of these partial-profit-grab exit points to be generally advantageous, or not?


(ProfGrabs-EXPE.png)



(ProfGrabs-FAST.png)



(ProfGrabs-FB.png)



(ProfGrabs-FISV.png)



(ProfGrabs-FOX.png)




Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits


Attachments
Attachments ProfGrabs-EXPE.png (60KB - 0 downloads)
Attachments ProfGrabs-FAST.png (56KB - 0 downloads)
Attachments ProfGrabs-FB.png (57KB - 0 downloads)
Attachments ProfGrabs-FISV.png (63KB - 0 downloads)
Attachments ProfGrabs-FOX.png (63KB - 0 downloads)
Top of the page Bottom of the page
JimDean
Posted 9/27/2017 8:51 AM (#8987 - in reply to #8986)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
Re your question about a broker stop ... I take it you mean a Stop-Market order with a threshold price-setting, or a MOO/BOO order, as opposed to a plain Market order.

So ... re Broker stops. Nirvana's Trade Plans and internal coding options offer *very limited* flexibility to utilize the scores of different order types that brokers typically support. The *ONLY* functional "virtual-price-level" order that OLang+OT/VT currently supports, despite many pleas for expansion, is a Stop-Market order (but server-based platforms OmniVest et al do not, at this time). *All* N platforms support MOO and Market orders ... the snapshots I posted presume MOO orders since they are simple and applicable universally to N tools. Nonetheless, it's a good question, and offers even better performance possibilities ...

Stop-Market orders have two important limitations:
1. They fire when ANY portion of the bar touches or crosses them .. ie High's for Shorts and Low's for Longs.
2. They ONLY fire when the touch/cross is in a LOSS direction ... ie Long-trade Low crosses *below* the preset-price-level.

On account of this, SM orders are usually sort of "counter-productive" for simplistic "exit if the profitable-extreme gets to this level" Profit-grabs, which work on the opposite end of the bar (Highs for Longs and Lows for Shorts). Furthermore, it is quite common for an extended tail (wick or shadow) to "reach out and touch" a "visually obvious" stop-level such as a prior pivot or prior bar's extreme or commonly-used MA or Fibonacci level, etc. This means that traders who use SM orders need to anticipate "frustrating" exits fairly often ... where the main body of the bar is in a "stay in the trade" status, but the "nasty" tail stretched out and bit them.

SO ... to answer your question ... YES, this can be set up using SM orders instead of Market orders. The sequence would be:

1. a bar in the trade would move into "profit-grab" territory, and that would result in a SM level being sent to the broker for the NEXT bar, where the level would presumably be somewhere between the "best" extended-price of the prior bar, and the "minimum" profit-grab threshold price (projected for the next bar). This is straightforward and do-able.

2. then the next bar hits ... and here is where the problem potentially occurs. If the next bar's open gaps against the trade significantly (which is a reasonably-likely situation when a true statistically-unlikely overextension has previously occurred), then the SM will fire but there's a good chance that the "sweet" profit-grab opportunity to be missed ... HOWEVER ...

3. if the next bar actually *extends* the price further ... ie the profit grab signal turned out to be premature, and the run continued apace ... and if the next bar's "bad side extreme" (ie Low in a Long trade) doesn't test the prior range, then in that case, the SM level and partial-exit set in #1 will be ignored (a good thing in this scenario) ... and usually, since things are moving fast and furious ... that next bar will itself fire off a new, *more* profitable "grab" signal.

IMHO this is an EXCELLENT suggestion. I had originally just "defaulted" my thinking to using MOO orders on the next-bar, since N seems to have gravitated so strongly in that direction (driven by OmniVest et al limitations). So, please consider that the release-code will be set up to work with three different *user-selectable* Trade Plan options:

1. Virtual-level Broker-SM orders for the next bar, as described above
2. Market-On-Open (aka BOO for intraday) Broker orders for the next bar, as shown on the charts
3. Market Orders that fire as soon as price on the "#1" bar above goes beyond the profit-grab threshold.

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
KenWilsdon
Posted 9/27/2017 8:52 AM (#8988 - in reply to #8983)
Subject: Profit-Grab Strategy Exits



Spectator

Posts: 7
0
Location:
: ,
I, for one, take partial profits on every profitable trade all the time, and a system such as this would be very helpful.

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
JimDean
Posted 9/27/2017 9:00 AM (#8989 - in reply to #8988)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
Thanks for your comment, Ken.

