Friday, December 21, 2007

Insider Crimes, Funny Money and Options Rackets

Insider Crimes, Funny Money and Options Rackets
March 22nd, 2007

WARNING: The following is not a recommendation to buy, sell or hold
any financial instrument. Trading stocks, currencies and options
involves serious risks.

For years, I've kept this information to myself, waiting for a chance
to look at it again more seriously and systematically. Unfortunately,
I don't really see myself being able to allocate the time to research
this stuff further.

I know that several Cryptogon readers are traders and/or otherwise
involved with the financial markets. Others of you are good at writing
software. There are several neuroscientists out there and at least one
rocket scientist-type guy or gal! HA. Seriously, it's true. Three
INTJs have written in, in response to the INTJ reference I made in a
previous post. These are the types of people who can nail this stuff
to the wall, to the extent that it can be nailed to the wall.

I thought that some of you might take a look at this information and
have the skills and tenacity to take it further than I was able to.
Maybe figure something out… and share it with the rest of us.

Anyway, here goes.


From around 1998 to 2003, I looked for a way out of having a day job
through trading, stocks mostly. I learned a lot of things the hard
way, but at least I learned. My goal wasn't to get rich. All I wanted
to do was make small profits consistently. Fifty bucks a day was
actually my goal. I managed to do this, and orders of magnitude
better, on occasion, for a while, but it was hard, time consuming and
incredibly stressful. Was trading (intraday) worth it? For me, not
really, but it was better than showing up to some office.

Was it gambling? In a word, yes, but I didn't know that at the time. I
thought that the tools and systems I used were working. Actually, they
weren't working. How did I know they weren't really working? Because I
started to lose money! They had just appeared to be working. I stopped
trading as soon as I lost confidence in the tools. (Maybe the tools
actually worked for a while, but then stopped working? That's
possible, I suppose, but not likely.)

I had always assumed the entire business of the markets to be a
massive criminal enterprise that allowed public participation simply
because it made a few people extremely wealthy. But, I wondered, when
do the fewest people make the most money in the shortest period of time?

During gaps.

A gap is a sudden, violent change in the price of a stock, future,
bond, etc. Gaps are caused by some event that temporarily creates a
state of disequilibrium in order flows. Gaps on stocks started to draw
my attention because I just knew that insiders (and others) were
pre-positioning themselves to profit from those moves.

I noticed that, quite regularly, unknown, crappy little companies
would go up 25%, 50%, 100%, 200% or more in a single session. You see
them almost every day on the top percent gainers board. For years, I
looked at those and shrugged. "Yep. People hit the lotto too," I'd say
to myself.

I watched another one pop, then, finally, I thought: "This is the
scene of a crime for sure."

I'm not talking about legal insider trading by corporate officers that
the financial press follows. I'm talking about the mistresses of
executives and $2000 per hour prostitutes who heard something over
pillow talk, friends at country clubs who get a tip during the
golfcart ride to the 18th hole, industrial espionage operators (former
spooks) who have these idiots under surveillance, low level IT staff
who read the executives' email, not for fun, but for profit, etc. etc.

The point is that LOTS of people who aren't corporate insiders know
the news before the market does. They buy before the news comes out.
This happens every day.

Thousands of symbols. Every couple of days at least one of them goes
up by 25% or more. How would I go about finding them?

Assume maximum criminal behavior and find stocks where the signal to
noise ratio is very high.

I came up with a system where I doubled a few times; I made nearly 3x
once. No losses. I could never bring myself to allocate large amounts
of capital to hold overnight on these "Risk of ruin" bets because I
didn't believe this was really working, even when I was hitting. (All
stock bets are, in my opinion, risk of ruin.) Was it luck each time?

Whatever it was, here's what I did:

I looked for cheap stocks that were flatlining or trading in a narrow
range near their 52 week lows. Then I looked for increases in volume
and churn. Churn is increasing volume with prices staying flat, or
increasing slightly. Under churn, prices stay relatively flat because
the buying is accompanied with selling to keep the price from running
away. There is, however, a net inflow of capital.

