Now Reading
Introducing Fractal Monger
90

Introducing Fractal Monger

by The MoleSeptember 22, 2014

In the past few months I have been immersing myself into the world of machine learning with a strong focus on data cross-validation. Quite frankly intellectually it’s been a brutal toil for me as my math skills had gotten a bit rusty since college and since this was completely new turf for me. I will spare you an account of all the trials and tribulations I have been through – as you can imagine this subject is not exactly considered light reading.

Repeatedly I found myself researching a particular promising avenue for weeks on end which in the end turned out to be nothing but the precipice leading to a steeper mountain of work hiding behind it. However without understanding the basic concepts the more advanced ones will simply confuse or elude you and the premise of their evolutions will not be clear. As I spent weeks absorbing a mountain of white papers I often only realized their significance (or lack thereof) many weeks later. As I started to put theory into practice some of the ideas I had developed over the years were confirmed – the ones which weren’t living up to the data were discarded.

At this stage I would not consider myself a quant boy of any sort but as they say I do understand enough to be dangerous. Given that, what I’m showing you today is only a first foray into developing more advanced systems and pertinent tools. And it addressed an issue most short term traders engage in at some point in their career: time series fractals. It’s the one aspect of technical analysis that has always fascinated me but which I never really could put to work.

When it comes to historical precedence most of us spend our time driving blind – we see a time series and we don’t have any idea as to whether it has occurred in the past and/or if it has any statistical meaning. Which doesn’t really matter if we have a functioning system that has shown an edge – perhaps that system actually implicitly takes advantage of one or more time series fractals. To be clear – I’m not talking about candle patterns here although one could easily investigate fractals on that basis. As a matter of fact it’s something I may do in the near future – but it’s not what I have been working on.

2014-09-22_EURUSD_fractal

This is what I put together over the weekend – it’s based on a converge of several fractal categorization approaches I have been working on. In a nutshell what it always boils down to is this: What are the odds of the next candle closing higher or lower? Well, in this case we actually know:

Current Fractal:

2014-09-22_EURUSD_fractal_graph

The current fractal has occurred 385 times in the past.
It ranks as the 45th most frequent fractal during the past 50000 bars.
The next candle has closed higher 171 times, lower 206 times, and equal 8 times.
Frequency of occurrence: 0.09%
Higher/lower ratio: 0.8
Lower/higher ratio: 1.2

That is the email message I sent myself for testing. It wasn’t as pretty as I’m still futzing with HTML formatted emails but you get the idea. What’s extremely cool here IMO is the fractal graph I’m generating on the fly. As you can see it’s based on four closes, which seems to be an optimal vector size in terms of frequency/occurrence and probabilities. The accompanying statistics tell me that the odds were pretty mixed on this one and that it had a low frequency. It did in the end close lower but I would not have traded this information.

2014-09-22_EURUSD_fractal2

Current Fractal:

The current fractal has occurred 617 times in the past.
It ranks as the 13th most frequent fractal during the past 50002 bars.
The next candle has closed higher 272 times, lower 316 times, and equal 29 times.
Frequency of occurrence: 0.03%
Higher/lower ratio: 0.9
Lower/higher ratio: 1.2

This is what I got an hour later – we wound up closing higher – again the statistical odds were below my threshold which is currently 1.5 (either on the H/L or L/H ratio). And that means outside of testing I would not send myself (or any of you) this fractal as trading on such low odds is tantamount to a coin toss.

But for the next week or so I would like to test it on an hourly basis without the threshold filter, meaning there will be one email per hour on the ES and the EUR/USD – you have been warned. If you like the concept/idea then feel free to sign up for Fractal Monger in the usual place, it’s free right now.

Over time of course we expect to run into fractals with higher statistical value – I actually don’t think we’ll get more than one or two in a week, even if we run it on two dozen symbols. But having done my homework I already know that some of them have very high odds and could easily tie into systems or discretionary trading. Let’s see if we can catch some of them.

Your input/comments/impressions are of course highly encouraged. See you later today.

It’s not too late – learn how to consistently bank coin without news, drama, and all the misinformation. If you are interested in becoming a subscriber then don’t waste time and sign up here. The Zero indicator service also offers access to all Gold posts, so you actually get double the bang for your buck.