I'm curious about several things, in regard to how you typically have been managing your trades, in light of all the descriptions I've given in prior posts:

1. Do you use same-bar Market orders, or do you use next-bar SM orders or MOO orders? Is there a strong reason you have for using one versus the others, based on your experience?

2. If you were to imagine the trades occurring as shown on the latest set of snapshots (hypothetical entries and flagged partial-profit exit points), are those noted exit points about the same, better, or worse than what you've previously been using?

3. What is the typical percentage of your position-size that you take out, on a "profit-grab"? Do you vary the percentage, and if so, what criteria do you use to determine it?

Your feedback about these things would be very helpful!

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
KenWilsdon
Posted 9/27/2017 10:25 AM (#8990 - in reply to #8983)
Subject: Profit-Grab Strategy Exits



Spectator

Posts: 7
0
Location:
: ,
Jim,

I will first describe my basic methodology, and then answer your questions sequentially. Hopefully this will make things clearer for you.

I use a percentage of total equity for my trades. for entries, I look at trend, cycle lows/highs, momentum, nearby support/resistance, relative momentum to market, volume, and finally momentum on the next higher time frame.

I "ALWAYS" enter upon a signal with a Stop Limit order one tick above the previous bar, with a limit of 5 ticks/pips/cents above the stop (I don't want to buy if the market has run away from my entry - there are enough good candidates, and it screws up my first take profit stop position, next paragraph). I use this stop above the previous bar as confirmation of momentum in my direction.

Upon entry, I place a fixed stop (market) 1 tick below the most recent cycle low on that time frame. Then I place a limit order for 1/2 of my order above entry on about a 1:1 ratio from fixed stop to entry, at just below/above obvious support/resistance, so that if the stock goes up and hits this first profit target, I have reduced my risk in the position to near zero, if it reverts to the fixed stop loss. This also helps my win/loss ratio. A "profit grab" trade.

With the remaining portion, I take 1/2 at the point where a cycle high is forming or where momentum is failing. I generally place a trailing stop on this portion at the prior Heikin-Ashi bar's low (with a stop market), but if the current Heikin-Ashi bar is the opposite color (e.g., red instead of green), I place it directly under the current bar. This allows for small Hiccups, but not reversals.

It is these first two trades where I see your exits as beneficial to my "system".

The remaining 1/4 I trail 1 tick behind the most recently formed pivot high/low, with a stop market order. This is my longer term portion that will go until the trend is turning, and I get stopped out.

So in answer to your questions:

1. I NEVER use MOO orders. Only stop market, limit, or stop limit orders. MOO orders do not give me confirmation that the security is moving in my direction at entry. I am at the mercy of the market makers on my MOO orders, and am not in control. I could be entering much farther away from my entry zone than I am comfortable doing, thus screwing up my risk/reward. Conversely, I WOULD consider MOC orders, as this is the time when professional money is setting final value for a security, and they rather than the market makers are in control. They are just finickier to place at the right time without software, and I have to be around my computer at that time if done manually.

2. Those points on the charts are almost identical to the places where I take my first and second TP trades.

3. Profit grab is 1/2 on first trade, one half of the remainder on second trade, remainder is long term. This does not vary for me.

With the above system, if I start with 4% of equity, I have 1% at the end for longer moves. This gives me diversification, plenty of profits, and a good win/loss ratio with a decent risk/reward for the total position. As I am always taking profits, I always have money available to take new trades that are forming. While it may not be the UTM, it works for me.

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
RossKovacs
Posted 9/27/2017 10:36 AM (#8991 - in reply to #8983)
Subject: Profit-Grab Strategy Exits



Veteran

Posts: 124
100
Location:
USA: GA, Snellville
The Profit-Grab Strategy Exits is an interesting concept. You clearly state "It's a MOO order, placed the night before based wholly on info from the earlier bar (does not depend on early-morning price action) ... therefore it's easy to implement."
1. Does "easy to implement" mean easy to program or easy for a trader to execute (or both)?
2. Could you point us to previous threads/posts where you have reviewed OT's various exit order options? I tried searching for them, but apparently could not find the best search terms to get the threads/posts.

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
JimDean
Posted 9/27/2017 10:49 AM (#8992 - in reply to #8990)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
Ken:

Thanks *very* much for explaining your approach ... it sounds really well thought-out to me, and my guess is that you probably have a pretty high "hit rate" (re making some profit) using that approach. Sounds like a great "low-ulcer-index" method.

I like the one-tick Limit-order entry aspects. Although OT/OV don't allow Limit levels to be defined "virtually" by OLang code ... the Trade Plan can readily be set up with a Limit one tick beyond a known price point ... so it's practical for a mechanical strategy.