This is the critical part: The activity had to be happening in the
absence of any news.

I'd read about Dow theory, so the minute I started to notice these
volume increases, without news, it made sense. What I was seeing was
the accumulation phase on these stocks, where people with inside
information were buying ahead of the good news.

I never looked at earnings. The company's products or services didn't
matter. All I was looking for was that very specific pattern.

The trick was finding stocks with a high enough signal to noise ratio
to spot the criminal insider buying. This nonsense happens on all
stocks, I'm sure, but it's too difficult to see most of the time.
That's why the flatliners and narrow rangers are nice. I picked stocks
near their 52 week lows because, well, for some reason they stopped
there before. Think of this as an up or out proposition. Oh yeah, it
could go to zero and you could lose your money.

Don't use stops on these bets. These are all or nothing gambles. Why?
Because market makers will often drive down the price before the big
move up. These stocks are very simple to manipulate. Just before a big
move, they might run the stops, just under where you've drawn your
support line, to scoop up a bunch of shares at a discount. (This just
happened on one today, as a matter of fact.) Double and triple bottoms
are built on the skulls of people who thought there were being smart
by setting stops down near a 52 week low. If you're reckless enough to
buy a crap company near a 52 week low and are worried about it
dropping more, forget this method. You'll get stopped out, and
minutes, hours or days later, the thing could suddenly double. Company
bought out. Drug approved. Government contract awarded. Whatever.

Why not wait for the fake dip before the big move up? Sometimes there
is no dip at all. Sometimes there is. Again, nobody said it was going
to be easy.

Why not use a screen that only looks for double bottoms AND a
follow-on positive slope OBV on no news?

* scratch whiskers *

Every time I think I've beaten this crap to death, there's always
another weird path to take. I'm no longer in a position to do 20 hour
research binges on this stuff, but some of you young whipper snappers
should try it.

The way to play this is with small bets on multiple setups with
capital that you assume will be lost. As soon as you get a pop on any
of them, sell immediately into the "dumb money" players who show up to
chase the news, which might just be some kind of criminal press
release. Don't try to parse the news. Don't even look at it. Get to
battlestations. Lock and load.

How much you take off the table is up to you. I would typically dump
it all in one shot. But you might consider selling 1/2 or 3/4 of what
you started with and set a trailing stop on the rest. You might also
draw a trend line connecting the bottoms of an on balance volume (OBV)
study to better plan your exit. Upward slope? Let it ride. Cross down?
Dump it. I liked watching this on one and three minute charts. You
just need some way of seeing that more dumb money is flowing into the
thing than insiders are pulling out.

There was no way to know what would happen an hour, a day or a week
later. Sometimes they would go higher or much higher, sometimes they
would crash back down. Once you get out, though, don't even think
about looking at the thing again. Spend your time looking for the next
one, trading sideways, near its 52 week low, with anomalous volume on
no news.

Can't find any? Ask your significant other to duct tape your hands to
the desk so you don't make any stupid, undisciplined trades. Actually,
if you have the time to dedicate to this type of work, you probably
don't have a significant other, or a company has provided you with a
padded room and just slides food and coffee under the door.

If you have access to Trade Station or some other scriptable tool, a
screen could be written to automate the search. [I have no reciprocal
relationship with the TradeStation Group.] FYI: TradeStation is not a
toy. It is the most sophisticated strategy testing and trading
platform available to the public. Any strategy can be coded and tested
using that software. When I was doing this, I was doing it manually,
individually inspecting hundreds of charts.

A note on volume: Screen for stocks with an average daily volume of at
least 25,000 shares. Many crap stocks barely trade at all, so any
activity will skew your volume studies and could give you false
positives. But if you see a dead one with volume picking up out of
nowhere (no news), with the price holding steady (churn), keep an eye
on it. Read about OBV. Then read about it again.

