Clemmo's Trading Odyssey

Jackanape

Instruments: Any, but start with the old chestnuts, EURUSD and XAUUSD
Timeframe: Trading on H1, maybe some fine tuning on M15. Scanning on D1. Phasing on W1. If this works you could and maybe should, try a lower trading timeframe.

'Phasing' shows I was still deep in the thrall of cycles mythos.

This system doesn’t worry about minima/maxima, as the trading rules basically take care of that. It finds the elbows (turning points) in bands and rides trends, even the tiny ones in ranging periods. It should be a ‘myopic’ system, to use the terms of economists because being able to predict future turns of any cycle isn’t crucial to its success.

Indicators: (use numbers that give a good band fit without being too ‘stringy’ (subjective) or leaving gaps between price peaks/troughs and the band walls.

1. TMAdCGmladen-arrowsMod a

1. Turn off the centerline colors, it’s a repainting siren liar
2. Then why use it?!

1. Don’t rely just on the red and blue dots. Sometimes they disappear or don’t show up for some reason. Look at the actual piercings. Maybe the line is coming down across the tops of some smaller peaks as price changes.
2. Tipu Renko

Optional:

1. Chaos Visual averages
2. Heiken Aishi candles (as per guttersnipe)

Process

1. Scan
2. Scan on W1/D1/ and look for price piercing (or approaching) a container band. Identify likely reversal points. Write them in a notepad/journal maybe. Watch/listen for signals
3. Also phase each symbol of interest, and note its dominant cycle and maybe plot a possible future price direction -no need to be too specific
4. Identify the most likely trade candidates. Put them in their own group, and switch to the H1 for those symbols. Or just stick with gold and EURUSD/USDJPY

Entry

1. The idea is to watch green band (H1) piercings, and take a trade from one side of the orange band (the container band) to the other, or as far as possible if that isn’t happening. You can also take higher TF (longer horizon) trades by watching the orange (H4) bands and taking a trade from one side of the yellow (D1) container band to the other. It might also be possible to trade lower than H1 (M5??) using the same principles but it might be too chaotic and more effort than it’s worth. It is worth investigating once you’ve mastered this system.
   1. Wait for a band piercing
   2. Wait for a confirming (color change) renko block signal OR
   3. A confirming signal from Jack;
   4. A confirming Jack signal is one that is either higher or lower than the previous signal of its type, suggesting a change of price direction.

2. So for example, suppose price pierces the green band at the top, it will likely generate a Jack sell signal (often, a cluster of them, but the whole cluster should be treated as a single signal). That signal is valid if and only if it is lower than the previous Jack sell signal. If that’s the case this is a valid entry.
3. Suppose price pierces the green band at the bottom, generating a buy signal. It also pierced the orange band, generating a higher TF signal. This is probably a local trough, and it may be a while before you get a Jack buy signal, so follow the Renko block lead. Once you get a green block, go long. You can enter one trade for the H1 TF and one for the H4.
4. Optional: Re-enter on future clusters of signals that confirm the band direction you’re trading from, as long as they satisfy the preceding criteria. Always use a lighter weight for every pyramiding trade.
5. The first trade is the trickiest. (why is that exactly?) It might be worthwhile to wait only for higher TF band touches and reversals, then follow the flow. Don’t exit too early, as it is hard to get back in sync.

The idea is to ride these trades for the max duration of their respective (half) cycles. Only add to new positions when price is validating it, and only when price hits the other band (buy/sell on dips). To do this, simply ignore long signals that are not higher on bottom band touch and ignore short signals that are not lower on top band touches. See explanation below. At a future date, if practical, modify the indi to ignore invalid alerts.

Demanding such a specific outcome from the chaos of the marketplace is asking for trouble.

Exiting
There are only two exit signal types:

1. Band piercings
2. Jack signals in which the order logic is contravened (see below). Wait for a buy signal to close a short or a sell signal to close a long unless price is zooming away from and/or distorting containing bands. If this second signal is itself a continuation of the original trend, then don’t exit. In this way, you need 2 different signals to exit on Jack signals.
3. It’s probably just fine to sneak some stops up well behind each swing high/low to lock in profit.
4. Note that Renko blocks are not exit signals. By the time a renko block is confirming an exit, price should have pierced a band or made a Jack signal that alerts to a changing trend. See Renko notes below

Fig. 1 Enter/Exit logic

Enter long on a ‘buy’ signal and enter short on a ‘sell signal’. Only enter on valid signals, near or at inner band or container band piercings. Only add to positions when the corresponding previous position was part of a favorable pattern (explained below).

For example.
(Fig.1) In the diagram above the first yellow dot to the far left is a bullish band piercing but it is not yet a valid signal as it requires a subsequent buy signal that is higher (a higher low or HL as it is labelled). The second dot is a valid buy signal and an entry point as it makes a higher low.

