|
|
| |
|
Kane Trading
on:
Simulation
Trading
I decided not to start out this article like
many of my articles, with the big red warning about it shaking up your beliefs.
I've seen a lot of thoughts expressed on this topic, and I find it just about
split for and against simulated trading. It's not quite 'racy' enough for me to
call some of the currently held beliefs 'myths', or to say the topic is
controversial, so I left out the warning. We all know I love to lean on
'myths', but I'm not going to play something up just to make it more
interesting, or whatever. I feel this is just a good, solid topic that I get
questions on, and I have some thoughts on it.
I'll start with a brief explanation of what I
am talking about here so we can all be on the same page, and then I'll get to
the pros and cons, or perhaps a better way to look at it, the limitations.
Simulation trading is simply using a trading engine simulator to duplicate, as
closely as possible, real trading. It usually has the same interface, if it is
a reasonably decent simulator, as the actual trading platform, and it will
start the user off with a simulated account value, preferably one where the
user decides on the initial starting amount.
The level of sophistication varies from
product to product, and some are quite good. I suggest one that factors in a
realistic commission, and some of the better ones I have seen designed the
engine to not only account for potential slippage, but also 'market impact'. If
you place a simulation trade to buy 10 contracts and at that instant there are
only 5 available on the ask, the trading engine can see this and jump the
remaining simulated fill up to the next level, where it assesses conditions
there. I will briefly discuss this more later on when we look at
limitations.
In the 'old days' before simulators were readily available this was
called 'paper trading' because it was done by hand, on a piece of paper. The
paper trader would write down what was thought to be a fair price for entry,
exit, account value after that, and so on, using the honor system. It wouldn't
make much sense to cheat, since one would only be fooling oneself. All in all
it wasn't too bad, unless one were daytrading the mini's or whatever with a
very short time horizon. Again, I'll discuss this a bit more when we get to
limitations. The point is, simulation trading is, for my purposes, just a much
better, electronic version of paper trading that also gives one 'live platform
experience'', too.
Now, why do this at all? Why not just trade real money right off
the bat? There are two schools of thought on this (actually there are more than
two, but I see two main camps), and all I'll do is present my thoughts on this.
One schools of thought is that you make a logical progression from formulating
a 'Trading Plan' with testing on a simulator, to trading very small money, to
trading increased amounts of money as the plan is proven in real time, actual
trading. The other school of thought is that simulation trading is essentially
worthless, and only trading real money is of value, be it small or not small
amounts. Where does Jim's thinking fall in this discussion?
I am fully and
solidly in the first camp. I am 100% for simulation trading, but I am for it
fully aware of the limitations it has. It is yet one more tool a trader can
use, and like any tool, it can't do everything, but it can be used quite
effectively for specific purposes. It is a tool I can't imagine not taking
advantage of, both for new traders and for experienced traders. Some of today's
simulators are so good, it can be quite a tool even for advanced work. Let's
take a look at some thoughts on this.
Those that say simulation trading is
worthless frequently cite that it doesn't take into account the psychological
issues of trading, the stress or effect of having real money on the line. You
know what I say about that? I agree. But does it then follow that it is useless
as a tool? They say the results won't be 'realistic'. So, guess what I say
about that? I agree. But, again, does it follow that it is useless? You see,
that is where I seem to differ from most people. I just don't seem to draw the
same conclusions that most draw.
Let's look at a hammer. I see it can help me pound nails
quite well, so I'm using it. Then my buddy comes by and points out that sure,
it's a great hammer, but I can't use it to screw in a screw. Oh, Geez, I didn't
think of that. I better get rid of this hammer, it's quite useless. Who would
conclude that? No one I know. Yet, with so many things, in this case simulation
trading, and with so many other things I have seen, that is exactly what a lot
of people do. I'm critiquing critical thinking here. Be very thoughtful, and
cautious, about the conclusions you draw. I feel a person's chances for success
go down, probably quite a lot, if they don't use good, solid critical thinking
when developing their 'Trading Plan', or a plan for any business.
