Running With the Bears
"It reminds me of the heady days of Sputnik and Yuri Gagarin when the world trembled at the sound of our rockets. Now they will tremble again -- at the sound of our silence. The order is: engage the silent drive." -- Sean Connery as "Captain Ramius" in "The Hunt for Red October"
For the past five years, my brother and trading partner Pete and I have been developing trading systems and scanners for unusual stock market activity. While the algorithms and processes we go through to ascertain the stocks and options most likely to be impacted by unusual activity have been a tremendous resource in identifying actionable ideas on the bullish side of the market (i.e., mostly in call trading), they have needed a little bit more work to help us better identify bearish activity.
At any given second, up to 120,000 quotes stream from the six U.S. options exchanges. The number of quotes coming from these exchanges has increased exponentially as market makers now stream their bids and offers to the exchanges, and each bid and offer for every call or put changes with the movement of the underlying stock.
To give you an idea of the frenetic pace of options trading, consider this: Stocks like Apple (AAPL), Yahoo! (YHOO) and Google (GOOG) trade 15 to 30 times per second! When you consider that Apple has more than 140 strike prices for its call and put options -- meaning, upward of 280 quotes change for every movement in the underlying stock -- you get the idea that quote traffic on the option side of Wall Street dwarfs that of the underlying stocks traded on the NYSE and Nasdaq.
My firm's computers take live feeds from every securities exchange in the U.S. and freezes those quotes to see whether each trade occurred on the bid price, on the offer price, or somewhere in between. The offer price is key, because it's higher and that says folks want in at practically any price before a big pop happens.
These programs, which we call "HeatSeeker" and "Distant Thunder," then conduct a second-by-second analysis of the data after running it through our algorithms to detect unusual buying patterns in calls. Those patterns have very frequently gotten us into trades made in advance of takeovers and other bullish situations alongside the "smart money" -- i.e., the "big" buyers who start piling into a trade in huge blocks.
We've been on the bandwagon with this "smart money" not because we were privy to some insider information, but because we rode the coattails of these institutional players. However, one frustration for us was that we were stacked heavily on the bull's side of trades because we hadn't found a way to use our metrics on the put, or bearish, side of the Street. Since balance is something that we don't just apply lip service to, we needed to perfect our techniques for bearish plays.
Smart money doesn't just play the upside; they make money in down markets, just like we would all like to do. And now, with our new put-tracking technology, we're catching up to our institutional investor friends no matter which direction they're heading in!
Open Season for Bear Hunting
As I've described, we have the data, computers, servers and the algorithms necessary to reveal the activity of every call or put that trades on any of the six U.S. options exchanges, but our challenge was how to use this technology to find which stocks were ready to roll over.
The puts presented such a challenge mostly because there are more reasons for institutional buyers to pay the offer price for puts. The speculative reason would be what we were looking for (just as it is for aggressive call buyers), as this could tip us off that someone wanted to get in on the short side of the market in a hurry and was willing to pay the offer price to own the put.
Another reason that someone would want to purchase puts on the offer would be for "married put" trades, in which the put buyer simultaneously buys shares of the underlying security. This is actually a bullish strategy that occasionally is referred to as a synthetic call, as it has the exact same risk profile as a naked call purchase.
The other reason is that an institutional investor who owns a large stake in a given stock may purchase the puts as a protective move -- either to lock in profits or ride out a rough patch in the markets. As you might imagine, the purchase of a put against a long stock position is also a bullish trade.
So, after tweaking our program and algorithm to take into account the volumes trading on the offer for puts versus calls, we came up with our first iteration of HeatSeeker for puts, which we codenamed "Red October," although we've been going back and forth between that and Depth Charge. Red October (which is still one of our favorite names for the service!) eliminated more than 80% of our false positives, but more adjustment was necessary before we were ready to battle test our "bear hunter."
'Put'ting Our Bullish Indicators to Work
No matter what we'll end up calling our service (i.e., Depth Charge, Red October or any other names that come our way), it is what it is: a way to play the options markets on the downside. Below is an example of bearish activity that our new program discovered.
Xilinx Inc. (XLNX)
After considerable back-testing of what came to be known as Depth Charge, we shared our first unusual put-trading activity report with the public on June 7, 2006. The Xilinx (XLNX) June 25 Puts were storming across our trading screens, and they seemed like a logical choice to use to debut our system.
Here were our notes from that day:
"Shares of the maker of programmable logic solutions are down 39 cents to $25.65, but there are storm clouds on the horizon, amigos! Our computers show 8,147 of the XLNX June 25 Puts changing hands versus an open interest of 2,927 contracts, and this tells us that someone -- a whole lot of someones -- are betting that XLNX could pull a move similar to the one it pulled in October 2005 when it announced sales significantly below Street estimates. That fiasco had sent shares of XLNX down 15%.
"The XLNX June 25 Puts are trading up 15 cents to 45 cents ahead of tomorrow's mid-quarter update. The average daily put volume last week in XLNX was 1,910 contracts, which further places today's action in perspective.
"We've simulated a $2 rally in XLNX on our models, and the puts drop to 10 cents -- and that's if these buyers of these puts are dead wrong! If they are only partially right and XLNX drops $1 to roughly $24.50, then the puts double and that's just a 3% drop, not 15%!"
The next day, those 45-cent puts ran up to 70 cents, where we recommended stepping out of half of the position. The stock ended up taking a turn for the better, though, when JPMorgan released some positive comments that, unfortunately (for us!) gave the stock a boost. The puts slipped a bit later in the day to between 55 cents and 65 cents, where we recommended closing the remainder of the position.
Not bad for a debut trade, and definitely a solid foundation to build upon.
We've made minor changes to Depth Charge and will continue to make changes to our metrics for unusual activity on both the put and call sides of trading. I've been quite pleased with how well the readings have done and how well they have been received.
Copyright 2011 Minyanville Media, Inc. All Rights Reserved.
Daily Recap Newsletter