Something that watches higher highs and higher lows using a renko brick or a point and figure chart is something worth building upon. Why have I mixed it all up with this other voodoo?

This system does not (yet) trade long and short simultaneously, so the subsequent sell signals are ignored. Note that they are also rising, confirming the bullish trend. The next buy signal is also a higher low, confirms the bullish trend and is an opportunity to pyramid. The centerline of this band (not pictured) would be rising bullishly. The other two buy signals are lower, but are part of a cluster, so they would be ignored. When price reaches the other side of the band, set a SL to break even.

If price hadn’t reached the other side, the trade would have failed, and you’d exit on the next bottom band break, a buy signal that is a deeper trough (lower than the previous buy signal clusters). In this case price does reach the other band and creates a new high and a (invalid) sell signal (a higher sell than the previous cluster and so, ignored). If it had begun retracing before creating a new high, thereby creating a valid sell signal (lower high) , we would exit the trade early or let the SL take us out for a small gain or breakeven.

The subsequent sell signals (dots 9-10) are unclear as they are basically level with each other. This system advises ignoring signals that aren’t clearly higher or lower than previous ones. The second sell alert labelled in Fig. 1 should probably be ignored. However, after price reaches its peak near the 11th dot (band piercing) it drops sharply, almost certainly hitting our SL and taking us out for a tiny profit or breakeven. The 12th dot is an invalid buy signal (lower than the previous buy signal cluster) and would signal an exit to this trade if the SL hadn’t been triggered for some reason.

The next entry opportunity (short) is after the third sell alert, where price pierces the top band, and makes a new lower high. Price continues to trend lower, giving us several more opportunities to add to our short position. We would continue adjusting the stop, securing profit, according to a mechanism I haven’t devised yet, and would still be in the trade at the end of the diagram. We would exit on a container band piercing, or a higher high -- a sell alert that is higher than its predecessor. Note this takes us out of the trade, but it isn’t a buy signal until price pierces the bottom band, and makes a higher low, higher than the previous buy signal.

If this is confusing, here’s another way to describe it.

1. Enter long on a new blue dot cluster that is higher than the most recent old one, and enter short on a red dot cluster that is lower than the most recent old one.
2. Exit long or short on a container band piercing or exit long trades on any red dot cluster that is lower than the most recent old one, or any blue dot cluster that is lower than the most recent old one.

This is the origin of one of my 'trade order' principles.

Alternate entry: Sometimes, after a band pierce, price moves too fast to make a retrace for the entry signal. So after such a piercing and reversal, watch for the supersmoother candles color to change, and enter in their signalled direction. Augh. complication.
Solved with Renko, hopefully.

Don't hope. Too complicated for sure.
I continue in much more detail talking about the Renko indicator. It probably deserves its own post.

Renko notes:

1. The general logic of the Renko block is to follow its counsel when it agrees with what happened at the most recent container band piercing. So if you got a piercing at the top of a cycle, and Renko turns red, that’s a good sell entry, no need to wait for the block to shift down. (bandsaw)
2. Ignore it when it changes colour before a container band piercing AND doesn’t shift up/down in the wrong direction.
3. If Renko changes to the wrong color AND shifts up/down in the wrong direction, exit, but hopefully, by then you’ve had the signals from Jack or a band piercing to get you out earlier.
4. Renko is good for confirming a trend is real (avoiding fakeouts) and staying in a trend, but not so good for detecting winds of change. It’s an intentional laggard. (like most every indicator)

Fig. 2 The classic elbow pattern (sign of a reversal)



Fig. 2 Notes:

1. Inner band pokes through outer band like an intestine invading the abdominal wall of an inguinal hernia patient.
2. Price forms a W (or M) with the point at the elbow crease. It rebounds and continues in the opposite direction.
3. Higher level trend is now changing. Inner band centerline reflects this change, and outer band centerline will catch up.
4. Don’t enter on leg 1, leg 2, or leg 3 of the M/W. Wait for the pattern to play out. (Renko block color change is a good confirmation). Let leg 4 carry on to halfway point (centerline) of container band AND THEN pierce through it (and retrace, possibly) and THEN enter. You do miss a big part of the upper move, but this should be a longer term uptrend, there are lots of profits to come and the time and money you save by staying out (and avoiding fakeouts) more than make up for it.

Highly Dubious Like All Patterns

1. Except part of it is much like the Sperandeo 123 where price breaks a recent significant local high or low.

Hi Clemmo

Great thread...and looking fwd to moseying thru it... And sure enough one of your very first posts I read here and I think ahhh someone who gets it.
As far as Im concerned the best loser wins...we learn alot about ourselves from the "stinging" losses.