Let's look
specifically at what I think simulation trading is good for, and not good for,
and what some limitations might be. When people start out and they are trying
to develop a 'Trading Plan', a methodology, they are really, in most cases,
just starting to walk, so to speak. As I have said in many of my writings, it
takes a lot of work, over time, to try to develop a working plan. This is not a
'get rich quick' profession. It is in many ways, as far as effort and time
investment, like any other tough profession such as being a doctor, commercial
airline pilot, professional athlete, and so on.
Now, if I told you I was studying
to be a commercial airline pilot and I spent the last x period of time training
in a flight simulator you'd probably say 'cool'. If I said I spent the same
period of time training in a trading simulator why would one say that had no
merit? It just doesn't follow. We all know that in a flight simulator it isn't
the same as real flying. But it is an interim step, to develop skills and be
sure you won't just immediately crash the plane. It has value, but only in the
context they use it in. I feel trading simulation also has a similar
place.
One
of the things I see when I observe new traders is that they trade way, and I
mean way before they have a complete, comprehensive, reasonably 'tested'
'Trading Plan'. They are like a pilot who feels the best way to learn to fly is
to taxi that baby down the runway and let's see what I can do, even if it's
only after two or three lessons. Do you have any idea what you must do before
you take a commercial jet up in the air by yourself? Yet many people think that
after reading a few books putting real capital on the line is appropriate.
Let's look at what I suggest.
I say a person should develop their 'Trading Plan'
completely, and do the chart work until they feel they have a complete plan.
His or her edge (or in my terms, group of small edges from across the entire
plan) should be clear. Now the plan should be 'tested' as best as possible, to
see if the perceived edge holds up in a trading situation. I see no point to
use real money here. The person is not experienced, and the plan is very raw at
this point. The goal right now is not to see if the person has psychological
issues, but to see if the 'Trading Plan' has an edge.
My experience tells me the chances
for a plan to have a solid edge right away are slim and none. It will take a
lot of work, refinement, and reworking to have an edge, if the person
can even develop an edge. Not all who try any profession will succeed.
It's not a given, by any stretch. And there is another aspect to look at here.
You want to mess your head up big time, here's a good way. 'Test' out a new
plan with real money when you are new to trading or have not had previous
success, and lose repeatedly. Most people will carry such psychological scars
from this that they will become very apprehensive in the future when they do
have a working plan that they may never get over it. I highly recommend
not doing this to yourself. You might be the exception, but most aren't.
This is one of the most common things I see.
If one starts out with a simulation account,
set to a small amount, they can practice. They can 'test' the plan and see if
it has a net positive outcome under simulation conditions. Now, this is no
guarantee whatsoever that it would perform that way with real trading, and that
is clearly understood from the start. This gets into limitations, which I will
discuss briefly here before we move on with the discussion. Simulated trading
is not real trading, but it mimics many aspects, and that can make it a useful
tool, as long as it is clear where it falls short.
The biggest issue I hear about is
that it doesn't take into account the psychological aspects of trading, the
'putting the money on the line' issue. That's 100% true, but we are now back to
my hammer example. The point is to find out if the 'Trading Plan' has a net
positive outcome, not if the potential trader is mentally fit at this point.
What use is trading real money with a plan that is not net positive outcome?
The simulator is a tool, and what the goal is with it is very specific here.
Now, how close is the simulation to real trading? How useful is this method of
'testing'?
That varies from simulator to simulator, and it also varies
depending on what you are trying to test. In some simulators the engine is
pretty good, and can take small to moderate 'market effect' into account, based
on your trade size. I don't think any can handle mini 'scalping', and I suggest
you don't even try to use them in that way. I think it is near impossible to
create an engine that could accurately reflect slippage, fills, and so on for
scalping, but since this is not part of the Kane Trading methodology this
doesn't come up for anything I discuss, and we'll assume that is a given
here.
