website header 2

website header 2

Monday, February 13, 2017


“Greed is not a financial issue; it’s a heart issue.”

Very few things in life allow you to be “close” and then be considered extremely successful; from a mathematical standpoint, it’s the “degrees of freedom” in the dynamic variables of throwing horse shoes or launching/tossing grenades in the field of battle, that allow the participants to be successful. Being close with a bullet, fired upon your enemy, is apt to get you killed.

Successful trading is so very difficult for the masses for the following 6 reasons; 1) absolutely nothing in life and/or educational experiences prepares you for what is necessary for success, 2) very few people actually understand probability theory correctly and are easily swayed and fall susceptible to the phenomena known as the “gamblers ruin”, 3) either “don’t want to know” or no have “no idea” how global markets work, who the players are, what “big money” [banks, hedge funds, & commercial dealers] can and will do to a market, and how volatility [not price] dictates how professional’s trade, 4) have an incorrect perception of the purpose and function of the mainstream financial media [CNBC, Bloomberg, Fox Business News, Reuters, etc.] and fall prey to “group think”, 5) have absolutely no idea what the multiple limitations or “downfalls” to their trading model are; whether its “seat of the pants”, trading news, perceived fundamentals, or a spaghetti like  conglomeration of traditional technical indicators. All they got is the “sizzle” of trading, with no steak in sight, and last but not least 6) even if you got the first five above covered, it doesn’t guarantee anything!

Which brings me back to my trading premise of successful trading; allow probability theory to get you “close” to the start and end of moves, with “degrees of statistical freedom” for maximum trade flexibility, that will almost insure … almost … maximum profit per the day’s trading [over a series of trades] given a certain lower threshold for volatility. If you get the necessary lower level of intraday volatility or exceed it in a 6+ hour trading session, and you execute your algorithm signals, your probability of profits [success] for the day is extremely high.

So, since everything depends on volatility, let’s explore 3 sets of volatility statistics; 1) GOLD ETF VIX from 2008 to the start of 2017 [source St. Louis Fed] to get a “big picture” look at gold, 2) traditional VIX figures [Source: CBOE] for the last year [2016] to get a little closer “time wise” to the present, and 3) historical intraday gold volatility over M5 intervals [Source: Serksnas Masters Dissertation, University of Oxford, 2013]. Directly below in succession, the 9 year look at gold volatility, the last year in gold volatility, and what we can expect in the very short term [intraday] from the M5 data in gold futures.

[Note: Click to enlarge charts]

The first and second charts show current levels of the gold VIX [via the ETF] to be very low currently; right around the 15% level from day to day, while at the same time telling you 2 important things; 1) don’t expect gold VIX to get much lower, and 2) it can and most certainly will get a whole lot more crazy [up] going forward.

In the third chart 150 = 12:30 GMT, and you can extrapolate times forward and backward; one striking feature of this dissertation proves something I have premised forever, namely that as you move towards the close of trading for the day [150 towards 300] liquidity dries up and moves in the market become less likely, thereby reducing the probability you will get a nice move in your profit direction to save your trading day. Start early, finish early and leave the game for the pigs to fight over the scraps!

I mentioned earlier statistical “degrees of freedom”; simply put, it’s the number of ways a dynamic variable can change in a final statistic without changing the outcome of the system. In pure math parlance, degrees of freedom = the geometry [dimension] of vector sub-spaces, but there’s no need to get into this so as to glue your eyelids together and get you mumbling to yourself something like, “hey, nobody said anything about math here, OK? … I’m outta here”!

Bottom line is that incorporating “degrees of freedom” into the trading algorithm [scalper’s] gives us a more dynamic approach modeling market behavior than a static “round peg, meet square hole” approach which only works under optimal circumstances and can lead to some ugly series of trades. So, what’s our big asset in terms of “degrees of freedom”? Simple: it’s the fact that trade entry is 3 or 3+ M1’s contra trend; therefore, we can get buy/sell signals with 3, 4, or 5 M1’s instead of a fixed number. Usually, anything past 5 will put the price vector fields “out of sync” [not going in the same direction] and negate the trade, and that usually is the time the market is ready to either roll over for a longer correction or change short term trend.

