|
PureBytes Links
Trading Reference Links
|
>Based on the comment above aobut risk not being initial margin, how do you
>define risk?
Amount of money you will lose on trade without impacting your ability
(financial and psychological) to continue trading after incurring a series
of losses in sequence. Amount - at portfolio level - that will keep you away
from margin calls. Amount - at portfolio level - that will help you meet the
margin call should the extraordinary happen - without affecting your
lifestyle.
Take your pick.
>Yesterday Gitanshu talked about gap risk and extrodinary
>events making hard stops ineffective to define risk. Although i understand
>your point Gitanshu, extrodinary risk is just that, Extrodinary. Must we
>plan for the worst in all trades since it will only happen once in a great
>while, or shouldn't we use portfolio diversification to lessen the
>probability of catastrophic loss?
I'm beginning to think this is getting theoretical... but one could,
theoretically, find a market without gaps and trade it. I think the returns
would be dampened since everybody knows there's no risk so why should there
be a risk premium, price volatility in your favor, etc.
I presume we're still talking specific to equities, so the comments below
are in that context.
Portfolio diversification:
Different traders have different ways - some don't diversify, and trade only
one market and one instrument in that market. They have to be very good at
what they do. Turning points, chart pattern recognition, trade management.
Some trade only one chart pattern but different trade sizing into taking
positions from that chart pattern.
That's some of us.
I achieve it by using strategy diversification. But it stretches the
available capital already.
Portfolio diversification is not how most of us trade since most of us are
undercapitalized relative to where we want our account to be - and need to
concentrate resources on 2-3 market/trade basis - nor do we have the breadth
of knowledge or the resource support needed to trade multiple markets large
sized positions or broken down positions like a mutual fund -
simultaneously.
Portfolio diversification increases complexity of account management and
dampens returns while giving us the fool's paradise comfort of thinking that
we are also reducing risk. The risk remains the same - or more - and for
that, we trade in our superior potential returns achievable from
concentration.
We're not a zillion dollar mutual fund to need diversification - most people
on this list trade under $1 million. You put $100k in 10 stock trades,
that's enough diversification already. On each of those $100k trades set
your stop at 2%, $2,000 - and you are a genius if you can catch all 10
entries perfectly and not get stopped out for a significant period of time.
I kind of find it difficult to believe that traders - repeat - traders -
diversify too much beyond that, because the trends are just not there - the
moves these days are magnified within compressed time, agility is the order
of the day and therefore focus is desired, not the machine gun spatter of
pick any stock, it will go in your favor - that existed in Fall 99.
Re: Extraordinary risk offset by diversification:
HOW DO WE KNOW that the "once in a while" is not right after we're in the
trade? How is that answer any different from each trade we take, whether it
is adding to a position or to a new trade while having an existing position?
I like to take care of each position against the fat tail event. I like to
go into a trade thinking I know what I am doing, but knowing that I cannot
possibly know it all and therefore ought to defend my stance against myself.
This applies only to stuff I carry home - and again I think the questions
are theoretical.
Over time I have come to employ a variety of strategies - let me rephrase
that - over time, I have sequentially worked through, used, learned,
eliminated, modified and finessed my choice of instrument and strategy to be
direction-neutral and make money to defined goals as opposed to getting
married to my belief in recognized chart patterns. When a direction happens
in the price action, my position takes me into it. There are times when this
strategy loses money - typically in tight congestions that last over 3
weeks. That is known. Hence I trade some other strategy in equal dollar
amount where I seek and find tight sideways markets with breakout
protection. This sometimes makes me sit out the breakout move from the
sideways market - sometimes, it is the move of the year - but I am so
focussed on preserving the thing on my position sheet that by the time I
realize it is the move of the year, it is over already.
Case in point is the spring crash - I made some money, but I know I
should've made a lot more. The fact that a lot of people lost big money is
meaningless to my P&L - or my ego. Being protected against a crash and
coming out of it unscathed by dollars or emotions is only half the job done.
Thus, strategy diversification.
I have come to view my trading as a profession that will last my lifetime.
It is too much fun to want to do anything else. Therefore I am more
concerned that my capital lasts that long and that I don't screw it up with
one or two extraordinary disasters each year.
