HFT

When everyone was throwing computing power at the 'momentum igniting', "can't lose", HFT-driven algo-trading world, it is perhaps little wonder that the opportunity to expropriate profits from an ever-decreasing pool of real money traders dwindles like a convertible arbitrage hedge fund in the 90s. Perhaps it was the writing on the wall we noted in February when GETCO's money-printing machine was reduced to pennies, but it seems the world of high-freaks is tearing itself apart. As the WSJ reports, two of the largest independent US high-frequency-trading firms are in early merger discussions - as a downturn in trading opportunities has spurred cutbacks. Of course, the firms claim this is a positive, "we're in accelerated growth mode, both organically and inorganically," but as one HFT analyst notes, "HFT is a volume business... If trading volumes are going down, that means it's tougher for prop shops to make money." The slowdown has forced some high-speed specialists to leave the business and with RGM and Allston (the two firms) having 280 staff, we suspect this week's initial claims may be bloated with PhDs.

To summarize, the SEC which admits it was clueless in analyzing the modern, fragmented market (yet which found definitively that the culprit for the May 2010 flash crash was Waddell and Reed, and nobody else, using what technology at the time, nobody knows), uses a platform developed by High Frequency Trading firm Tradeworx... to reach a conclusion that High Frequency Trading firms are innocent of every flash crash resulting from an HFT algo gone haywire...

"I am going to hit on some of the landmines that you can encounter within order-matching engines, and then I am going to give a forecast on, at least from my perspective, what’s going to happen over the course of 2013"

It started as your everyday hexagonal discussion on CNBC with the anchors up-in-arms over the fact that (shocker) some firms can pay for early access to critical economic data items. The disdain for the 'rich' was palpable as Bernstein, Sullivan, and then Cramer all exclaimed both their amazement and surprise that this was even possible. That was when Santelli stepped into the ring and explained - in what was a relatively well-behaved exclamation - that not only was the fact that early data releases were well-known to every real trader (as opposed to those who pretend for TV) but that the issue was absolutely not about 'early access' but about HFT. When we first brought the perils of HFT to the attention of the broader trading community in 2009, it was the stuff of conspiracy theory - but now (as with many other things) it is conspiracy fact and in a few short minutes, Rick Santelli showed off his co-hosts ignorance of the real market and opened many new eyes to the damage that HFT can do in a market that is, well, anything but Reg-FD fair and balanced to all.

Trading has a few simple rules - do the opposite of Goldman's Thomas Stolper; don't fight the Fed; and buy low, sell high. However, as this series of charts from Nanex shows, it is the latter rule that is the easiest to comprehend and yet - thanks to massive and obvious HFT manipulation - is an extremely difficult thing to do. As Nanex's Eric Hunsader notes, high frequency trading algos do not get much clearer than this as the machines buy low (from you) and sell high (to you) each and every millisecond of the trading day.

Bravo again to Jim for his expert work in helping people make money, just not the people he claims, not his viewers, another P.T. Barnum Show folks.  Good thing this one hasn't shown he is running a business taking subsidies from the Conneticut government based on the number of employees he has, hat's off to you too Keith Mccullough.

This was one helluva week. Nevertheless current markets are still hooked on QE.

On Monday we brought to you proof of a 15 millisecond frontrunning of the Mfg ISM number by what turned out to be HFT clients of Reuters which admitted subsequently it had "inadvertently" leaked the number to select clients. However, that was child's play compared to the absolute market farce that happened today which we can visualize courtesy of Nanex, and which impacted gold, ES, and Treasury Futures altogether.

Back on Monday, following the huge miss in the Manufacturing ISM, in collaboration with Nanex, we exposed yet another instance of blatant headline data frontrunning in "15 Milliseconds Of HFT Fame: Watch Today's Early Leak Of The ISM Print" where we showed aggressive trading amounting to tens of millions in notional contracts ahead of the 10am release of the key economic indicator. We assumed that just like every other lament about a market that is front-run by those "who have the means", manipulated (by the Fed of course - remember when that was just a conspiracy theory: good times) and simply broken, it would disappear in the ether forever. After all: why bring attention to facts when hopium is sufficient for the E-Trade baby to retire rich and famous before it has hit 2.  We were delighted to learn that CNBC's Eamon Javers picked up the torch and actually did some further investigating, which in turn led to an actual admission out of Reuters that it "inadvertently" sent out the data to "a select group of high frequency traders, many of whom immediately traded on the information before it was available to the wider market, CNBC has learned." Inadvertently? The humor just never stops.