Anyways hope your staying warm Been quite a winter here so far in Canada ! (BC) warm now lol...

Cheers
Blue

Edit : and your book thread is Top Notch...Thanks for all your work..

Thanks a bunch! We got through the cold snap here but I'm looking forward to spending more time on the coast.

London Open Straddle

Not all my strategies even in the olden days were complex and unreadable.

Can't get much simpler than that. I still like this idea and I think I'll build it. Of course there are already hundreds if not thousands of London Open strategies and experts. I could just find one or buy one but there are reasons not to go this route.

1. Most of the stuff that's out there is too complicated, and filled with frills that are not only unnecessary but possibly can harm performance. This is a natural part of the development process, caused by the temptation to add things to 'fix' problems with the EA that are actually inherent problems with trading. I've written about this before. Often the best version of an EA starts with 0.0xx.
2. There's nothing quite like knowing exactly how something works. If something goes wrong it's your fault and you know what to fix or at least where to look.
3. The intellectual challenge of developing a skill. If you stop using your brain, especially at my age, it will stop working.

...short duration (usually 1 day to expiry (ie, 1 dte)) straddle, strangle, and iron condor trades on the liquid index options (SPX the fave due to its large size) have gotten very popular with options traders. Entry can be made in the morning of the expiry day or at the end of the day before. In this market that is fond of sideways action, but that nonetheless has some good-sized moves in the sideways chop, you can often make money on both sides of a strangle. I traded several of them in my thread today, starting here https://www.forexfactory.com/thread/...8#post14268738

Some day soon I'm going to start educating myself about options. At least I keep saying that. Your thread has been an inspiration for years.

Jotunheimr
Can you join the realm of giants? Image credit: Framestore Viz dev


In the book thread the subject of moving averages has come up. I have had an off-again/on-again relationship with them since I started and for now, we're off, but sometimes I find one and I think...maybe, baby, we can still make it work...
This system is based on just such a moving average. It's a cunningly, subtly modified version of the standard MT4 simple/smoothed/EMA average. I found it to be too 'signally' for my taste so I turn off everything but the line colour changes. It was coded by some genius named Masemus. Wherever you are Masemus - thanks for wasting a lot of my time ... but in an awesome way. Attached for your pleasure.

Here I am comparing it to some other MA. Probably some type of supersmoother. What do you think? It looks even smoother than the smoother. That's smooth baby. Oh no. No. I can't get hurt again.
edit: no you can see from the screenshot it's a TMA! I don't even know who you are anymore, moving average.

Entries:

1. Start trading when the signal line (Jotun) is sloped
2. If price has already moved off the line wait for it to return
3. Enter with the signals from Jotun (red - enter short / green - enter long)
   1. Aggressive entry - as soon as price crosses and closes on the other side of Jotun

   
   
   

4. Add-on trades (single units) with every swing high or low that retreats away from the Jotun line. Often, these will be signalled by ATR flags.

I complexified this by adding an ATR channel breakout indicator; but you can just see when price has moved far enough.

1. ***The distance to the line is variable, but if the flag/swing is in the outside half of the band then it’s too low probability to bother with. Always expect a return to the Jotun and don’t feel bad for missing out if it takes longer than expected . If you’re feeling gutsy, use any old swing and if not, insist on contact with the line with no close on the wrong side. If price closes on the wrong side don’t add-on or if you have already, close it.
2. Long trades taken under the dotted line (MA array) call it Ifingr, and short trades taken above the river Ifingr are ‘underdogs’. Trade these trades lighter or more pessimistically than the initial trades.

Exits:

1. After an ATR signal take partial profits (probably ⅓) - stalk the arrow a bit to see if it will continue but exit at first sign of retrace. Also exit any add-on trades on this signal.
2. Do this on any repeated ATR signals or if price goes what you consider a sufficient distance beyond the first signal. Don’t be scared to take profits.
3. The last trade of the initial trade is the ‘lotto ticket’ trade. Don’t allow it to go negative, place appropriate stops. In fact, consider allowing no trades to go negative once a partial profit has been taken.
4. Underdog trades stop out on a cross and close on the wrong side of Jotun. Initial trades stop out on a cross and close of Ifingr or a signal change from Jotun if Ifingr isn’t positioned to be a suitable stop line.
5. Update: (tentative) If there’s an ATR flag and price doesn’t reverse in ‘short order’ (2 bars, or without a a second ATR flag, let’s say) then exit everything, abandon your plans and go with the flow. Especially since, by then, most likely the Jotun will be suggesting you do so.

Besides being both complicated and inflexible, like most systems based on simple indicator signals, the basic problem is that moving averages love you when the trend is friendly but break it off when the range comes back to town. Avoid this temptress! I didn't take my own good advice and tried to make it work. With variations.

