Trader's Performance Index(TPI)

[virtualmoney]

Quote from nonlinear5:

This is better, but still biased. Compare traders A and B:

Trader A returned 40%, had 30% Max DD, but his account also experienced 25% drawdowns every single day.

Trader B returned 40%, had 35% Max DD, and no other drawdowns greater than 1%.

According to your formula, trader A is better than trader B, while the opposite is true.

a max 35% DD is a max 35% DD no matter how low his DD 1% is on any other days, meaning he or his system cannot accustom to a new situation which he has never encountered.

Reminds me of a martingale automated EA, low DD, consistent return until somewhere in Aug last year just blew up compared to another EA with higher daily DD but survived the subprime whipsaw.

 

[Moneyball]

Quote from virtualmoney:

You can sweep everything under the carpet but at the end of the day, the bottom line is people want to see is %R accounted in some form in the calculation which they can easily obtain from the accumulated balance data over any period.

Who are "people"? Most prop (chop shop, as you like to call it) traders couldn't care less what others want to see- the bottom line is, net P&L is the most important for prop traders. You can't spend %'s.

If you're talking about managing other people's money, raising capital, "true" prop, etc., that's a different story.

 

[Dustin]

Quote from virtualmoney:

You can sweep everything under the carpet but at the end of the day

Alright...not so helpful. Anyone with ideas on how to create a measure of performance for prop traders besides the old way (straight $ gains, or cents per share), please post your ideas. I think it would be interesting.

 

[virtualmoney]

Quote from Moneyball:

Who are "people"? Most prop (chop shop, as you like to call it) traders couldn't care less what others want to see- the bottom line is, net P&L is the most important for prop traders. You can't spend %'s.
If you're talking about managing other people's money, raising capital, etc., that's a different story.

If they couldn't care less what others want to see, why bother even measuring their own performance? In fact, they should STFU to avoid attention.

 

[Dustin]

Quote from virtualmoney:

If they couldn't care less what others want to see, why bother even measuring their own performance? In fact, they should STFU to avoid attention.

A measure of performance against peers is useful and interesting information.

 

[BJL]

Quote from nonlinear5:

This is better, but still biased. Compare traders A and B:

Trader A returned 40%, had 30% Max DD, but his account also experienced 25% drawdowns every single day.

Trader B returned 40%, had 35% Max DD, and no other drawdowns greater than 1%.

According to your formula, trader A is better than trader B, while the opposite is true.

Actually, I think trader B is a very dangerous trader likely to blow up in the future.

 

[nonlinear5]

Here is what I suggest:

TPI = PP * (1 - (GrossLoss/GrossProfit)) * sqrt(TimePeriodLength) * sqrt(TotalTrades)

where PP is percent of profitable trades.

 

[virtualmoney]

Quote from nonlinear5:

Here is what I suggest:

TPI = PP * (1 - (GrossLoss/GrossProfit)) * sqrt(TimePeriodLength) * sqrt(TotalTrades)
where PP is percent of profitable trades.

TotalTrades? hmm...What happens if the trader is a scalper with an itchy finger on the mouse?

PP? what about non-directional spread traders?

 

[Moneyball]

Quote from virtualmoney:

If they couldn't care less what others want to see, why bother even measuring their own performance? In fact, they should STFU to avoid attention.

I'm not sure who you're talking about now, but a lot of prop traders don't measure their own performance in terms other than P&L (and many do try and avoid attention). Some go a step further and look at things like cents per share, etc, but things like % return, sharpe ratio and DD% isn't very important to a lot of these traders. Some have virtually no DD at all, especially if you look at weekly or monthly.

For instance, if Prop Trader A has $10 MM in BP and Trader B has only $5 MM, but they each make $250K in a year, is trader B twice the trader as trader A? No, and I don't think most traders walk around asking "how much BP do you have?" as a measuring stick. Besides, BP available and BP used are two entirely different things- I know traders with $10MM available but they rarely use more than $2-3MM at any point in time. Also, most of these prop traders hold things for minutes or hours, rarely overnight. How do you factor that in?

I'd be interested to see if some one came up with something to measure performance this way, but I still think net P&L is the most important factor to these prop traders. For other traders, other measures are much more important, and your formula may be a decent one for that. I'm just speaking about your average prop guy.

 

[JaiSreeram]

Quote from virtualmoney:

Is there a standard performance indicator for a trader?

I notice that trader's perfomance based on %return or sharpe ratio are not sufficient to gauge the quality of a trader.

(sharpe ratio alone cannot indicate trader is generating enough cash returns)

May I propose a "Trader's Performance Index" (TPI) ?

TPI = %R X (%R / %DD) X %t,

for example 90% return, 30% max DD, 12 months time frame

TPI =0.9 X 0.9 / 0.3 X 1.2 = 3.24

Can you please explain the symbols used above and what is %return or sharpe ratio? Moreover, I thought trading is a more objective term and can not be reduced to mathematical terms as conditions differ in various markets and countries. So, is it possible to standardize the above for all traders?