I was doing this on NASDAQ listed stocks in the $1 to $3 dollar range;
stocks colloquially referred to as dog food. I don't see why this
technique couldn't work on the serious pink sheet garbage penny stocks
as well. I mean, what's the difference between a $2.25 NASDAQ stock
and a penny stock? About two bucks.

Another note on volume: Your volume studies are the key to this. If
you can apply two moving averages to daily volume, that will spot the
weirdness. Make your long MA something like 13 periods and your short
(trigger) MA something like 5 or 3 periods. If you get a fast period
MA cross on the volume, ON NO NEWS, that might be something. You're
looking for sustained, anomalous volume. One day spikes may or may not
be something. There's no way to know on those. But when that average
daily volume starts breaking out, that's interesting. Up sloping OBV
on no news is interesting.

What period? Screen using daily closes. There's no need to gear up for
an intraday-style gun battle during the screening process. (TR and I
tried to do it in real time using TradeStation's incredible Radar, but
we failed. We couldn't filter the cranks from the real moves. Our
tools were finding real moves and cranks. Sadly, for us, they looked
the same in fog of war. We progressed to the Magic Mystery Dot, which
was very interesting and weird, but didn't quite work. It kinda
worked, but I nearly pissed myself trading based on its signals.) You
can quietly sit back, look at the daily data and pick these out very
carefully. The criminals need to take their time on accumulation. They
can't do it too fast or the price will go up, or smarty pants people
with their volume studies on their radar screens might spot it. The
setups are usually slow and steady, they happen over days or weeks.

Once you're in, just set price alarms using some tool and don't think
about them again. Initially, set your alarms somewhere above the top
of recent ranges that the stocks have been oscillating in, or just
above the flatline if it's really just trading sideways. If the price
breaks out on no news, continue to hold; this could just be a
temporary crank that will fade fast. If the breakout occurs on the
news, dump it, even if you're only making cigarette butts and bottle
caps. A weak pop on news just means that there aren't enough suckers
willing to take the bait. How do you know it's a weak pop? What's the
slope like on your one and three minute OBV charts after the news?
Flat to down? Buddy, you're juggling hand grenades with the pins
pulled at that stage.

Also, because these moves may involve very serious criminal activity
you want to get clear of the pig before the exchange halts trading and
the SEC gets in on the act. By the time the authorities are parsing
the news and taking note of the order flow, you want to be outside in
your hammock, sipping a cold beer.

Again, nobody said it was going to be easy.

Note: The above strategy should work on the short side too, in theory,
but I have only ever worked the long angle. Why? To play this short,
there will be more noise, you'll have to play with more expensive
stocks and to not use stops when short could result in unthinkable
disaster. You could potentially lose more than your initial bet.


FOREX is the best pure gamble now, for intraday action, but you have
to sit there for hours and just be able to react to momentum/panic
moves on news. FOREX is for people who have the temperament of a
sniper. Most of it is boring to the point of being unbearable.
Finally, unexpectedly, you'll get window of opportunity, lasting
seconds, to take one shot.

That's FOREX. I've only used a realtime simulator for FOREX and never
traded with real money. I tried all of my old daytrading tricks (multi
period stochastics, MAs, Ichimoku, trend following, etc.) and none of
it seemed to work for me. What looked good, though, was a straddle of
USD/CHF and EUR/USD just before news.

USD/CHF and EUR/USD move in opposite directions almost all of the
time. Get in long (or short- I don't think it matters) on both sets of
pairs, just before news. Once scheduled economic news hits the market,
there could be a violent reaction one way of the other. Immediately
dump the position that's losing and let the winning pair ride for some
number of PIPs. Were talking seconds to minutes. Get out. Turn off
your computer. You're done for the day.

FOREX is probably the best venue for news driven action. You will make
or lose the most money on FOREX. It's easily the noisiest, whip sawing
fake out thing I've ever seen, but sometimes, those pairs just run and
keep on running one way. Outside of news, though, that thing is a crap

You can experiment with a decent real time trading platform at [I have no reciprocal relationship with fxcm.] You just run
the thing with a practice account that is loaded with 50,000 pretend
dollars. It's totally free.