Worried that manipulated official data is the only thing one has to "predict" on a day to day basis in a world drenched with "Baffle with BS", where China expanding and contracting at the same time is perfectly normal, and where Chicago PMI soaring by an 8 sigma beat to multi year highs precedes by one day the lowest US manufacturing print in 4 years? Turns out that's not all - in addition to everything else, one should also realize that key market moving data continues to be disseminated ahead of its official release time to those who have the "funds" and the interest in trading on early leaks. Take today's key economic data point: the Manufacturing ISM. As Nanex shows, trading in SPY exploded at 09:59:59.985, which is 15 milliseconds before the ISM's Manufacturing number released at 10:00:00. Activity in the eMini (traded in Chicago), exploded at 09:59:59.992, which is 8 milliseconds before the news release, but 7 milliseconds after SPY. Surely someone decided to perform a massive headfake and like a plunging goaltender during a penalty kick just happened to guess the direction right. That, or the clock on the CQS tape is just a little off. Oh, and this is merely today's example of early distribution of data to those who have the means(and the funds) to trade on it. Everyone else - well, the saying involving a sucker, a poker table and confusion, is quite applicable right now...

Ever feel like you can't put that math PhD to good use anymore and make money scalping ahead of order flow, sub-pennying and frontrunning retail in normal and dark pool markets because volumes are just off 1929 levels? Then the Chicago Fed has an offer you just can't refuse. And since money printers can't be choosers, the Fed may also have a spot for those who tried their hand at the New Media (i.e., churning slideshows): "Develop presentations and clarify complex issues for broad audiences." Yet what is most interesting is the following requirement: "Interact with highly informed and technically skilled outside stakeholders while preserving the reputation and credibility of the Reserve Bank." We'll just let that one slide...

•     550,000 SPY shares
•     10,000 June 2013 eMini futures contracts
•     1,400 Nasdaq 100 futures contracts
•     800 Dow Jones futures contracts
•     350 Russell 2000 futures contracts
•     125 S&P 400 Midcap futures contracts
•     300 Crude Oil futures contracts
•     900 Dollar Index futures contracts
•     800 Gold futures contracts
•     10,000 10yr T-Note futures contracts
•     2,500 5yr T-Note futures contracts
•     3,500 T-Bond futures contracts
•     5,000 Eurodollar futures contracts
•     750 Japanese Yen futures contracts
•     600 Euro futures contracts

...understand the national threat that is our fragmented and perverted equity market microstructure that is driven by such esoteric order-types such a Post No Preference Blind Limit Order created through the buddy system of exchange/order volume producer.

The longest ongoing government "sting" operation against a hedge fund, possibly in all of history, that which absolutely everyone has known about for years now i.e., against Steve Cohen's SAC and its Bernie Madoff-esque series of profitable years (at least until recently that is, when "expert networks" no longer accept any calls originating out of Connecticut or New York), may be coming to an end, following what the WSJ reports may be an imminent filing of criminal charges against the hedge fund. "U.S. prosecutors are considering possible criminal charges against SAC Capital Advisors LP as a result of the government's insider-trading investigation of the hedge-fund firm, according to people familiar with the matter.  It isn't clear what led prosecutors to warn the Stamford, Conn., hedge-fund operator that it could be charged criminally. But the move is the strongest sign yet that prosecutors and the Federal Bureau of Investigation are trying to ratchet up the pressure as a five-year deadline looms to file the most serious charges related to trading that allegedly involved Mr. Cohen."

In a world in which fundamentals no longer drive risk prices (that task is left to central banks, and HFT stop hunts and momentum ignition patterns) or anything for that matter, it only makes sense that the day on which Japan posted a better than expected annualized, adjusted Q1 GDP of 3.5% compared to the expected 2.7% that the Nikkei would be down, following days of relentless surges higher. Of course, Japan's GDP wasn't really the stellar result many portrayed it to be, with the sequential rise coming in at 0.9%, just modestly higher than the 0.7% expected, although when reporting actual, nominal figures, it was up by just 0.4%, or below the 0.5% expected, meaning the entire annualized beat came from the gratuitous fudging of the deflator which was far lower than the -0.9% expected at -1.2%: so higher than expected deflation leading to an adjustment which implies more inflation - a perfect Keynesian mess. In other words, yet another largely made up number designed exclusively to stimulate "confidence" in the economy and to get the Japanese population to spend, even with wages stagnant and hardly rising in line with the "adjusted" growth. And since none of the above matters with risk levels set entirely by FX rates, in this case the USDJPY, the early strength in the Yen is what caused the Japanese stock market to close red.