Cheers,


About The Author
The Mole
Mole created Evil Speculator amidst the chaos of the financial crisis in early August of 2008. His vision for Evil Speculator is a refuge of reason, hands-on trading knowledge, and inspiration for traders of all ages and stripes. You can follow him and his nefarious schemes at the usual social media waterholes.
Enjoyed this post? Consider a small donation to keep those evil deeds coming!

BTC: 1MwMJifeBU3YziDoLLu8S54Vg4cbnJxvpL
BCH: qqxflhnr0jcfj4nejw75klmpcsfsp68exukcr0a29e
ETH: 0x9D0824b9553346df7EFB6B76DBAd1E2763bE6Ef1
LTC: LUuoD6sDWgbqSgnpo5hceYPnTD9MAvxi6c

  • evilasevildoes

    shweet

  • DollarChaser

    fascinating stuff mole. ive signed up.

  • Skynard

    Will be long @ 4065 retest (2000 SPX equivalent) bounce teritory.

  • mugabe

    Good stuff Mole- good to keep the brain ticking over.

  • http://gerb-reloaded.blogspot.com/ Gold_Gerb

    Germans it must be in their blood.

    http://www.youtube.com/watch?v=5k90KvZ_UWg

    I look forward to the results, good or bad. watch your step, Ohne Dich..

  • http://gerb-reloaded.blogspot.com/ Gold_Gerb

    ..and the bullish percent rolls back over (down intraday)

    http://stockcharts.com/h-sc/ui?s=$BPSPX&p=D&yr=0&mn=4&dy=0&id=p61511725933

  • Billabong

    Update: Closed EBAY for +0.57R and MOS for -0.18R. Opened short positions in QID (NQ), DXD (YM), AAPL and SCO (CL).

    The short oil is one where my long bias is fighting the sell short signal…

  • Skynard

    Remaining long for a gap fill, 5 min div carving out

  • Skynard

    Retest or potential gap fill. Small signal range

  • Skynard

    Short /CC w/ a tight stop

  • SerenityNow

    how do you define a stop when you short at contract highs?

  • Skynard

    Generally, on a squeeze of this nature if it sells off a potential for a retest is likely. So 10 pips above the high would suffice. Low risk trade or you can scalp the bitch.

  • http://gerb-reloaded.blogspot.com/ Gold_Gerb

    Bullish Percentage – Energy
    looking for a bottom..
    [BPSPX]
    http://stockcharts.com/h-sc/ui?s=$BPENER&p=D&yr=2&mn=6&dy=0&id=p96330904244

  • Skynard

    Bull flag

  • Objectively_Bias

    Text book trend day. What did the alert come in at?

  • Kidd Cudi
  • Scott Phillips

    This is a superb entry the best one you have posted yet. I would trade this with a tight stop initially, then relax the stop, then tighten it again once you have 5R

  • https://evilspeculator.com molecool

    Seems like everyone is taking the day off – see you tomorrow then :-/

  • https://evilspeculator.com molecool

    NYSE TICK closed 79.1% below its zero mark. LOW probability of a Down Trend Day!
    Sample size: 2699 1-Tick bars.

  • https://evilspeculator.com molecool

    Guten Morgen – I’m going to send you some statistical data on the SPX after dinner. Very interesting…

  • Scott Phillips

    Cool!

  • Objectively_Bias

    Great thanks! Do you have data on how accurate these alerts are (Low/Med/High)?

  • Billabong

    Nice set-up … it’s one I would’ve taken.

  • http://gerb-reloaded.blogspot.com/ Gold_Gerb

    a multiple of 20 is 60.
    an alternative lens, for short targets.

    http://stockcharts.com/h-sc/ui?s=FCN&p=D&yr=1&mn=0&dy=0&id=p11443271105

    -GG

  • Scott Phillips

    This measures a lack of two sided price action at the open. When they happen they are accurate (very) but can be a trend day without that buy/sell imbalance at the open.

  • Scott Phillips

    This raises an interesting question. When should you use targets and when should you trail a stop?

