|
PureBytes Links
Trading Reference Links
|
Peter Zwag wrote:
This is a massive topic...
Why do you guys want to re-invent the wheel?
I don't think its so much that new traders want to re-invent trading,
but traders have differing requirements, differing capital,
differing technical understanding, different risk aversion,
different time availability etc. etc.
There are people who are systems designers by profession.
While I admit much of what's out there is not good, there's also
a number of good things with broker statements to match system trades.
I think most peoples concern with a purchased "locked" set of trading
rules, is the rules are never revealed,
nor often is even the logic behind the rules. I for one have an issue
entrusting my money to a rule set I cannot
view and/or edit.
Fortunately the internet with its increasingly sophisticated search
engines is allowing people to swap notes
on software, courses, "gurus" with an efficiency never seen before, I
would personally like a rating system
introduced where software/courses/gurus were ranked against a common
criteria, this would eliminate a lot
of the lies that perpetuate the trading "industry".
I wouldn't be the first to suggest most of the "classic" (1929) trading
books were released by failed
traders looking to recover losses. It seems nothing short of an entire
industry of charlatans has developed
to prey on the innocent trader, from poor quality data feeds, over
priced software, books, seminars,
endless "gurus" etc.
Good luck to you designing systems. While the end system might be simple,
the task of thinking it up is not easy at all. Believe me, cause I've been
at it for many years.
I think almost everyone on the list is in the same boat.
Personally, if your going to spend your hard earnt money, spend some more of
it and lease or buy something that works.
I would consider a system I could view the code of (under NDA etc) and
get to understand, I have issue
with the fund manager approach, especially if the fund manager is a
"black box" inside my PC.
Better to do this and give it to "the market" I think that's stats if 90% of people loose most of their initial capital in 6 months trading futures.
It's the classic fund manager argument, and for the most part it is
true, but I would re-phrase it more plainly
to read "A professional fund manager has less of a chance of loosing
your money than you do".
In short, this is what I feel you should be looking for in system trading
1) large sample of trades. At lease 1000 trades in sample size.
the more the better
2) Long period of time in sample size. Preferably the full length of
whatever contract you trade
3) Have sufficient capital. 2 or 3 times the max drawdown
4) A profit factor of at least 1.5, preferable 2 or more ($win / $loss)
5) A $return / max $drawdown of at least 10
When it comes to fund managers, I would look for 10+ years history of
the fund, this in my experience
sorts the men from the boys. It's to easy to mold results to fit a set
of known data, just look in the back
of Futures magazine.
For systems, longer term results with broker statements are a minimum
requirement. On top of this you
need to know what the data requirements are, a decent data feed
(Reuters/Bloomberg etc) is ~ $25k/yr
plus if your recording a lot of tick data, your quickly going to be
looking at hardware costing another $10+ k etc.
Happy trading
None of the above should read that I am down on trading, I am though a
little disappointed with the number
of charlatans in the industry that thrive thought the industry's lack of
any common standards for assessment.
Go in with your eyes open!
Justin
----- Original Message -----
From: "Justin Fanning" <Justin@xxxxxxxxxxxxxxxx>
To: "Riley Robertson" <rileyrobertson@xxxxxxxxxxx>
Cc: <omega-list@xxxxxxxxxx>
Sent: Tuesday, September 14, 2004 10:44 AM
Subject: Re: system trader
Riley Robertson wrote:
> Hi List & Justin:
> When you suggest it is the System trader that can make a living,
> do you mean a trader who knows computer programming so well
> that he can write his own systems and use it?
> or a trader who is very disciplined and his attitude+trading habit is
very systematic?
I consider anyone who operates by a set of *written* mechanical rules,
to be a
system trader. Some I have seen employ teams of guys working around the
clock,
more commonly now people are using computers.
> If it is the former, does it mean all the discretionary traders are
losers?
The consistent big winners (in my observations) were all mechanical
system traders,
discretionary traders can hit the odd big win, but (in my opinion) lack
consistency
and are long term losers.
> If so, how can the discretionary traders pay someone to write the
programming for him
> without that programmer using the formula himself and trade with it?
The standard legal framework of Non Disclosure Agreements (NDA's) plus
someone's
professional reputation is usually enough to keep most things secret.
Not to mention the
average man on the street known nothing of data feeds, brokers, exchange
rules, contract
specifications etc. Trading isn't exactly something you can get your
head around in 5 mins.
The use of external programmers is something I would like to do more of
myself,
in fact I would like to sub-out my whole IT operation and concentrate
more on the
core trading.
Often I find myself modifying my trading ideas to fit in with
TradeStation,
I have explored the path of professional programmers before and no doubt
will again.
The freedom to dictate exactly how you want the code to run, knowing it
won't be you
writing it is quite liberating for the mind!
Having said all of this... I am constantly stunned by the simplicity of
trading systems
used by banks & other institutions, often as simple as a single moving
average.
> RR
Justin
|