One thing I've been debating is whether to code up the scaling-in/out based on fixed sizes for each step, or using a declining amount such as you've described, or using an increasing amount which I've heard from some other traders that they prefer. The simplest to set up in a Trade Plan is the fixed-sizes approach, if more than just one or two extra entry/exit steps are potentially in play ... that is, a trade during a really long trend-run that never has any extreme retracements might have a lot of points at which the trader would scale out partially, then scale in again (maybe more), then scale out and so on. Building an OT/VT Trade Plan to handle this eventuality is a big chore if all the sizes for each step are the same ... it becomes a nightmare (ie hundreds of TP-steps) if the individual in/out sizes might come in different combinations depending on how price action progresses.

So, with that in mind, I'll probably start out with a fixed size for each in/out step AFTER the initial trade ... the fixed-size would be user-defined as X-percent of that initial entry. Later, if demand appears, I can adapt for different patterns ... but in order to keep TP's reasonable, the decreasing/increasing options probably will force full exits and subsequent re-entries along the way, for a really extended trend.

Thanks again for your feedback ... very helpful!

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
JimDean
Posted 9/27/2017 11:10 AM (#8993 - in reply to #8991)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
Hi, Ross:

By easy to implement, my primary thought was its universality for all the N platforms ... currently OVest/OFunds only allow MOO and market orders ... and the market orders only hit when N happens to run a recalc on their servers. So, MOO is their standard. Also, SM virtual-level orders take a little more decision-making (by programmer, or via an extra user-input) on the programming side ... a choice has to be made about what price to use for the order. Having said that ... if the trading is being done via OT or VT, then either approach is practical to implement, and my prior response to Salim's question re using SM orders shows that they offer some real benefits.

Re other posts ... I don't think I've ever written "a guide to OT order types". There should be some explanation in OT Pro Help. The comments I've made, here &/or on the N forums, primarily relate to the bugs/issues with using virtual-level Limit orders for entry. It's been cussed and discussed a lot over the years. For some reason, tho N is aware, they've never chosen to fix it. I call it a "fix" since the TP options allow for a user to *try* to make it work ... but the OLang keywords don't have a way to define a "LimitLevel" or "EntryLevel" or whatever. Sigh.

I don't have time now to search the forum but I'd start with the word "Limit" as a search term - again, many of my posts have been on the OT forum since I was trying to get them to fix it. I might have written some posts on scaling in and out (try "scaling|scale" search) ... and I seem to recall having posted some scale-in/out Trade Plan examples. I hope that helps.



Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
JimDean
Posted 9/29/2017 2:59 PM (#8994 - in reply to #8984)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
Followup to "question 1" mentioned previously, where I was asked if the routine would allow for stoploss-gaps to be optionally defined as ATR-multiples, Price-percentages, or raw Price-deltas. I gave a general reply but since then I've come up with a nifty method to not only provide all three modes, but also allow implementation of some logic from my "SmartSize" algorithm ... all this with just two inputs.

This input method implements something I believe is the *best* approach for stock trading stops ... to specify *both* an ATR-mult and also a Price-Pct, either of which would be desirable/acceptable in appropriate circumstances ... then the program calc's both of them, to determine which is the *most* restrictive for a given symbol on a given bar, and uses that most-restrictive gap-size. This helps assure reasonable stoploss-gap sizes for a Focus List with widely varying symbol prices (ie $5 vs $500), and widely varying volatilities (which can change based on symbol &/or current window of time).

The two input-parameters will be:
FxdLsAtrMlt10ths (slider=5-25,def=15) ... to define the fixed-loss full-exit gap as a mult of ATR/10 (ie "15" input means 1.5x)
' if negative, interpret this input as a price-delta multiplier (using next-input as pwr-of-ten)
FxdLsPrcPct10ths (slider=10-50,def=30) ... to define fixed-loss full-exit gap as a mult of Price/1000 (ie "30" input means 3%)
' if AtrMult negative, use this as price-delta pwr-of-ten: AtrMul=-15 and PrcPct=-3 => $0.015)

When both inputs are nonzero and positive, then the the stop-level is set to be the "closest to Entry" between the ATR-mult eval and the Prc-Pct eval. If only one input is positive, the routine uses just that method (ATRmult or PrcPct). Also note - as with all other OT routines, you can type in whatever # you want, even if it's outside of the slider range. The range is set up to help guide the user to reasonable or recommended values.