  • Scott Phillips

    Good day for me today

  • Billabong

    The other question is on shorting itself. Shorts for the last couple of years have been covered quickly … BTFD. At some point it will reverse. I have no idea where we are on this timeline, but I do know many of the stocks I track are already down 20%+…

  • http://gerb-reloaded.blogspot.com/ Gold_Gerb

    I don’t think there is a ‘correct’ answer. but I will say it could break into two BIG categories. range & trend. target for range. trend use trail.

  • steve J

    Hi Scott, Mole
    Firstly, some great discussions on here over the past couple of days which has prompted me to post. I’m after some advice, I have a back tested system that I am happy with and I am now learning to follow it. I still drop a discretionary ( and anticipatory trade) here or there based on my ‘Lens’, but one of my short comings is in not keeping a trading journal. I have tried to keep them in the past, but find that when I add to or reduce a position I’m never quite sure what the best approach is to record these in-trade adjustments. Therefore I am drawn to ‘massaging’ the journal to so that it makes sense (e.g average entry/exit) but this doesn’t reflect the original scenario, and finally I stop recording trades. Do you have any advise on how to log in-trade adjustments.
    Thanks
    SJ

  • Scott Phillips

    If you don’t keep accurate trade journals you will not be a successful trader, period 🙂

    Listen to your reasons for not doing this most essential of tasks and see how they really sound like rationalizations.

    The question arises, why are you doing in trade adjustments? If you don’t log the effect of them, you don’t know if it is a good or bad idea on balance.

    What you should do with each additional entry is treat it as a separate trade (which it is) with a stop whereever the stop is at the time of entry. You should not be trading for anything other than trivial size for the first 3-6 months with your new system.

  • Scott Phillips

    Exactly so. Targets are only appropriate in ranges, and low volatility ones especially. A *breakout* from a tight range, however, a target is very inappropriate. It could be a 10R, a 5R or a 2R, no way to tell

  • Scott Phillips

    Good point 🙂 One more reason why I hate trading stocks

  • steve J

    Thanks Scott
    Logging additional entries as separate trades makes sense. I know the trading journal should be there for my benefit and therefore if it isn’t reflecting reality it isn’t going to help.
    But, are you saying where I close out half R based on a certain rule set, that I then log this as a separate trade and not part of the original one? Also if I have closed half a position it is here that I will tend to re-enter. For example, if I close half my position at 2R and re-enter at 1R do I log that, including the original, as three separate trades.
    SJ

  • Scott Phillips

    When you close out half a trade just average it out. Let’s say you exit 1/2 @ 1.2R, 1/2 @ 2R, the remainder at 4R… you would have .3R + 1R + 1R = 2.3R

    When you take an additional position, treat it as a totally separate thing.

    Remember this isn’t something that will “help your trading”. So I’m not sugar coating it. There has never been a profitable trader in all of history that does not keep detailed and accurate trade journals and spend the time to go through them.

    You will not be the first.

  • newbfxtrader

    Good one.

  • newbfxtrader

    Fractals are way above my pay grade. But if you find it works I can give it a shot.

  • Kidd Cudi

    Unless he’s the oracle.

  • Scott Phillips

    Yes, the alpha and omega. Blessed is he

  • Scott Phillips

    New setup today – short ZB Z4 entry stop 136’19 stop loss 137’06 risk $593 per contract

  • strider

    No wonder you took a vacation. Thanks for sharing it. I’ll be signing up and paying special attention to EURUSD.

  • newbfxtrader

    I dont get that entry. Nice green candle today.

  • steve J

    I’m getting closer then, in my current journal I have columns for up to three exits. Just need to treat the additional entries as new trades
    By the way I make your maths above 1/2* 1.2R + 1/4*2R + 1/4 * 4R
    = 0.6R + 0.5R + 1R = 2.1R?

  • Scott Phillips
  • Scott Phillips

    oops, my math was bad

  • Scott Phillips

    Also important is not just to collect meaningless data you never take the time to review. You are trying to generate actionable intel. This means setting aside time for review on a weekly/monthly basis

  • newbfxtrader

    Thats tomorrows candle!

  • SerenityNow

    Apologies if you mentioned this, I’m new… how are you filtering for trend? I see you’re short Corn, am I right in assuming today’s Hammer would not be a long signal?

