Every week Floating Path looks to highlight some of the crazier examples of HFT running amok in the markets. We typically do so with the help of Nanex which monitors, analyzes, and visualizes high-frequency trading market data.
The past few weeks we’ve seen more and more evidence of cracks in the infrastructure (or microstructure) of the stock market and its thousands of miles of fiber optic cables and servers. The three hour outage of the NASDAQ OMX Group, Inc. (NASDAQ:NDAQ) and the system-wide option market blackoutare telling scenarios of just how stressed and vulnerable the fragmented markets have become.
On September 16th this was also the case in the trading of Cisco Systems, Inc. (NASDAQ:CSCO) and the resulting issues from two exchanges.
The move was so explosive that it caused trade reports from 2 exchanges (Edge and Edge-X) to disappear, then reappear suddenly, 2 seconds later. This is an indication of a network failure somewhere between the exchange and input to the SIP (securities information processor, or consolidated feed).
More Fed Leak Data
The blatant leaking of the last FOMC decision has made headlines throughout financial and even mainstream media. The charts below show the cumulative value of stocks and futures traded in the first 100 milliseconds immediately after the news release and also compares this event with the past two FOMC meetings.
Within a millisecond of the 2pm release of the September 18, 2013 FOMC Meeting Announcement, the stock market exploded, trading nearly $400 Million worth of stock in a tenth of a second (a blink of an eye is 3 times longer), and almost $1 Billion worth in 2 seconds. Over in Chicago, futures trading also exploded, with about $5 Billion trading in a tenth of a second and more than $10 Billion in 2 seconds.
Retail Option Traders Get Hosed
We’ve highlighted before how quote spamming is rampant in the wild option market when the quote to trade ratio in SPDR S&P 500 ETF Trust (NYSE:SPY) reached 14,000:1. The following charts show roughly the first minute of trading in two different Google Inc (NASDAQ:GOOG) puts. The orders executed were placed at the open, yet didn’t get filled when they should have, and the small retail order was stuck with a much lower quality price. In the first scenario, the trader sells for about $1.50 ($150 per contract) lower than he should have. In the second, the retail trader pays about $1.00 ($100 per contract) more than necessary as earlier, better quotes are ignored.