Some interesting facts

[hahaha]

I did some statistical stuff, covering 6474 in play horse races.... from June to the end of November...

I was looking for an odd, a"safe" odd where on long term trading I would hit a profit, backing a horse who would first reach specific odd. Statisticly, for this period, the most safe odd would be 1,19. Meaning, if you would bet everytime, on every race two pounds, you would earn 399,04 pounds. You would win 5608 races ( 2131 pounds) and loosing 866 races (1732 pounds). Warning 1: I assumed you would loose everytime, second or third or etc. horse would go bellow this odd, which in reallity would not happen. Warning 2: You would need to be matched on every race at this specific odd, which in reallity would not happen as well

So called safe odds are from 1,03 to 1,29. All other odds would in this period bring negative result, considering above method.

I used this method in practice as well. If you were trading in last weeks and were loosing or winning on 1,19.... it was me

However I found out, it is not worth it. You just don't get matched.

Does someone else have any statistical fact?

[wizardofmann]

I started trading three and a half years ago but gave up after three months because unless you are prepared to risk extremely high liabilities, it's a mugs game!

I have been gathering statistics for well over three years and the only thing I can agree with is the matter of getting matched. All my statistical data is fully documented and shows that since 19th July this year to date, my bot has only managed to get matched 22.65% of the time. If I could manage to up that figure to just over 30%, I would be laughing all the way to the bank: if being the operative word!!

I have tried placing bets all the way from 1.01 to 1.5 and found no one particular figure to be 'safe odds.' But then my system does not depend on asking for specific odds. And whatever odds my system asks for only gets returned at the asking odds 51.51% of the time.

I have 'mountains and mountains' of statistical figures but at least 95% of it is not worth the 'paper' it is written on!

[doris_day]

You said "I have 'mountains and mountains' of statistical figures but at least 95% of it is not worth the 'paper' it is written on!".
I would disagree that its not worth anything. Knowing stuff that leads to a negative outcome is just as important as stuff that leads to positive outcomes.
All your stats are useful and the more that lead to negative expectations means when you do find a positive expectation it'll probably last longer and be more rewarding.

[negapo]

Nice conversation.
I have also done a lot of work for the last 7 years, it has being a bumpy ride but in the end it paid off.
I think doris_day is right. Like Fama said (the guy from the financials markets and the EMH): Sometimes knowing how things don't work is as valuable as knowing how things work (like when you dis-confirm something that was so obvious).