  • mugabe

    no wonder he’s coining it.

  • SerenityNow

    noooooo…. big up candle is Friday.

  • newbfxtrader

    Yeah. My bad.

  • newbfxtrader

    Hes down under. He gets the candle before us 🙂

  • mugabe

    that’s the type of chart service I want- not the ones with crappy historical prices!

  • mugabe

    I like it too.

    Also, if you look closely, the long-billed duck formation formed by the BBs makes it aesthetically pleasing:)

  • https://evilspeculator.com molecool

    Based on the response I don’t think any of you guys realize the significance of what this means. You are able to take entries or even bet binary options for instance with real statistical evidence behind you. I’ll post some statistics on the spoos and the EUR/USD tomorrow.

  • https://evilspeculator.com molecool

    I’m starting the think the day crew are a bunch of slackers. The night crew has been kicking ass and the way it’s looking I may as well move myself to Australia.

  • newbfxtrader

    Not even going to pretend I understood how it works but fractals seem to repeat so I guess you are looking for an edge there.

  • https://evilspeculator.com molecool

    Mate, I took one day off after leaving Sunday. Not a vacation really…

  • OJuice

    On that note:

    Not a current set-up, but one from the S&P on Friday to consider in the future.
    – Daily BB flat
    – Gap up with noticeable volume expansion
    – Sell Close on 1 hr, stop at local high
    – Target for retest of lower daily BB

    I haven’t had a chance to take a look at a serious back test. On a quick inspection, there was a similar pattern on March 21.

  • BobbyLow

    Hey Scott it’s great to see that you had another good day. 🙂

    I had a bunch of housekeeping stuff to do so I couldn’t be active on the blog during the day today but I’m pleased to say that I am back on track with Forex which had actually begun Friday afternoon.

    The funny thing was right after JJ posted about the GBP/JPY being a possible short last night, I had begun to answer him that I went short on Friday. Then your reply hit saying that it was more or less marginal and I had an “oh shit I’m swimming upstream moment” and pulled my reply to JJ. But I held on to the short GBP/JPY position anyway and it is still marginal. LOL

    But I also have a Long USD/CAD, Long EUR/AUD and Short AUD/USD. These three are doing fine. The GBP/JPY has been kind of flat lined for about 26 Hours now so it might be coiling up for a move (prefferably down) but who knows? This is a crazy business and that’s part of what makes it so fascinating. 🙂

  • Scott Phillips

    Ahem

  • Scott Phillips

    You need about 100 instances for a real investigation. 50 instances on pen and paper with 3 columns, win,scratch, loss without calculating exact numbers just a tick for win/loss is sufficient to see if the idea has merit or not

  • Scott Phillips

    Bobby I implore you to not take any notice of what I’m doing. Your system is very good, and works in a different way to mine. You take some early counter trend entries, I trade solely with the trend. Neither is definitively a better approach.

    That being said I’m with you long on USDCAD, and I wanted to use discretion to go long EURAUD but I figured I would be cautious it wasn’t quite a setup on my system. AUDUSD looks great, powerful new trend for sure.

    Have a look at AUDNZD.

  • Scott Phillips

    There are many ways of filtering for trend, all with positives and negatives. I can’t think of a single one of them that would call corn an uptrend.

    http://screencast.com/t/FezMsjU3lx

  • Scott Phillips

    HAHAHAHA my crystal ball – oh shit!

  • SerenityNow

    Agree and didn’t mean to suggest the trend was up, just a poorly worded comment on my part. Meant to simply confirm that a Hammer would only be relevant to you in an uptrend

  • Scott Phillips

    Pro tip. The OHLC data you get on most charting services, while accurate for stocks and FX isn’t even close to accurate on futures. The “closing price” is mostly the “settlement price” which is a totally different thing, a custom VWAP of the last 5 mins of pit trading.

  • Scott Phillips

    Binaries, for sure

  • Scott Phillips

    Yes correct 🙂 Others here like to trade counter trend, I believe it is like picking up pennies in front of a steamroller.

