P/L unit statistic

C2
1

Can we have a closer look at this P/L unit statistic please, in the form of a tip (similar to W/L ratio etc.) that shows up on the front page.



For eg., my system Midas Long-Term Value has P/L unit as $1,153.88 Just curious how this is calculated.



rgds, Pal

Midas Long-Term Value (www.collective2.com/go/money)

Midas Short-Term Value (www.collective2.com/go/money1)

2

I was just looking at this value for my trading system, and although I don’t know how Matthew calculates it, my guess is that it should currently be 4 times bigger for my system “Daytrading WIZ”. Of course, I am making lots of assumptions about how I would figure it out.



So, just how do you calculate it?

3

Since MK has been silent about the calculation part of this P/L unit statistic, I can only guess.



It seems to me that the P/L per unit can be considered to define the ‘payoff’ structure of the portfolio and if so, it is in my opinion a vital statistic and as somebody said, a real eye opener. By ‘payoff’ structure I don’t mean ‘payoff ratio’ which is the same as the ‘profit factor’. I hope somebody would chime in here and correct me if I’m wrong.



rgds, Pal

4

Sorry. I didn’t mean to be silent. Just bogged down.



Anyway, P/L unit is not really a big deal. It is simply a measure of how much profit the trading system generally generates per ‘whatever’ that it trades. So, if the system generally trades stocks, and generates only .03 cents per profitable trade, you have to say to yourself: wait a second, .03 cents sounds okay – but then I have to pay .02 cents commission each turn (for a deep-discount broker!) and I’m sure that I’ll have more slippage than the Platonic Ideal World of Collective2 creates…



In a sense, P/L per unit is a measure of how ‘robust’ the system is – that is, how likely it will create real profits in the real world. A very low P/L per unit is a sign that there’s not much room for error: commissions plus slippage will eat you alive.



In contrast, let’s say I found a system that trades electronic futures contracts, and that generates $50 of P/L per contract traded. Since commissions are likely only $5 per round turn, even with decent slippage I’m still going to do okay.



There are some pretty obvious weaknesses with this measurement: it averages P/L across all types of trading instruments, so for systems that trade lots of different stuff, it’s not that helpful.



But it should be regarded as a rough measurement that helps traders decide if a system is for them – that is, whether it demands perfection in execution, or allows you to be more human.



Matthew