BUT, when the first input is negative, both inputs are interpreted differently ... the first input is a "scalar" multiplier, and the second is the power-of-ten exponent (positive or negative) to use with the scalar ... thus allowing very small numbers (if exponent is negative) for forex (ie 0.000002) or tick-based (ie 0.015) trades. Currently OT/VT input interfaces only show two decimal places (even if you type in more) ... this solves that native input-parameter display problem as well. Furthermore, it makes the use of the "slider" more handy ... each increment is a tenth, not a hundredth.

Not sure anyone really cares, but I think this will be a good standard to use, moving forward.

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
JimDean
Posted 10/5/2017 12:49 PM (#8995 - in reply to #8994)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
Further followup enhancements to the determination of initial Fixed-Loss stops ...

Prior discussions explained a two-input means of providing for:
ATR-mult gap stop, OR
Pct-price gap stop, OR
Define both above and use the minimum of the two, OR
Price-points gap stop

I've decided that providing a really powerful and flexible solution to this is worthwhile, so it can be used in multiple stop routines. So, I've taken it a couple of steps further, in addition to all the options above ... it only required one additional input.

The new input defines how many lookback-bars should be used to determine a prior "support" level. There are *many* ways of defining a support level ... I've chosen a "safe" middle-of-the-road approach that isn't too complex to code or explain. For a Long trade, the support level is: ( LLV(Open) + LLV(Close) + 2*LLV(Low) )/4, for the input lookback periods. This approach provides roughly equal bias for the Body-edges and the Extreme-prices. Same for Short trades, but HHV and High instead of Low. By adding this option, we can take the recent "absolute" price action into account, either in conjunction with ATR &/or PctPrc, or independently. This actually adds four more permutations to the flexibility, with just one new input (total = 8)

Finally, I've decided to allow the user to "force" any one or more of the three categories (ATRmult, PctPrc, Support) to be used even if some other nonzero input-value defines a "tighter" gap. That is, if the ATRmult is input as a *negative* value, then its gap would be the smallest possible, even if PctPrice &/or Support might be tighter for a given symbol on a given bar. If more than one input is flagged this way (ie negative), then the *loosest* (ie max) gap of the negative-flagged values is used.

This provides a lot more permutations, since zero, positive or negative values can be combined as inputs in many ways ... I'm not going to try to count them ... probably at least a couple dozen.

I am hoping that this flexibility covers enough ground that it would satisfy any kind of need, yet it's easy enough to understand so that it can be used in its simplest options without any head-scratching required. Comments and questions welcome.

Oh yeah ... one more thing ...

I'm also setting up a "Guru" input that can be used instead of specifying all three of these ... it will have a four choices: Small, Medium, Large, Huge. The other inputs will be varied automatically in a sensible way (if they are present at all, or if they are zero). This is for the folks who want the benefit of the method but don't want to hassle with lots of inputs. :~)

Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
JimDean
Posted 10/26/2017 7:22 PM (#9019 - in reply to #8995)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville
Here is a little taste of what the completed product's output will look like. This particular chart uses Default input values (except for one) and I've deliberately picked a Symbol (ALGN) with an entry date that results in an incredibly long and profitable trade. I've set it for Market Orders, so the various exits and addons occur on the same bar that the condition is found. This kind of long run is very rare, but it serves as a nice "strawman". The one non-default value is set so that Add-Ons are allowed further into the trade than the Default woud normally allow ... to show for this very non-ordinary trade (which is still going on!), how precisely the logic picks perfect scaling-in points.

Since the TT forum restricts the size of displayable snaps to 800 pixels wide, the small PNG-file pictures you see below are much too zoomed-out to clearly resolve. So, I'm also attaching large (1920x1080) JPG files of the same thing, for your viewing pleasure. Two versions of each ... one with just the "dots" (and some small hashmarks to highlight them), and the other with the dots plus all the relevant explanatory lines that are very informative once you've learned about them. For now: Big Yellow Dot with gold "tails" appear for initial Entry and final Exit (there is no final exit for this trade, yet). Silver dots with brown "tails" underneath are AddOns; White dots with small white "tails" beneath are Partial Exits (mostly Profit-Grabs, but a few are "stagnant zone" Loss-Cuts). I'm not going to try to explain all the "diagnostic" lines on the busier chart ... this is just a "teaser" posting :~>

I am *not* yet ready to provide a lot of detailed explanations or answer any but the most general questions, so please be patient. I've deliberately blacked out the parameters-list at the top of the chart, because I don't want people to draw conclusions from it. Nutshell re parameters ... there are four "versions" of this routine offered. Three are "simple" subsets, and the fourth is a "power" version. Two of the simple versions implement different "subsets" of the logic; the third simple version implements the full logic but limits the number of parameters. All three of the Simple versions have approximately half the number of logical-input parameters that the Power version exposes.

