I've observed that the more bets you have, the less POT you'll be able to achieve. Using ratings for example, and betting in most races, a POT less than 10% seems to be a great result. I have run systems through my database and have come up with quite a few different methods that show POT's well in excess of 50% but each system showing only one or two bets per month. Cumulatively speaking these systems add up to one or two bets per week. There's a lot of work for one or two bets per week, and Bhagwan has stated many times on this forum that the best systems seem to have only a few bets each moth.

So, the more bets you want, you'll end up with a skinny POT ..... the fewer bets the greater the POT.

I've recently come up with a system with a great POT that targets first up runners but only has 79 bets over 12 years. The S/R is in excess of 30% and the max div is $9.90 so the system has legs. BUT, who on here only wants to bet once or twice per month.

It reminds me of a rule in the Zurcih Axioms, which states you should only play for meaningful stakes.

Make of this post what you will ..... I'm sure I'll be criticised for not posting my systems.

At the end of the day it comes down to making money.

Compare Woolworths vs a Car yard.

The car yard might have a larger POT then Woolies when you compare sales and costs of goods sold but does it mean its better???

Its an individuals preference. Six or half a dozen!

In regards to your 79 bets - I would have more confidence in 5000 bets in 1 month than 79 bets over 100 years.

The question is "what do you want out of it?"

As stated you can find systems with great POT and they are all low number of bets. The otherside of this is a small POT but volume. If you are after a greater profit and better security then you are better with the latter. I would always choose to have 10% of a million over 50% of a thousand.

Alternatively you have to ask how did the greater POT system get its profit. If it was from longshots, how are you going to replicate the returns? To go down this path you need many systems with similar POTs and then you need to automate the selection process and have a method of ensuring every selection is bet. Sounds simple. Get a dozen systems with 50%+ POT, generate the selections via BetSelector or the like, then back them and sit back with a stubby and watch the profits come in. The reality is that you will need deep pockets and in the week that you go away with the missus all the longshots come in. The smaller the number of bets the more unlikely it will replicate itself.

Well Max was the man!

If you could simulate your 79 bet's in 12 years system out to say 1000 bets I wonder what your long term POT would be. If you could break a 1000 bet system into groups of 79 bets I wonder how often the POT would be well up above the the 1000 bet average.

__________________
Pixie

"It's worth remembering that profit isn't profit until it's spent off the racecourse." -- Crash

A lot of the system sellers like bazza rely on low volume of bets (Av 18 - 35 bets a year ) to drag out the punting torture to its end degree before you go broke.

In saying that though one could call these "Special bets" and perhaps have a much larger stake whilst continuing to use the trusted Wt Ratings.
i.e Barry Watsons excellent mechanical system.

Cheers
darky

Is the duration of the testing period important? I would think so. I'd rather have a successful system compiled over 12 years of data than one which had many selections over a period of 6 months.

And how far back did bazza's data go darkydog2002?

All his systems only went back 2 years.
Funnily enough they all fell in a hole once released.

I'm assuming the Win S/R stayed about the same but the dividend dropped. I suppose you could backfit a system over 2 years. The max div must be taken out of the equation. None of my systems rely on one long shot winner, in fact my price restriction doesn't allow for long poriced winners, but I still take out the max div.

I kinda of disagree with you Barny on that statement.

I would have more confidence in certain systems with 5000 bets over 6 months than that of a system with 100 bets over 12 years.

Over 6 months I would be more mindful with the seasons of racing.

You could easily back-test a system to work on Melbourne cup over 100 years of information. I still would not have confidence in that holding up in the 2012 Melbourne Cup.

Vortech, there was a terrific thread on polls and how little the sample size needed to be. Most of my systems have one or two bets per month max. I agree it's not ideal, but the POT makes it enticing.