Inherent in the construction of the gold algorithm [scalper’s + regular; the first defines entry, the second allows the “free trade”] is the fact that nobody can identify consistently the day’s high and/or low; so why attempt to do something that is not possible? I know only a few things with certainty as I come to my screen to trade each and every day; 1) it’s going to go up and down during the trading day, 2) I got 6 … maybe 8 hours [tops] … to make money before the market most likely dries up and gets quiet, and 3) the algorithm as developed is a whole hell of a lot smarter than I am at deciding whether to be long or short at a specific time within a short term trend.

Ultimately, though, the algorithm’s main goal & premise is to spot those trading opportunities where the concept of “buying at 1, and selling at 3” are present, and have a very high probability of success; not a guarantee. “Just because you see clouds in the sky doesn’t mean it’s going to rain”. Remember always one of the central theorems of trading, which is just because A = B does not mean [≠] B = A; trading is not associative or commutative like elementary arithmetic.

Now, if you think the aforementioned above is a challenge, wait … here’s the real challenge … “using the heart of the New York trading day to determine when to ‘walk away’ when you are at or have exceeded your profit goal” … cuz now you’ve just wandered into the non computable arena where probability wave functions become excruciatingly complex and unsolvable.

In the final analysis, I’m not competing with the HFT’s, trying to steamroll commercials, front run stops, or guess what’s going to happen; I’m simply allowing the volatility in the marketplace to “work” and give me my opportunities within a statistical reference point inside a short term trend. It doesn’t mean I have to take every signal; the algorithm allows for maximum flexibility for just about everybody. But what it does do via signals is give me a very high probability of success trade, and when combined with other signals and trades, gives me the best chance of walking out the door at the end of the day a winner; and that’s what this game is all about.

Turning to today’s market … very low activity overnight as China comes back from Golden Week Holiday … I pointed this out the other day, Asia for the most part is now unwilling to push gold higher almost every night … is it a “buyers strike” or have they finally … finally … figured out that they are being played like a violin by New York bullion dealers and simply waiting for said dealers to push it lower in New York for better pricing? … if it’s the latter, Oh are we in for some volatile times as lines are gonna get drawn in the sand and somebody is going to have a bad case of “butt hurt” going forward.

Here we are about an hour before the NY open, and I’m trying to remember a non Holiday trading day where market activity has been this quiet; absolutely zero on the economic docket for release today, so I have no idea what drives this past the $4.40 range for the last 12+ hours. At some point here, things are just going to explode.

Over an hour into NY trading, and to some it appears there have been algorithm signals to make; not so, as the first rule of trading is we have to see an HVALUE for the day greater than $5, and here at about 8:40 EST we finally see the HVALUE go above $5. If you trade before this threshold, you are begging to get whipsawed from reversals, as there is a much higher probability of a reversal day when the current HVALUE that is below $5 becomes the LVALUE quickly and leaves you with a losing position.

But it isn’t just reversals; when you get a market that is “coiled”, and the HVALUES are very low, it doesn’t take much of anything in terms of a price movement on the M1 to get the price vector fields to shift and thereby cause false positives. Here in gold, in the scheme of things, what’s a 40 cent move mean on a normal or above average volatility day? SHORT ANSWER: “Nada”. You can’t give the same answer when the HVALUE is below or hovering around $5. And as we all know, staying clear of “false positive” signals when things are slow to crawling is a key test of your patience and discipline. “Don’t do stupid poo poo”!

Finally, first trade of the day directly below.

I’m really not liking this kind of trading action; nothing for minutes on end and then BOOM! price lunges lower quickly before it’s a rinse & repeat operation; not willing to let this short trade ride at these levels, and therefore treat it as a scalp.

Examples 2 & 3 directly below were basically scalps with a few pennies to show for it; quite frankly, this is one of the worst action days I’ve ever seen in gold. There just isn’t any up/down action here; it’s nothing but the dealers versus the specs, and I don’t care what happens from here on out, I don’t want to be in this going into the NY afternoon.

Beach beckons … I’m outta here … until tomorrow.

Have a great day everybody!



No comments:

Post a Comment