Ok guys, I am back again with more useless ramblings.

Would you yourself begin to bet on each/if any of the following?:

SYS BETS RETURN POT% 1sts S/R
A. 245 $332 36% 51 20.82%
B. 240 $336 40% 45 18.75%
C. 291 $376 29% 43 14.78%
D. 80 $117 46% 24 30.00%
E. 160 $242 51% 48 30.00%

Most of those systems used the same filters which deleted approx. 100 bets,
that returned approx. 80-120 units.

I cannot find anywhere on the internet something like a table of %
chances of a system failing in relation to S/R's.

Personally, I think the POT% is irrelevant to knowing the long term chances
of a system succeeding or failing. Obviously the higher POT% the better,
but the S/R is what it is all about. Knowing that after 300 or 500 bets
that you can expect a reasonably consistent S/R, then obviously you take
your minimum price based on that S/R.
If your system has had 500 bets and the S/R is 20%, you'd say that
would pretty well hold up long term.
If your minimum price on that S/R is $10.00 or $7.00 then the POT% is irrelevant.
But kaching, kaching either way.

If your system has had 150 bets, 25% S/R, could you safely say you
are rolling a 4 sided dice with $5.80 on each successful roll long term?
Or has your system just been lucky.

So, I tried to make a table myself. Have a checksies, what do you think?
My POT% on any system is reduced by the Fail% on attached sheet.
The higher S/R, the less chance of a system failing on my excel formula.

Sound good, or am I full of ****.
To me the hardest thing in this horse racing game is not making any system
successful, but knowing the chances of it continuing successfully long term.


PS. I don't like the idea of another 40 years of working life.
This is my life's yellow brick road, it's like I woke up one day and said,
"I'll never give in until I am making a living on the nags."
I know it can be done it just takes time.. forget the doomsdayers, it is
too hard for them or they have a hidden agenda.

So.. my table, any thoughts on it? Adjustments that could be made?
Are my systems ready, or not yet?

Attached Files

FAIL %.xls (30.5 KB, 473 views)

Try the free download from the twonix dot com dot au site.
It's exactly what you're looking for.

Compare your sheet with this using (say) 60/300

http://members.aol.com/johnp71/confint.html

It says your observed strike rate of 20% could be "really as bad" as 15.62%.

Therefore you'd want a ROT of 20/15.62 = 28% POT to compensate.

28% is far bigger than your reduction of 10%.

Also just bet A+B+C+D+E with your disposable income on AusTote with its introductory 2% rake.

The worst thing that can happen is you going crazy after some beginner's luck.

It is more character-building if you lose a bit as that should force you to research harder to lift your game.

Saved the ROR calculator long ago.

My comp says:

C:\WINDOWS\SYSTEM32\AUTOEXEC.NT.
The system file is not suitable for running MS-DOS and Microsoft Windows applications.

Thanks for that jfc, may be just what I've been longing for.

Cheers.

So let me get this right jfc,
this formula is saying to me that a system after 300 bets with a
strike rate of 20% will probably long term settle to a strike rate
of 15.62% - 24.98%. This is considered 95% accurate.

So basically a 1 in 20 chance
that the strike rate will fall below 15.62%. Correct?

Sahasastar,

I don't want to pretend to be some sort of authority on this, but:

Lower down on that page are confidence limits.

Currently they are 2.5% at each end.

So you have a 1 in 40 chance your strike rate will fall below 15.62%.

You may be justified in tuning those parameters to something higher, to stop hanging around forever.

There's nothing that wrong in starting with A, B, C, D, E then sacking any non-performing ones along the way, as that is a logical way of reducing your stakes in bad trots.

Hey to all, thanks for you help.
I was on to something, but wasn't sure of correct %'s to use.

This table should be pretty accurate and a great guide to knowing
how confident one should be that their system is on the money,
or close to it.

Feel free to use it if you wish.

Thanks again, I'm off to make my fortune.

Attached Files

System Confidence Percentages.xls (41.5 KB, 472 views)

For those of us too slow witted/lazy/confused to work it out could you give us an example of how to use it using figures on your sheet. Like if I have a sample of 400 bets showing a strike rate of 18% there is a figure of 25.4%. What does that mean?

Cheers,
KV

Hi KV!
I hate to admit, I am in the same boat myself.
Sometimes I feel all those years at uni were wasted. Don't want be personal but some statistical stuff on this forum is, I am afraid beyond my grasp.
Of course I could be just too stupid, or else, did time pass me by!? Think not, last I looked I still had a job, A good one too! And we use stats in our programming work.
Or is it just a failure to communicate? Most likely!
Good punting.

It is saying:

That observed strike rate of 18% is 25.4% better than the low confidence limit.

Suggesting you need a POT of at least 25.4% to be confident that you have a non-negative edge.

Earlier on this thread I showed how to use the Internet tool to get confidence limits for strike rates.