I've been there, done that, and don't want to go back/do that. That's my
experience of what it takes - 1 or 2 bad trades a year that are real
bloopers - and it takes 5-6 months to bounce back. It gets compounded
because you trade small on the comeback trail - so even if your win/loss is
good you're making money back slower. But your psychology won't allow you to
trade larger.
Too many 5-6 months bouncing back, and I might as well buy an Index Fund and
find a paying job...
Since this discussion is primarily for directional oriented traders, all I
can say is this:
It takes many many incrementally successful trades (high % won/lost AND high
amount won/amount lost) to build an account of some size. It only takes one
extraordinary event to wipe it all out, especially if you're using leverage.
I like to believe that I do not have the consistency of direction picking,
and therefore like to think that my next trade may be a loser - regardless
of how many accurate calls I may have made in the recent past.
For eg, I know that YHOO put in a textbook key reversal bar yesterday. I go
long YHOO sometime yesterday - for simplicity sake, lets take my entry at
$100.
So my stop would be $98, if I want to risk 2% on the trade.
Looking at the chart, $98 is pretty good to stop myself out at, it is 2
points below the most recent swing low. The last time price traded $98 was
11/15/99, when it broke out on a range expansion bar through $98 on its way
to the $250 zone. Below 98, there is really nothing to stop it until - say -
$60, so I'd rather lose the $2 and tell myself I'm wrong - than lose $30 and
double down.
Question: It gaps up in my favor at $120. Where do I trail my stop? $120 is
above the highs of the last 3 bars on the daily. So going by the concept of
higher lows being the new uptrend, I start by trailing it at the lows of
each of the bars (113, 115 and 121 - rounded) it busted through before the
open today. Moving averages are out, since YHOO is below both the 50 and 200
mas in the $130 area and the 20 is below the 50.
Lets fire up a 5 minute chart. So far it is a trend day, about to hit a
brick wall called the 50 day ma on the daily. Shall I exit? Shall I stay? If
I exit, I will be naked without a position if it busts through the overhead
resistance.
Will it? Its overbought - but 4 days ago it was oversold, and it still
dropped $30.
The distance between my stop and current price is much more than the 2% I
was initially willing to lose when I entered the trade. So should I widen my
stop given the additional volatility? Should I narrow the stop given the
large range? Should I simply let myself be stopped out at the gap's lower
boundary?
How can there be a single answer to this set of events?
Next question:
Let us say you saw YHOO breaking down from the 50 and 200 ma's on a range
expansion bar on 6/22, You went short the next day at the open ($129), you
set your stop $1 above the prior bar's high ($142) and trail it down each
day using the falling 200/50 mas as your stop out. So far you'd still be
short, having enjoyed a $60 round trip. If you used trailing prior bar's
high for a stop, you got stopped out prematurely in the congestion after the
first thrust down and missed the real acceleration down move.
Price accelerated in your favor until yesterday. You are an eod trader,
looking at eod charts and placing orders accordingly. You see yesterday that
price may have made a key reversal BUT the next real support is at $60 - so
you phone in your order to cover if price trades 1 tick above prior bar's
high at 110 instead of the trend following ma at $130. The broker will only
execute in the day session since the broker does not accept night orders
(lets just assume this for the moment).
You wake up and are stopped out of your short at $120 - which is $10 worse
than your protective stop. Your profit went from $24 to $9. Your risk grew
(typical of trend following methods) as the trade went in your favor.
Your reward grew if you overrode your rules and covered at the close
yesterday because of the key reversal bar.
YHOO too much of a bronco for comfort? Lets take slow and stodgy PFE.
Next question: You were also short ARBA because of the same chart patterns.
Now it opens up $25 tomorrow morning. Boom. Two days in a row, your P&L went
from x to x minus significant y.
Forget the internet stocks.
Take stocks of your choice, and post some examples of your questions. Take
some real life cases, put them out here on the list for comments, and see
what you learn. Better yet, why don't you tell us what you would do in the
above examples, or other examples of your choice - given the benefit of
history...
Without practical application, all this is just so much bandwidth on a
charting package's list.
Gitanshu
|