  • Scott Phillips

    quack quack

  • Scott Phillips

    On a serious note, as Ivan pointed out in a discussion on crude a week ago (and was right) that big up candle in a downtrend expended a lot of energy. Big baller bottom callers got long on the break of that candle and now are in danger. If it start to fall from here those new traders (weak hands) should capitulate

  • newbfxtrader

    Yeah. Unless it can recover pretty soon I am looking at last years April lows around 86 or so.

  • Scott Phillips

    Day 50 of new system, account is up 39%, $236,500

  • Skynard

    /NQ daily, saved at the NLSL. BB tight as a you know what:)

  • BobbyLow

    I just took a look at the AUD/NZD and if this Pair had been in my stable, I would have gone short September 17th at the close of the 1PM EDT Hourly Candle at 1.1069 with a 34 Pip Stop. Then I would have been stopped out between 6 and 7 PM tonight at 1.0951 for a +3.47 R Gain. This would have been a sweet trade.

    I notice that your average price is 1.1047 so we would have been very close there. 🙂

    Currently running 4 Open Forex Positions which is my Max. I’m also waiting to go Long /ZN via UST which is getting close on the Daily and could be in play as early as COB tomorrow. The Daily on /CL keeps trying to find a bottom but isn’t in any hurry. When it does and shows some evidence that it has turned around it could be one of those nice long running trades even if it is a little herky jerky in the beginning. However, whenever this one gets ripe is anyone’s guess.

  • Skynard

    Looks like ST support, looking at 86.

  • BobbyLow

    Could be Sky. But every year certain patterns repeat in Oil but not necessarily during the same times of the year. In 2013 the high was in August at 112 then a real decline began in September and lasted through November. This year the high was 107 in June and then began a serious decline in July.

    So the 2013 Decline occurred in September, October, and November. 2014 decline occurs in July, August and September? I think if you look back, you’ll find these long periods of sustained declines and then long periods of rising prices with some choppy shit in between. This could all mean something or it could mean nothing but I do pay attention to this.

    The bottom line is that there is no way I would call a bottom here but when we get some kind of sustainable buying that lasts for more than a day or two, it might be time for me to go long at this point.

  • Skynard

    Watching divergences, looking for something ST to shake the weak hands before the next plummet:) Allot of support at the Weekly.

  • Scott Phillips

    I love to see the discipline you apply to your max open positions and your “stable”

  • http://gerb-reloaded.blogspot.com/ Gold_Gerb

    could be interesting going into Month end.

    http://s27.postimg.org/d0xecnvvn/NYUD_2014.png

    I am, the one who knocks

  • phylum

    Sure it was ….”The first day seemed like a week and the second day seemed like five days. And the third day seemed like a week again and the fourth day seemed like eight days. And the fifth day you went to see your mother and that seemed just like a day, and then you came back and later on the sixth day, in the evening, when we saw each other, that started seeming like two days, so in the evening it seemed like two days spilling over into the next day and that started seeming like four days, so at the end of the sixth day on into the seventh day, it seemed like a total of five days. And the sixth day seemed like a week and a half. ….”

  • https://evilspeculator.com molecool

    I think you’re on track for the moon base before 2015.

  • https://evilspeculator.com molecool

    I’m sure there’s space in one of your closets.

  • https://evilspeculator.com molecool

    I’ll explain today.

  • Scott Phillips

    Agree

  • strider

    I didn’t word that right. I meant thanks for sharing fractal monger.

  • mugabe

    is it hard to keep the emotions in check? (serious question)

    PS Would LOVE to have that problem!

  • Scott Phillips

    Not really. Brett Steenbarger talks about how 90% of the mental problems of trading come as a result of bad trading practice. Emotions are still important, but because I trade correctly every time no exceptions it is easier.

    Why are there no books about the mental game of brain surgery?

  • Skynard

    Hum, if support holds the bulls may have one more hard knock left. Shall see, have a look at /SB. Woooohoooo

  • https://evilspeculator.com molecool

    ¤ø„¸¸„ø¤º°¨¤ø„¸¸„ø¤º°¨

    ¨°º¤ S H A K E N ¤º°¨

    ¸„ø¤º° B A K E !“°º¤ø„¸

    ¸„ø¤º°¨¤ø„¸¸„ø¤º°º¤ø„¸