I
have found that a good simulator is quite accurate for anything from my lower
timeframe 'intraday swing trading' of the mini's right up to longer-term swing
trading and position trading. It serves my purpose to 'test' a methodology out
to see how it performs. No, the fills will never be the exact same as one would
get in a real trading situation, but the goal is to get a reasonable estimate
of whether the plan is net positive outcome or not. I see no point to move
forward with the process until this step is established. I feel that simulation
trading will, by its very nature, just about always yield better results than
real trading. But in my experience a good simulator will be close.
So, if the plan
'fails' in a simulator and the real world results would be less, why move to
real money? How could that be sensible? The first step is to develop consistent
results in a simulated environment, knowing full well that real world results
won't be as 'good', and that real world trading may bring about psychological
issues and so on that aren't exposed yet. (I have found, though, that many
people do get nervous using a simulator, not because they could lose money, but
because, in a way 'this is it', and they may 'fail' or 'succeed', and hence
sometimes psychological issues do show themselves here.) There is no point to
try to run if one can't walk yet.
Once the plan has shown itself over time, under various
market conditions, to have a net positive expected value, the trader may want
to then move on to real trading with actual money. One great advantage to
simulation trading, especially if it is done on a simulator that is the same
platform as the actual trading platform, is that the use of the platform can
become second nature, with a lot of experience now under the belt. I am amazed
at how many new traders will trade live with real money on a platform they
don't even understand, barely understand, or don't fully understand. I just
don't get that. So, simulation trading assists the learning, in real time, of
the platform, and I rarely ever hear this pointed out as a positive for
simulation trading.
Now, only the trader can decide for him or herself when it is time
to trade real money, and how much to work with. That is dependent on the
individual circumstances, and other than saying never trade with money
you can't afford to lose with essentially no material impact on your lifestyle
('scared money never wins'), I can't add anything to the already copious
published material on this topic. I am for trading as small as possible to
start, and seeing how that goes before gradually moving up in size, according
to the 'Trading Plan' goals. Trading is curious in that it can be very
'streaky', and it is hard to draw conclusions without a lot of data. I've seen
people start out small, get on a streak right off, and then plunge in big, only
to be handed their heads.
Keep in mind, too, that depending on your plan and you
own personal financial situation (the details of this are outside the scope of
this article, but thoroughly do your research, your due diligence, and consult
all necessary financial professionals before you make your own decisions on
this), one contract of a mini or say a currency future may be way too
big to start, in my opinion.
This gets into undercapitalization, maximum risk
calculations, and so on, which is also outside the scope of this article, but
be aware, even if this bothers you, that 'too big' is relative to your
particular situation, and that just because you may be undercapitalized, or
capitalized such that the minimum unit size of the issue you want to trade is
not small for your purposes doesn't mean you should just ignore the rules.
Saying things like 'well, this is all the money I have to work with, so I'm
just going to go for it anyway' is as close to a surefire guarantee of failure
as I can think of, not to mention potential financial disaster.
To me, there is a
process to go through to go from 'I'd like to be a trader' to 'swinging a big
line' using the maximum guidelines set about in one's 'Trading Plan'. I think a
good quality simulator has a place in that process, as long as what the tool's
capabilities and limitations are is fully understood. Remember the hammer
analogy. A tool is only as useful as it's proper implementation in a
well-constructed plan implies, and no more. Know your tools, and use them
appropriately. I think simulation trading is a great tool, and I still use it
myself when I want to test the viability of a new concept once it is past the
'I found this cool new thing on the charts' stage. I suggest you test it out
with an open mind and see how it may fit into your own 'Trading Plan'.
|
|
|
| |
|
|
 NOTE: Reading this page or
any page on the Kane Trading website, or utilizing this website and any
material
contained herein in any way, shall constitute an
acknowledgment that you have read, understood and agreed
to all
the disclaimers,
terms & conditions, and
policies of this site.
|
|
|
This
website is best viewed with MSIE 6.0, text size set to medium, and screen
resolution set to 1024 by 768.
Copyright ©
2007 Kane Trading. All rights reserved.
|
|