Variation:
When Jotun is flat, trade the range, using the ATR signals as contrarian signals and their arrows as the boundaries of the range, or just make the ATR channels visible. For safety, take trades only in the direction of the likely larger trend.
Update: I think this is vital. Or avoid trading.

Update: when trading ranges, don’t do it too late in the day (>6 hours) and respect the Jotun.

I think you are asking too much from this relationship! Hey this one has some testing notes!

"Initial testing is promising, some experiments with other MA types might be tried - screens attached. Jotun is remarkably coherent even if it is too flat near peaks and troughs."
but then later
"Update: 2018-11-19 more simulated testing isn’t going so hot, a few crash and burns. I’m not sure if it’s my lack of patience or if entry/exits need to be defined more precisely."
And then I suggest using Hurst cycles to gauge trend direction and it all goes to hell. Oh, where did we go wrong? How did our love go bad? Was it when I stopped believing moving averages have predictive value?

I don't understand what you mean, can you expand on this ?

It's a cunningly, subtly modified version of the standard MT4 simple/smoothed/EMA average.
Well, I took a look at the code for #JoGET_MA2b# (mtf + alerts).mq4 and decided likely that a simpler MA will tell the same story. Pictured is the JoGet with period 16, method simple, and Hull MA(20,1), both with Close price as input. You can see the HMA is less "signally", as you prefer. the HMA is yellow/magenta.

Besides being both complicated and inflexible, like most systems based on simple indicator signals, the basic problem is that moving averages love you when the trend is friendly but break it off when the range comes back to town. Avoid this temptress! I didn't take my own good advice and tried to make it work. With variations.

I use MULTIPLE INSTANCES of MAs that show me a "smoothed view" of price for 4 TFs at once: Nothing complicated at all about that. And I include a short-and long-term ATR for guidance about ranging periods. The most "complicated" aspect is determining the TF to use to give the direction bias and then checking where the TF for entries is--and really, that's not a complicated task to perform, either.

The direction TF and entry TF are both dynamic. So as price makes its swings back-and-forth on smaller TFs, the entry TF dynamically shifts. Eventually, the direction bias TF will also shift. But again, nothing complicated about it.

Order From Chaos (Focus Edition)

Attached Image (click to enlarge)

I have to assume that the Focus Edition of this idea means that there was an even more un-focused edition at one point which is thankfully lost to history.

This is one of those rabbit-hole, deep-dive, attempts to get my head around an impossible topic that results in hundreds of digressions and few solid truths.

The main indicators I was exploring at the time were:

1. Chaos visual averages
2. Target Bands

Both are attached for your amusement. I don't have the source for the Chaos. The source of chaos is probably best left to philosophy or advanced mathematics.
I later learned that Chaos is simply the Williams Percent Range indicator with two filter length levels. I still think William's idea is clever. Useful though? I'm no longer so sure. Its main attraction is that it calls attention to the likely low or high based on how much (the percent) of the range was attained in preceding periods. However, due to 'spectral dilation' there is always another low or another high. You can go broke waiting for the low or the high to come in.

The target bands are a seductively adaptable triangular moving average band that adjusts as price changes, so they're always accurate, and always wrong. If you stuff one into another and layer them a few deep you get a very clear picture of how price will play out in the future. A crisp, credible lie.

Paraphrasing 4 pages of notes. I don't seem to have the original template for this. It would have looked something like this, but with a lot more bands.

1. Meant for M5 trading
2. Determine the dominant trend
3. Place 2 trades - one scalp and one swing
4. WPR1 (the faster line) is the scalp trigger; I wasn't sure if I should use the Chaos signal or a band bounce
5. 2:1 volume ratio for scalp to swing. About 0.1 lots for $2500 for the scalp; 0.2 for the swing; aggressive!

Entry
(scalps)

1. Wait for price to move to a band edge
2. Wait for WPR1 to go above 30 or below -30
3. Wait for price to curl back towards the centerline then enter

(swings)

1. wait for WPR2 to go above 40 or below -40
2. wait for WPR1 to curl back to centre then enter
3. Don’t be surprised if new highs/lows are printed after initial entry. You could either enter a lighter trade on new ones, or exit and re-enter. Or do nothing and wait for price to revert back to the mean. (what’s statisically/strategically preferable?)

Exit

1. On a scalp failure on a countertrend trade;
   1. failure defined as a piercing of M15 band or failure to make new higher high/lower low

2. Hold on a scalp failure on a with-trend trade because they might work out when the trend resumes;
   1. exit if WPR1 exits the 30-40 zone on the correct side but price didn’t behave.

