CityPips overstates Performance Results

I subscribed to CityPips signal service and compared results to what they post on their "Our Results" page. Their "Results" page claims (as of 9/18/09) that they've made 3,662 pips since 5/1/08. They arrive at this figure by subtracing losing pips from winning pips. But as you'll see, their winning pips are usually a smaller-position-size pip than their losing pips. They cleverly don't mention position size on their Results page.
Here's how the scam works:
Signals contain two profit targets and one stop loss.
When the first profit target is hit, stop is moved to break even on the remaining position.
To keep the math simple, let's say you trade $1.00 per pip,
and you take 1/2 off at the first profit target.
Let's say the signal has profit targets of +30 and +50, and a stop loss of -40. On 9/15, CityPips issued such a signal and it played out like this:
CityPips: We are closing 1/2 of our AUD/USD trade prior to the report release to preserve profits at +30 pips!
Move S/L to B/E.
So at this point, you're up $15.00, correct? 30 pips on half the position.
The next email received from City pips said:
Our remaining 1/2 position of the AUD/USD trade stopped at B/E. Total gain for trade at +30 pips.
On their Results page, they did indeed post +30, even though only part of your position realized those 30 pips. Yet on a loser, your full position size realizes the loss.
So when you look at the results page, and if you trade minilots, you have to see that losing pips are $1 per pip, but winning pips may only be $0.50 per pip, depending on what portion you take off at the first profit target. So losing pips and winning pips are not apples and apples, they're apples and oranges. Therefore the number of "winning" pips is meaningless.
If you win 4,000 fifty-cent pips and lose 2,000 one-dollar pips, how many net pips did you win? According to CityPips, that would be 2,000 pips gained. How stupid do they think we are? Of course, by the time you find this out, you've already paid for at least one month.
Yes, I asked CityPips about this. And they admitted to it, sort of. (I have the emails.) They try to make it sound good by saying "Joe, that is correct, our results do not take into account money management techniques such as closing 1/2 or 2/3 of your position at certain levels to lock in profits. The pip count is indiscriminate of position sizing." and "we have found that our subscribers like this simple approach of calculating pips."
I went back and forth with them 7 times. Bottom line, they're going to keep comparing partial-position-size pips won to full-position-size pips lost, as though a pip is a pip is a pip.

 

There's a great book titled How to Lie with Statistics. I think it's commonly used by certain signals providers.