|
PureBytes Links
Trading Reference Links
|
Paul,
Perhaps the confusion may lie in what Tharp factors into his concept of
"expectancy." When YOU use the words "... I can imagine a negative
expectancy game that would make money ...", I assume you mean percentage of
winners versus percentage of losers. So as an example, I imagine you may
be thinking to cut your losses when wrong, and perhaps add to the size of
your positions when they are proving correct.
However, this is not what Tharp is considering when he says "expectancey."
Tharp's expectancey consists of 4 pieces. Only one of the pieces is
Percentage of winners. The other 3 factors are:
1- relative size of profits versus loses.
2- cost of making trade and
3- how often you get the opportunity to trade.
Tharp is certainly agreeing with you when you say "position sizing is
important," as he says it is the most important aspect of sytem design.(page
306).
Regards,
Barry
----- Original Message -----
From: Paul Altman <paulha@xxxxxxxxxxxxx>
To: <omega-list@xxxxxxxxxx>
Sent: Friday, May 19, 2000 2:55 PM
Subject: Tharp's Expectancy
> Would one of you mathematicians kindly take a moment and explain to me why
> we'd require that a system demonstrate positive expected return
> historically, before we'd put any money on it?
>
> I'm specifically looking at Tharp's definition of expectancy, where he
> states: (p.148) "Expectancy is a way of comparing trading systems while
> factoring out the effects of time, position sizing, and the fact that one
> is trading various instruments that have different prices."
>
> Using _his_ definition, I think I can imagine a negative expectancy game
> that would make money, since you could conceivably have a position sizing
> algorithm that would "know" when to enormously increase your bet size on
> winning trades, even if those winning %'s were very small.
>
> Obviously, all things being equal, I guess you'd want a higher expectancy
> rather than a lower one. But why would you insist it be positive? If I'm
> reading him correctly, he's saying that you don't want to play a negative
> expectancy at all. With his definition removing position sizing as an
> input to the expectancy formula, I don't fully understand his reasoning.
>
> Paul
>
|