3. Swing failure countertrend
   1. WPR2 reaches 40/-40 zone on the correct side and price is still not behaving.
   2. Touches H4 bands

4. Swing failure with-trend
   1. Pierces D1 bands
   2. Trend change on D1 according to WPR2

   
   

5. Scalp success exit; wait for
   1. WPR1 to curl towards centerline above/below 30/-30 in the correct end-zone.
   2. Wait for price to pierce the M5 band in the correct direction.
   3. Don’t rush to re-enter. Try to ascertain price has reversed.

6. Swing success exit. Wait for either:
   1. WPR2 to change colour
   2. Price pierces H4 band; only if this band isn’t already invading the lower TF bands (which occurs during reversals). If it does this and returns towards the H4 centerline (doesn’t have to reach it) you might be safe placing another swing trade rather than waiting for a new signal. Needs testing.
   3. You don’t have to be as careful on re-entry since you’re giving this trade lots of room to get on the right side.

This is far too ambitious, at least for me. Trying to scalp and swing trade within a semi-flexible system? Checking things from M5 to H4. What a headache. And it creates an endless sequence of what-ifs and discretionary choices.

1. What do you do when a trade was with-trend when it started but dominant trend changed when it failed (making it counter-trend - in other words -a breakout reversal)?
2. How would you re-enter then? Wait for a new signal? Jump on?
3. What if price fails on the wrong side of the band a second time? Wait for a certain number of candles?
4. Should a trailing stop be added after price pierces a band?
5. Constrain a trade by time? Fixed point stops?

Scan the daily setup on D1 and consider what direction it’s favoring. Price next to a band, will move away from it, price close to a higher TF band will be attracted to it.
Nah.

When the bands are wide price seems to be ranging - it makes sense - it’s indecisive. When bands are narrow price is trending - it is zeroing in on a target, and there is little interest in one side of the price auction. So - counter-trade with wide bands, and stick with the current trend when bands are narrow.
Yes, but how often does the price trend when you're in the trade?

I think I abandoned this after 9 bad trades in a row. However the lessons I learned from trading it spurred me to write another two pages of notes some of which are interesting.

THINGS LEARNED FROM BAD (and GOOD) TRADES

1. Avoid trading against trend with narrow bands
2. Avoid trading against WPR - esp. Against both WPRs! (I mean, if you're using an indicator, use it, otherwise, what's the point?)
3. Wait for the curl up; Wait for the band touch (in other words, follow the rules of your own system? Uh, yeah)
4. Don’t make trades with price in the middle of bands (trade only at extremes)
5. Don't panic when price retraces if you’re trading with-trend (with WPR). Leave the trades open and be patient. Exit when trend changes and the trades are hopeless.
6. Trade with trend, whenever practical
7. Most often when you look at a chart you’re looking at an expired setup or a bad entry point, because, when conditions are right, price moves to where it’s going fairly quickly, then it rests and waits for conditions to move again.
8. So look for future setups, use stop/limit entries to get into trades when conditions are good again.
9. If you’re present at the computer, not tired, focused, then you don’t need to use stop losses. (well, if you're present at the computer that's the only time you should be trading, whether you use server-side stops or not; otherwise offload your duties to hardware)
10. Follow with-trend trades to their completion. (this is still an open question; what is their completion? When do we tug the leash?)
11. Be careful if you switch timeframes to get perspective, to use only the WPR of one timeframe to make decisions at the moment of truth. Don’t for example, work on M5 then switch down to M1 and then forget that you switched, and then panic that it is now pointing in the opposite direction, and prematurely exit a trade. (LOL. That's pure inexperience.)
12. Don’t place trades near the end of your waking day when you’ll be too tired to see them through to their completion. If there are trades you can’t watch, hedge them, and come back fresh. (Or just exit.)
13. Be wary of following patterns into the next day - things tend to reverse
14. Don’t try to predict.
15. Quick reaction is the key
16. You should not have to think (or worry) about any trade. Just do what the system dictates. Don’t pay attention to the score. (well, pull the plug after 9 bad trades if that's unusual)
17. If it’s not clear what to do on the current TF, either don’t trade or move up to another TF.
18. If you can’t watch trades closely trade on H4/D1 according to the WPR
19. If you’re down on the day, the week, the month, don’t try to catch up by trading heavier; you’ll just dig yourself deeper. Instead, look at the trades you’ve made and identify which error you’re making, then don’t make it again.
20. Good trades will look good quickly - figure out how long on average it takes - (1-9 bars - so 45min. max!) and if your trade isn’t behaving, cut it. Better to lose out on some runs then to get stuck in a turkey. (yes, and no. The market's appetite for mind games is nearly limitless; maybe say wait for a better opportunity to present itself before cutting)
21. The rarer the event (higher TF the band break) the bigger the trade opportunity, in theory. For that reason, if a particular trade encompasses several band breaks, place multiple trade lots so you can exit them at different points. (yes, but be careful; what you think is a big opportunity might be for the other guy)
22. 2 or 3 peaks/nadirs are usually the most you will encounter before a profitable price reversion.(unless your account is about to blow up; then it will go as far as it needs to)
23. Pick an arbitrary number of points (or candles) that you will let price drift before pulling the plug
24. A trade that drifts away, returns to point of origin, hitting the band and bouncing away in the wrong direction should be considered a failure and cut quickly. Don’t be forgiving of lazy or contrary price behaviour. You can always get in again. For now don’t be too concerned with fees for frequent trades. (another open question; the market will play mind games with as many people as it can; sometimes it's better just to go do something else and come back when things are moving)
25. Single points of contact are much more likely counter-trend. With-trend you should expect multiple bases and even multiple stations, simultaneously. Which is why when trading with-trend we wait on the band edges, expecting price to continue until it reaches its range max and maybe a bit more. (stations are the coloured lines at either end of the WPR1 oscillator, and bases are the vertical lines that cluster near a peak/trough. You can see a bullish base formed in the screenshot above.)
26. When price is pulling out of a station/base it might be a good idea to zoom into M1 to identify the point of price pulling away. (nah, no.)
27. Big question - if price is trending in the opposite direction of the last closed trade, at what point do you start making with-trend trades again? The WPR1 might not make another base/station for a while, and the whole time price will be moving in the direction of the trend. So - place a with-trend trade the moment both WPRs are in agreement (i.e. after WPR1 curls back towards the direction of the last base/station) or make a rule that WPR1 has to at least cross the midpoint before curling back? Enter on every curl back and exit when the new base/station is made? This is a difficult nut and crucial to the system success. (message to me, 5 years ago: no man. It's not important at all. Just find something else to trade that isn't ambiguous.)
28. If you make a hail mary move you have to stay awake (or set a timer to wake up) to follow it. If you can’t do that, hedge and try again later. (no man, just don't do that. There's always another trading day)
29. Maybe we need to have a bias towards dominant trend (did we not just say that?)

