1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Changes to the FPA Performance Testing Statistics

Discussion in 'Performance Testing of EAs, Signals, Managed Forex' started by AsstModerator, Mar 28, 2012.

  1. AsstModerator

    AsstModerator FPA Forums and Reviews Admin

    Dec 11, 2007
    Likes Received:
    The FPA's Performance Testing statistics keep generating questions and controversy.

    First, the history...

    At the very beginning of the Performance Testing program, the formula was to simply subtracted the Initial Deposit from the Current Equity and dividing by Initial Deposit obtained Gross Equity Growth in %. Weekly numbers were derived from division of the Gross Equity by time. This formula worked fine for the fist tests but as the testing intervals became longer one alert members pointed out that it didn't include the compounding.

    This led to the second version which included compounding to calculate weekly profit or loss. This worked very well for demo and live accounts with no withdrawals. However, the FPA's Performance Testing program became more and more popular. Many account managers and signal service providers wanted to have their real accounts monitored by FPA. Some EAs also showed live accounts. Withdrawals are often the part of the money management strategies of account managers. Signals and EA sellers also sometimes took out money, so the formula was further adjusted for withdrawals. Then people noticed that deposits were throwing the numbers off.

    This brings us to the current formula. Maybe you traded for a year with a $100 account, made good money, and then added 10,000. It wouldn't be fair to use either $100 or $10,100 as the Initial Deposit for the calculations. The programmers came up with an artificial value called "Time Weighted Deposit. The name is mostly self-explanatory. Each deposit in the sum is first multiplied by the time weights factors where the first deposit weight is equal 1 and all other weights are less then one and proportional to the time that each deposit was made.

    All of these improvements give a much better picture of equity changes, but there are still some problems. Large deposits throw the numbers off for too long of a time. A smart service provider can calculate how much money to add to an account to move the test up in the rankings on the Performance Test listings pages.

    After much debate here at FPA HQ, it's been decided that actual profit or loss from trading activity would be the better way to sort the test performance. New formulas will be put into place that will greatly reduce the impact of large deposits. Later this week or sometime next week, the test pages and external banners will change to reflect the profit/loss value. This will cause some major changes to the ranking order of tests.

    Please come back soon and see the results.

Share This Page