Hi All,

After taking a bunch of 'leads' from different threads, I worked up a system that over the last three months has shown a reasonable profit at level stakes, win only betting.

Now my curiosity has got the better of me and I was hopeful another forumite could run the system through their database?

Here it is:

1. Metro racing Saturday (Syd/Mlb/Brs) or Sunday if major day (e.g. Underwood Stks);
2. No maidens, 2YO, Hurdles or Steeples;
3. All track conditions;
4. Selection must be:
a. Top 5 Average Prizemoney
b. Place Percentage 50% or better
c. Weighted 56kgs or below
d. Barrier 9 or inside;
e. Last Start Winner.

Any insight would be appreciated. Thanks in advance.

Luckyboy

d. Barrier 9 or inside;

why ?

puntz,

It's purely a statistic that I have been using under test and it is standing up to date.


Luckyboy

[ This Message was edited by: puntz on 2003-11-09 17:09 ]

Ran it through 500 races
21% SR
-43% LOT

The 50% rule is the thing that realy kills it.

Works better if you have the top ranked place getter which had won by 2.5+ lengths or more last start.

__________________
Cheers.

Rather than just increasing the amount of filters [the usual method to improve S/R on systems posted], how about adding in 'Must be in the top 5 pre-post odds' and dropping the place % compleatly [ a 50% place S/R in the bush is meaningless in town ]? That should improve the S/R a lot.

The 56kg. rule could be another area for improvement because weight becomes increasingly important and relevent with increased distance [above 1600m. 56kg. is probably too high]. As there are no distance rules the Weight rule doesn't really relate to anything.

Barrier rule relates only to the fact that 1-9 barriers win more due to field sizes distorting the true stats because those barriers get run more than those above 9 [see barrier discussions on this forum]. That being said, it is probably a good rule because of the above and worth knowing about anyway [in case you didn't already].

Cheers.



[ This Message was edited by: crash on 2003-11-07 08:16 ]

Bhagwan,

Thanks for the check and the suggestions.

I'll keep plugging away.


Luckyboy

<Qoute>
"Works better if you have the top ranked place getter which had won by 2.5+ lengths or more last start"

Interesting idea Bahgwan. Not seem many systems with "beaten length" features for some time. One of my early ratings systems used this, interesting to see it mentioned again.
But you must also filter then, if it was the same jockey riding the horse that won.
It might be worth considering.

[ This Message was edited by: crash on 2003-11-09 18:06 ]

Puntz,

The reason you never see a beaten length rule [far more logical than a place rule] is that it can't be run through a data base [backfitted]. I use 2.5 lengths from winner and importantly 'running at least 8th at the 800m' last start which puts the horse into the 'on the pace ' at least catagory. Like to see the stats freaks run that one on their data bases.
Totaly sound and rational rules continualy escape stats. backfitting [but win races]. You will NEVER see the armchair statsmen in any tipping comp. Why ? I wonder. Until they PROVE their point, their arrogance and methods should be ignored [frustrated mathematictians with a dribble from the corner of their mouths showing].

Ho Ho.

Cheers.