Anyway, this shows me that I've known some things for at least five years and yet here I am still working on it. Why is that? Clearly there's a difference between knowing a thing and doing that thing. Also, good common sense is probably not enough.

, happy new year mate

Overhill Underhill

Attached Image (click to enlarge)

You draw far too much attention to yourself, Mr. "Underhill."

The basic idea is to load up with short orders at the top of the likely range, and long orders at the bottom of the likely range. While price dithers between the 5 min. bands you take regular small profits so that you don’t feel like you’re wasting your time.

I'm still chasing this dream. Also I probably should have introduced this before 'Order &Chaos' because this one was written first and explains things better in more detail.

Looks a bit risky here.

We need to define some terms.

WPR1 & WPR2 - the two Williams Percent Range oscillator lines in the Chaos indicator (uploaded earlier)
Station - when WPR2 makes a range high/low and its line changes colour to dodger blue/rosy brown. Usually accompanied by some vertical lines of the same colour. The entire coloured line is the ‘station’.
Base - when WPR1 does much the same thing and its line changes colour to blue/orange.
‘Leaving the station/base’ - when the WPR1/2 line colour reverts back to its usual grey/beige, indicating that price is leaving its local range high/low and is less likely to resume the previous trend. Returning to a station or base is the opposite situation, where price forms a second (or more) base or station shortly after the first.
The hill - the price wave usually consisting of a local peak followed by a trough or vice versa
Overhill - local price range peak
Underhill- local price range trough
Anchor - the final trade placed overhill/underhill that secures your margin and must not be removed until price has definitely left its station/base. Otherwise in a strong trend, you will find things get one-sided fast. An account-endangering situation. If this happens a new anchor has to be placed, at possibly the worst possible point, and things can go from bad to worse. Knowing when to pull up the anchor is one of the crucial aspects of trading this system.
Run up/Run down - one sequence of placing one or more trades underhill/overhill, followed by placing the opposite trades at the opposite side of the hill, followed by closing the original trades including the anchor. Ideally you will do a run up followed by a run down, and so forth perpetually.
Cratcheting - see Xmas Carol trading system. Making trades bouncing off two sides of a set of bands.
End Zones - the 30-40 and 40-50 zones on the CVA indicator. These are usually the areas where bases/stations are formed.

