|
PureBytes Links
Trading Reference Links
|
That's where I was getting my information, but there are additional
steps to follow (that I unintentionally left out in my example) if
you're calculating expectancy for trading situations.
In the example of betting $1,000 on the (50/50 even-money payout)
outcome from flipping a fair coin, the risk and reward are the ENTIRE
$1,000, EVERY TIME, which makes figuring the expectancy a snap.
A trader who buys $1,000 worth of stock doesn't typically have the
entire amount at risk, though. If he uses a stop of 10%, his risk is
only $100, not $1,000. If he takes a profit at 10%, he only makes
$100 and not $1,000. With $10,000 in equity, his 10% profit or loss
is now $1,000. The bet-size is constantly changing and expectancy
has to be "normalized" to get the expected dollar of profit per
dollar risked, and the only way to do that is by doing a trade-by-
trade analysis.
Fred's example assumes that all trades have a risk and reward that's
a fixed percentage of equity, normalizing the bet size. If I'm
finally understanding this clearly.:)
Luck,
Sebastian
--- In amibroker@xxxxxxxxxxxxxxx, "RR" <stocks@xxxx> wrote:
> You can read about expectancy versus expectancy per dollar risked
in TYWTFF
> by Van Tharp pages 148 - 152.
>
> Expectancy per dollar risked gives you the truer picture of your
trading
> system.
>
> -----Original Message-----
> From: amibroker@xxxxxxxxxxxxxxx [mailto:amibroker@xxxxxxxxxxxxxxx]
On Behalf
> Of sebastiandanconia
> Sent: Saturday, September 03, 2005 10:26 AM
> To: amibroker@xxxxxxxxxxxxxxx
> Subject: [amibroker] (OT) Re: AmiBroker Support
>
>
> You know, Fred, I think you're right. My example has major flaws
> that I didn't recognize and your example addresses them
> appropriately. Sorry, and thanks for taking the time and effort.:)
>
>
> Luck,
>
> Sebastian
>
>
>
> --- In amibroker@xxxxxxxxxxxxxxx, "Fred" <ftonetti@xxxx> wrote:
> > One can "look at it" any way one wants.
> >
> > Over a short period of time almost statistics are meaningless ...
> >
> > Let's look at a trader who over a longer period of time has a
75/25
> > win/loss ratio and starts w/$1000. Let's assume he has 3 winning
> > trades of 10% each followed by a losing trade of 10% and then the
> > cycle repeats.
> >
> > After 100 trades he has $117,391 w/the average win being 2446.55
> > (10%) average loss being 2682.53 (10%) ...
> >
> > You prefer an expectancy calc of:
> >
> > 0.75 * 2446.55 - 0.25 * 2682.53 = 1163.90 i.e. < 1% of his equity
> >
> > or:
> >
> > 0.75 * 10 - 0.25 * 10 = 5% i.e. ~ $5869
> >
> > If we take the same example over 500 trades then we have an
> > expectancy that's around 0.1% using $ to the same 5% using %.
> >
> > The longer the example, the more out of whack this gets ...
> >
> >
> >
> > --- In amibroker@xxxxxxxxxxxxxxx, "sebastiandanconia"
> > <sebastiandanconia@xxxx> wrote:
> > > Is it? I'm not sure.
> > >
> > > Example:
> > >
> > > A trader has a system with a 50/50 win/loss ratio. Starting
with
> > > $10,000, he has three losing trades in a row with each costing
> him
> > > 10% of his existing equity. Three losses, each 10%, $1,000,
> $900,
> > > $810. Average loss -10%, or (1000+900+810)/3 = $903.33.
> > >
> > > From his current equity of $7290, he has three winning trades
> that
> > > bring him back even to $10,000 (it's $9999.87, to be precise).
> > Each
> > > trade is a win of 11.111% of his existing equity,
> > > (809.99+899.99+999.90)/3 = $903.29.
> > >
> > > Using an expectancy formula calculated with percentage profit of
> > > winning and losing trades would lead him to believe he had an
> edge
> > of
> > > 1.111%, but he obviously doesn't since the average dollar amount
> of
> > > his wins and losses are virtually the same.
> > >
> > > Or if he took a larger hit of 30% on his first trade, bringing
> his
> > > equity down from $10,000 to $7,000 and then had a big win of
> 42.85%
> > > bringing him back even, he'd think he had a REALLY big edge when
> in
> > > reality the dollar amount of both trades were the same.
> > >
> > > Or am I looking at this the wrong way? I'm not exactly a
> > bookie.:)
> > >
> > >
> > > Luck,
> > >
> > > Sebastian
> > >
> > >
> > >
> > >
> > > --- In amibroker@xxxxxxxxxxxxxxx, "Fred" <ftonetti@xxxx> wrote:
> > > > Using $ as opposed to % in formulas like this is really only
> good
> > > for
> > > > the futures traders who tend to trade the same number of
> > contracts
> > > or
> > > > dollar amounts for awhile ...
> > > >
> > > > --- In amibroker@xxxxxxxxxxxxxxx, "sebastiandanconia"
> > > > <sebastiandanconia@xxxx> wrote:
> > > > > Sorry, I had to edit my first reply. Let's try this
again.:)
> > > > >
> > > > > In your expectancy formula, are you calculating Avg Profit
> and
> > Avg
> > > > > Loss as % (as it appears in your formula)? Because
expectancy
> > > > > formulas quote the profits and losses in dollar amounts,
like
> > > this:
> > > > >
> > > > > (%Wins x Avg Profit $) - (%Losses x Avg Loss $)
> > > > >
> > > > > Could this account for the difference of opinion that you
and
> > > Tomasz
> > > > > have? Expectancy tells you what dollar amount you can expect
> to
> > > win
> > > > > or lose per dollar amount risked. However, the dollar
amount
> > > > risked
> > > > > doesn't vary with leverage. Buying $10,000 worth of a
mutual
> > > fund
> > >
> > >
> > >
> > > > in
> > > > > a cash account puts $10,000 at risk, but so does buying
> $10,000
> > > of
> > > > > that fund on margin with $5,000, it's just that you're
> > borrowing
> > > > the
> > > > > other $5K. Same thing when buying $10,000 worth of a mutual
> > fund
> > > > > that uses 2X leverage, still the same $10K at risk. The
> amount
> > > of
> > > > > money risked is always the same, so the expectancy has to be
> > the
> > > > > same.
> > > > >
> > > > >
> > > > > Luck,
> > > > >
> > > > > Sebastian
> > > > >
> > > > >
> > > > >
>
>
>
>
>
> Please note that this group is for discussion between users only.
>
> To get support from AmiBroker please send an e-mail directly to
> SUPPORT {at} amibroker.com
>
> For other support material please check also:
> http://www.amibroker.com/support.html
>
>
> Yahoo! Groups Links
------------------------ Yahoo! Groups Sponsor --------------------~-->
Make a difference. Find and fund world-changing projects at GlobalGiving.
http://us.click.yahoo.com/PcNrnD/PbOLAA/cosFAA/GHeqlB/TM
--------------------------------------------------------------------~->
Please note that this group is for discussion between users only.
To get support from AmiBroker please send an e-mail directly to
SUPPORT {at} amibroker.com
For other support material please check also:
http://www.amibroker.com/support.html
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/amibroker/
<*> To unsubscribe from this group, send an email to:
amibroker-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|