I've built this much like MTV ... it uses Expert inputs with qualitative input sliders (small to large, few to many, tight to wide, etc) that provide detailed control of the various logic-patterns, and groups those Experts under three Guru sliders which automatically and intelligently "fill in" the Expert inputs that are not explicitly set by the user. This has the same incredibly flexible structure that MTV does, yet makes it very easy to pick up on how to use it just by "playing with it".

The four versions will each be offered in two forms ... for the discretionary trader, an Indicator version allows the user to pinpoint the Signal bar for the trade via a nifty input-slider (or by typing in a YyMmDd date); for the mechanical trader, the TradePlan version will incorporate this routine as a "Stop" into a *very* complex TradePlan (Barry's comment was that I am doing things that never have been done before with the TP). The TradePlan version can be made part of a fully-automated Strategy, or can be kicked off by a discretionary manual Order placed via OT's green/red triangles. Both next-bar MOO and same-bar Mkt orders are fully supported (same routine, different TP's) , and I'm likely to also offer another TP version that uses StopMarket and maybe even Limit orders.

The biggest deal about this routine, in terms of "going where no Nirvanian has gone before", is that it implements very sophisticated Scaling-Out and Scaling-In rules (ie Partial Exits and in-trade Add-Ons). The user has very comprehensive control over all this.

Although there is a lot of debate about the efficacy of Partial-Exits (and entries), I believe in this simple philosophy: "You don't know what's coming!" So, when in a profitable trade that happens to reach what appears to be a good point to take out some winnings ... either due to your personal desired Reward/Risk ratio or due to a statistically-unsustainable profitable excursion during the trade ... then, DO IT ... but don't take out a huge chunk (or close the trade) unless you have good reason to believe it's about to reverse dramatically (remember the simple philosophy, though). Also, I believe that when a trade "stagnates" (often after one of those "sweet spots"), then it's a good idea to take a bit off the table (since the floor might be getting ready to drop out). The key to taking partial exits (either ProfitGrabs or LossCuts) is to ALSO provide for scaling back in via subsequent Add-Ons, while the trade is still live, and when you find a sweet spot after a retrace has started to recover. *THIS ROUTINE DOES ALL THAT*, and more.

Will it be usable in OmniVest? At this point, no ... because of an arbitrary decision by Nirvana not to accept DLL's as a part of an Elite Trader upload. However, if the value of DLL's such as this one and MTV and others that I'll be offering in the next several months "pan out" in OT/VT, then possibly Nirvana will reconsider that policy ... on the basis of performance improvements the tools offer, and "pointed" requests from users. Architecturally, there should be no reason why OV couldn't accept DLL's in the ET upload ... after all, the various N plugins are DLL's.

(ProfGrab JustDots Small.png)



(ProfGrab Dots+Lines Small.png)




Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits


Attachments
Attachments ProfGrab JustDots Small.png (16KB - 1 downloads)
Attachments ProfGrab Dots+Lines Small.png (24KB - 1 downloads)
Attachments ProfGrab JustDots Big.jpg (234KB - 6 downloads)
Attachments ProfGrab Dots+Lines Big.jpg (306KB - 6 downloads)
Top of the page Bottom of the page
JimDean
Posted 10/27/2017 8:12 AM (#9020 - in reply to #9019)
Subject: Profit-Grab Strategy Exits



Owner/Admin

Posts: 3557
200010005002525
Location:
USA: GA, Lawrenceville

********** Special Beta-Testing Opportunity ***********


If any of you are confident that you would like to use the ProfitGrabs Trade-Management package that has been previewed in the prior post (http://tradetight.org/forums/thread-view.asp?tid=1406#M9019), and would like to be a beta tester over the next few weeks as I polish it up, receiving a 30% discount on the released product when it is complete (no discount on optional Subscription), then please email me. I need a few more people who are readily available via email &/or phone so that I can get timely feedback on pre-releases, and who are willing do some testing so that the release version will be stable and won't need updates.


Thread moved by JimDean on 10/5/2017 3:45 AM from General Discussion > Miscellaneous Discussions > Profit-Grab Strategy Exits

Top of the page Bottom of the page
Jump to page : 1
Now viewing page 1 [50 msgs/pg]
( E-mail a Link | Printer Version | Search Room )

Owner of site: Jim Dean -- Forum content is confidential, and may not be distributed without written permission.