Rationale:

1. If you think about it, there is no point in placing long trades anywhere but at the lowest point that price will reach during your trading session, and short trades should be placed at the highest point.
2. Easier said than done, but anything else is less than efficient.
3. Grid trading systems for example, add unnecessary risk and make inefficient use of margin by placing trades at many points of a range instead of at local maxima.
4. Luckily the Chaos Visual Averages indicator (thanks Mladen) is pretty good at detecting the local peaks/troughs of a trading range. I think it’s basically two W%R (williams percent range) oscillators, one fast and one slow. I say local, because the CVA cannot predict the future, and a new local extremum can form after a first one has been identified, causing a grid of prices to form on the wrong side of the trend direction.
5. One way to guard against this is to keep at least one trade (the anchor) on the opposite side of the ‘hill’ that gains value as price trends. The other way is to place as few trades at the local maxima as possible. The latter is done by placing trades only when price is ‘leaving the station (or base)’. Since it’s rare to have more than 3 or 4 bases or stations in a row without at least one intervening station/base on the opposite side, this preserves a balance. Since trades are mostly being placed at price extremes, this can be very efficient, cutting down on overtrading, and ensuring a good profit ratio.
6. To ensure that you are earning profits during ranging periods (the usual state of price action) we also place trades at the top and bottom of a set of bands - in this case, 5 minute target bands. The other bands in this system are simply to help identify higher timeframe extrema and to gauge the dominant trend and are not strictly needed.

Entry

1. Much like OFC, wait for price to 'leave' a station and curl towards the centre.
2. Re-enter whenever a new base/station forms and price leaves it.
3. Hint: use higher TF to see how likely it is that the trend is going to reverse - you can tell by how far away it is from a higher TF band colour (LOL )
4. Cratchit when price is ranging
5. If WPR2 is lazily meandering around the midpoint without making many (or any) or small stations/bases that’s a good indication that it’s Cratchit time.
6. Enter when price touches or pierces one side of the band and then starts to return. The target band indicator produces a little dot that gives you a signal.

Exit

1. Exit when price has left the station/base on the other side of the hill, unless you only have one position left open on that side (the anchor)
2. Is it better to exit at every base, every station, or only exit base trades at base exit points and only exit stations at station exit points? Gut feeling is it’s better only to exit at stations for both base and station trades, but test it. (I wish I had a definitive answer today)
3. Remove the anchor when you feel that price will not make any more bases/stations on that side of the hill. This is more art than science. You could specify waiting until price has left the 30 zone (although this will fail often), or is approaching the midpoint line. Maybe it’s when price ‘slides’ past the target band wall and continues to the opposite end of the hill. If WPR2 still has lots of room before peaking/bottoming then there’s a good chance there will be more than one base. How do you predict multiple stations? I don’t have a solution (phasing analysis - category interactions? Examine higher TF). Look at the curve of the bands. If they’re starting to curl up/down that could be a sign. This is imperfect as is any system that isn’t the holy grail. Work on this. Maybe an RSI type indicator would help identify when trends are running out of steam, but I doubt it will be better than CVA.
4. It’s an acceptable cost of doing business to remove the first trades placed overhill/underhill if they’re now well below or above current price and there is little hope that price will revert to catch them. If you still have your anchor (you do don’t you?!) then the value of that final trade will more than make up for the pruning of old base/station trades. Generally speaking you only want one trade per base/station but it’s not hard to get faked out and if price is really trending hard there could be a lot of bases/stations in a row. Use your human brain to determine what’s going on. Usually a trend of that magnitude is caused by some fundamental news. So check your calendar and plan ahead. If you have a strong dominant trend bias, you might just leave those stragglers in place since you’re sure that when price resumes the dominant trend they will get picked up. Don’t be wrong though! They can become account killers. ("Don't be wrong" is not good advice for traders. Or trading systems.)
5. It’s a good situation to be in if you have a minimum of two trades anchoring the opposite position since if you take profits on one, and it turns out to be a massive fakeout, you still have the other one to provide anchoring. Since you will now have stations and bases on the other side of the trade, becoming profitable as price returns to your anchor point, you needn’t worry about being slow to finally take profits on the anchor, maybe even wait for WPR1 to cross the centerline.
6. Basically just do whatever it takes to avoid a situation where you have a bunch of trades on one side of price momentum and nothing on the other side in case price suddenly moves a lot in that direction and you are neither profiting from this big move nor safe from a margin call if price never reverts back.
7. Watch WPR2 (and to a lesser extent WPR1) if it curls away before entering the end zone. This might be the end of its run, so take profits (but not the anchor!) and remove the anchor when there is no doubt that it has finished the run. If it recovers and resumes the trend, be extra conservative when placing new trades until the run is complete.
8. For Cratchits:
   1. Exit when price touches the other side of the band
   2. You could be slower on the trigger for with-trend trades as they’ll often bust through the bands and keep going.
   3. Generally cratchits are just something to bolster income and so don’t let them interfere with run trades, and don’t let them get on the wrong side of the anchor.
   4. If the cratchits are cluttering up your real estate because there’s a run trade in the way, and this bothers/distracts you just wait for price to move away before starting to cratchit.
   5. Watch out for a range that suddenly starts to trend. Close the cratchit before it becomes a liability.