I actually wrote this as part of a comment. I'll just reproduce it
here. I initially named the market maker, but I'm removing it now. It
doesn't matter who it was. Why? Well, that firm frightens me. There's
no way to know what they're capable of doing. I'll just call them Big

One day, things were looking good long on Amazon. We got a bit of a
pullback during a multilane uptrend, lots of support. I got in heavy,
1000 shares. It wobbled around a bit, then the futures took off. Up
and away.

I thought, "Wooo hoooo!" this is going to be a good one.

Buyers pouring in!


What's this?

Big Firm is on the ask.

HAHA Big Firm is up to his old tricks, that guy! Not going to fool me…
* hands quivering at that point *

Futures rising. The Nasdaq 100 is climbing. It's ok. It's ok. Stand
fast. Stand fast. (I was viewing the battlespace on three displays.)

Big Firm is selling AMZN. He's keeping the lid on Amazon right now. Is
he just trying to front load shares for himself before the pop??? What

Actually, the price is dropping. The support on the bid side, outside
the spread, started to disappear on Level 2.

"F*ck this shit man." I sent a sell limit order outside the spread.
(This is a way to jump the line that forms at the exits when the
theater is on fire.) I was lucky, someone took it at the market for a
loss of only ten cents or $100 on that 1000 shares.

Seconds later, the long red candles started dripping down my
multiperiod charts. Time sales cascaded red with large block sales. *
my hair is standing on end even now remembering this, I nearly got
decapitated on this one * The idiots in the chat room spat out things






Big Firm single handedly moved Amazon down nearly $2 in a couple of
minutes, through hourly and daily supports. Well, not single handedly.
They just used their deep pockets to frighten lots of traders into
dumping. And the move took on a life of its own. I just sat back and
watched it happen, in awe…

Why did They do it?

Someone told the chatroom I was in that it was about options
expiration, and that Big Firm had obviously bought a bunch of out of
the money puts that were about to expire worthless.

But guess what? Instead of expiring worthless, Big Firm just drove the
price of the stock down until those options were in the money! I
watched this happen. Amazon did that move on its own, without follow
through from the futures or other tier 1 stocks.

As usual, don't take my word for it:

Stock Price Clustering on Option Expiration Dates

This paper presents striking evidence that option trading changes
the prices of underlying stocks. In particular, we show that on
expiration dates the closing prices of stocks with listed options
cluster at option strike prices. On each expiration date, the returns
of optionable stocks are altered by an average of at least 16.5 basis
points, which translates into aggregate market capitalization shifts
on the order of $9 billion. We provide evidence that hedge
re-balancing by option market-makers and stock price manipulation by
firm proprietary traders contribute to the clustering.

That was it for me. I could think of better things to do than dog
paddling in a shark tank.


Maybe this could be visualized in a way (heat chart) that would show
where the largest clusters of out of the money options were sitting.
This would have to be a real time screen of the NASDAQ 100. (Forget it
on smaller stocks.) You're looking for stocks that have large pools of
options that are out of the money and for stocks that begin to move
toward those option clusters IN OPPOSITION TO THE FUTURES.

How do you play?

If a large cluster of options is below the market price, the futures
are rising, and the tier one stock is wobbling or starting to fall,
buy puts somewhere down there. If the large cluster of options is
above the market price, the futures are dropping, and the stock price
is wobbling or starting to rise, buy calls.

The fact that the stock starts moving toward those options in
opposition to the futures shows you that a powerful market maker or
two might be refusing to let their options expire worthless.

If you knew where the options cluster was, and, therefore, the
motivation behind the bullshit crank of a move, I suppose you could
play the stock directly.

Of the three dangerous trading schemes above, this last one would be
the most difficult to program. It might also be the most profitable.

Well, guys, that's it. I need to go build a bean frame and collect
some cow manure.

No comments: