I half agree with this statement. The best (easy) way to test anything is to use the number of bets and the next question is always how many bets is a statistically significant number.

I usually say around 1000-2000 should be enough to get a feel for a system BUT also a minimum of 3-6 months is a good timeframe. For example there are some systems that work during spring carnival time but fall over the rest of the year.

To tell the truth it is one thing I have lacked on checking this progressive staking plan. While it has had over 1200 bets it has only been over the timeframe of a month. I thought rather than test it on prior random months I would just check it against the next couple of months.

As I have said all along I am trying to see if using a progressive staking plan will work with the favs on betfair (both backing and laying). There may be some reason it works that has nothing to do with the staking. It may work because the prices are so close to breakeven that as they sway in and out of profit from each side(back and lay) it provides the edge the progressive staking plan needs to make a profit. The only way I am going to prove or disprove any theory is to run a semi-live test period which I am trying to do but there seems to be a lot of people hell bent on condemning ideas before they even have some test results.

Good Luck.