G'day All

I've been taking a more disciplined approach to my betting the last couple of months and my question is: How do you know if a system is viable? I've read somewhere about something called the 'chi-Squared' test anyone familiar with it and it's workings?
I operate about 15 systems together and in the month of January
Bets: 445
Wins: 126
Win strike rate: 28.3
Return: 579.4
Longest losing win sequence: 15
Highest priced winner: 19.00

These systems were designed using a program that has been mentioned on here before and created over 1 to 2 years of data, every year since the initial period they have shown a profit, basically from 2002 onwards. I tend to believe that the results can hold over the long term but is there a formula that people use to determine if any system has potential. Also for the mathmaticians or anyone with more knowledge than myself could you explain in simple laymans terms what 'win drawdown and maximum drawdown ' mean in relation to graphs or profit/loss.

Thanks and see what happens to Sunshine Lover in the second at Cessnock.

With a name like Sunshine Lover, it won't like the slow track!

Wocka wocka wocka!!

(to quote the great Fozzy Bear)

Cabernet today Duritz?

Nanook,

Maximum Drawdown of a betting bank is basically the biggest DIP from a bank's peak point (in the period in question) to its lowest point (in that same period). It is not necessarily the difference between the highest point EVER and the lowest point EVER but it could be. The low has to come AFTER the high and the maximum Drawdown is the largest recorded DROP or descent (hopefully before the bank recovers and surpasses its former glory!).

More than likely, there would be a number of collects and small upward movements in the bank balance within the maximum drawdown period unless, of course, the poor old bank goes into extreme (vertical) freefall and you begin to ponder on the sanity of betting at all.

I've got no idea about "chi-squared" or "win drawdown". Sounds like they could damage the brain, though!

Cheers,
QFB

Nanook,

This is the method I use.

Take the maximum win dividend and divide it by your profit, this gives the number of winners that make up the profit...the more the better.

Now you might ask why not use the average win dividend etc.

This is to show if the profit comes from a couple of fluke winners or not, so it may not be a trend, just an anomoly in future expectation.

Example: One of my systems, I divide the profit by the maximum win dividend and come up with 22 winners. This is very good, and continues to show profit in realtime. Other systems with very low scores of say less than 5, have dropped into loss or become shockers, because the profit relied on an unreal average of events.

Hope this helps.

__________________
RaceCensus - powerful system testing software.
Now with over 413,000 Metropolitan, Provincial and Country races!
http://www.propun.com.au/horse_raci...ng_systems.html
*RaceCensus now updated to 31/01/2025
Video overview of RaceCensus here:
http://www.youtube.com/watch?v=W821YP_b0Pg

QuickFunBet,
Thanks for your reply, I suspected it may have been something similar to what you wrote and your confirmation is appreciated.

Chrome Prince,
Thanks as well to you, but could you tell me if that figure of 5 is a definite cut off point? or how do you determine the cut off point? In my above fiqures I get the number 7, have I worked that out correctly as I have the mathmatical capacity of a thimble.

Cheers