   
   

9. Stops:
   1. It would be perfectly viable to trade this without stop losses or take-profit limits since whenever price busts past your expected high/low you simply create a new order at the expected maximum, but if that isn’t your style and you want to play it safer:
   2. Cratchits - place stops about 7 pips above or below the band
   3. Run trades - place stops WHERE? Maybe this isn’t an option.

   
   

Summary:

1. heap as much position weight at the top and bottom of the range as safely possible
2. exit it all then rinse and repeat
3. band-trade when there is no trend
4. this system avoids the need to think and/or worry. You are only watching for a trio of conditions
   1. is price leaving a base or station?
   2. is WPR2/1 signalling a change of trend direction?
   3. is price touching or bouncing off a band?
   4. If the answer is no to all 3 of these, just keep one eye on the charts, and one on whatever else you do while you trade. You can also use pending orders to ensure you get the best execution at the likely top/bottom of the ranges/band positions.

   
   

Reducing cognitive load should reduce mistakes.
Last but not least, practice and have all your trading backups/failsafes in place.

Can this be automated? (almost assuredly, with the right team)

Now that I reread this, it isn't so absurd. This SHOULD be tradeable, but in practice a trending instrument tends to keep trending; so the Chaos visual averages are much better for scalping/mean reversion. That's why I created a 'Reverse Underhill'. But I didn't document it.

Reverse Underhill


Well, here's the template and the remaining indicators you'd need to reconstruct it. Maybe you can reverse engineer how this was supposed to work? I'm fairly confident if I didn't write it down it wasn't worth worrying about.

Happy New Year to you, and best wishes for 2023!

Just an addendum to the Underhill story. I found this attempt to document something about it. It dates this system to late August 2017.
https://docs.google.com/spreadsheets...it?usp=sharing

It shows how easy it is to make vague predictions about the future and to be mostly right; it also supports the notion that if you want to make long-term predictions use the highest timeframes you can.

Killzone NY

Kurt Russell has 24 8 hours to escape the killzone and emerge with 150 pips….
No, but this is a true day trade strategy that requires you be up at the same time every day. Think you can handle it Daysleeper?

This was inspired by something in an 'Inner Circle Trader' video.

Step-by-step.

1. If you haven’t already, make the template. It’s like Luke Skywalker making his own lightsaber. (you don't have to build it; template included)
   1. Add 3 indicators (attached)

   
   
   

2. Set up Fibs as per Huddleston. See Fib level notes.
3. Now trace a dashed trendline (white) from the start of the day (0:00 server time) to the 14:00 NY open line. Whatever slope it indicates is your trend bias for the day. No slope at all? Flip a coin, or skip it today, or use some other method to select a bias. Probably the latter. Once you have your slope bias consider the following. Does it agree with the distance from the band? Good. Does it not agree? Ugh. Either dial down the position or choose not to trade, or make up your mind some other way, possibly by observing what happens right after the open.
4. Either way, you are going to observe price (on M5) until time reaches the trigger point (15:30) unless you have reason to act sooner, usually you won’t. At that point you should have a local minimum and maximum (swing high and low) to draw a fibonacci retracement/extension in the direction of the expected remaining daily trend.
5. Enter when price retraces back into the 62-70.9 percent range. It often gets to the deepest target point, so no need to be hasty, but watch carefully during this time.
6. Let’s try a 3 strikes systems for s/l. When price moves against you past the 100.00 line (“point of no return”) where you’d usually put a s/l line, make a note of it, that’s strike one. If price retraces back across the point of no return, good, and if it goes back the wrong way, that’s strike two. The third time it goes across the line, kill the trade, you’re done, and Plissken is really dead this time.
7. Take partial profits (¼?) at the 0.00 first profit scaling line, then target 1, target 2 and if you get all the way to symmetrical swing you might consider just letting it run with the stop back at target2.
8. That’s it, I think - a daily trade that is usually closed every day. In/out, surgical. Stack the pips, repeat the following day.

Variations:

1. Basket trade it
2. Try doing this at the LON open - probably riskier, probably not more profitable.
3. And/Or walk this back an hour and test again each time, til you find the sweet spot
4. Trade this on the simulator and note where you get most of the losses. Maybe you can use some of your other indis to avoid trading at this point.
5. Draw the sloped line from LON open to NY Open only?
6. Create a scanner where fibs are pre-drawn and alerts are set for retracements on every instrument? Then it’s just a matter of deciding if the setups are still valid before entry.

Does this work? Let me know! I couldn't get anything Huddleston/Inner Circle Trader showed me to perform more than randomly. I don't think Huddleston knows much except how to fleece subscribers. This is just a 'buy the dip' idea though, so it might be fine.