OZmium Sports Betting and Horse Racing Forums - System of the Year For 2011.

OZmium Sports Betting and Horse Racing Forums (http://forums.ozmium.com.au/index.php)

- Horse Race Betting Systems (http://forums.ozmium.com.au/forumdisplay.php?f=10)

- - System of the Year For 2011. (http://forums.ozmium.com.au/showthread.php?t=23023)

I think Darky might be right in this instance, I got these LAY results with the system below:

The Rules used were : horseNumber <= 4 and distance < 1001 and lastStart = 1

There were 38142 horse form lines in the database at this point in time. We update when we get time.

There were 180 selections for the System
There were 146 which Lost (did not win race) for the System for a strike rate of 81.11%
You had to pay out $153.86 but brought in $171 after 5% commission. This means a Profit or loss of $17.14 or a percentage profit/loss of 9.52%

The Schmile

Don't want to seem like I'm arguing, but like to point out, that
the data is not accurate price wise.
When I run systems that I know win laying, they lose using the data, so I'm not sure what's going on there.
I'm comparing actual lay bets with the data, and it's way out.
I'm not knocking UB, but if the prices aren't accurate, it puts people at a disadvantage.
I think you need to be recording either the last traded price or use the Betfair weighted average, or Betfair SP, the markets aren't mature enough if the races don't run exactly on time.
Secondly, it's not all last start winners in the field, it's the first one you come to.
It's also last start 2nds.
Using the criteria <=4 includes multiples.

But for the sake of the argument let's do it that way.

Code:


ALL OF THEM AT TAB PRICES 
(that's any TAB #)
01/01/2011-03/12/2011
                 WIN      PLACE
Races Bet:    1523      1516
Races Won:     565      1036
S.R./Race:      37.1%     68.3%
Outlay($):    3396.00   3386.00
Return   :    2727.90   2872.02
$ Profit :    -668.10   -513.98
% P.O.T. :     -19.7%    -15.2%

Code:


Just using TAB # 1 of qualifiers 
Just at TAB prices
SYSTEM RESULTS FOR: SPECIALS_DARKY 01/01/2011-03/12/2011
                 WIN      PLACE
Races Bet:     432       427
Races Won:      95       212
S.R./Race:      22.0%     49.6%
Outlay($):     432.00    427.00
Return   :     347.20    367.24
$ Profit :     -84.80    -59.76
% P.O.T. :     -19.6%    -14.0%

But if you're laying into better prices and paying commission, it's not going to work, because the fancied one's will kill you on strike rate and the unfancied one's will kill you on price.

Code:


Just using TAB # 1 of qualifiers 
Just at TAB prices, WHEN THEY ARE FAVOURITE
01/01/2011-03/12/2011
                 WIN      PLACE
Races Bet:     172       169
Races Won:      61       107
S.R./Race:      35.5%     63.3%
Outlay($):     172.00    169.00
Return   :     136.40    142.90
$ Profit :     -35.60    -26.10
% P.O.T. :     -20.7%    -15.4%

The strike rate is better than just favourite.

Here are all possible bets today, will use BFSP as the tally...

Code:


DARKY: 03/12/2011 WAGGA 
R 6 # 2 192x HEAD OVER               
DARKY: 03/12/2011 TOOWOOMBA 
R 1 # 6 67x1 CHARMING BAY
R 3 # 2    2 SAVAGE              
DARKY: 03/12/2011 SCONE 
R 7 # 5 421x MAGASIN     
DARKY: 03/12/2011 MORPHETTVILLE 
AR 2 # 2 1241 IMPERIAL FURY  
AR 8 # 1 25x2 BIT OF A DUDE     
DARKY: 03/12/2011 GOLD COAST 
R 3 # 4 25x2 QUIZETTE  
DARKY: 03/12/2011 EAGLE FARM 
R 1 # 1  331 OPTIONALITY 
R 5 # 1 11x2 ROCKET TO GLORY  
DARKY: 03/12/2011 CRANBOURNE
R 3 # 2 3432 LITTLE HOPE         
DARKY: 03/12/2011 ASCOT 
R 1 # 2 16x2 DE LATTE      
R 4 # 3 2162 THE ENTOURAGE

Here are the results when laying the field using UB's prices to demonstrate:

There were 38142 horse form lines in the database at this point in time. We update when we get time.

There were 38142 selections for the System
There were 34258 which Lost (did not win race) for the System for a strike rate of 89.82%
You had to pay out $42188.79 but brought in $36234.9 after 5% commission. This means a Profit or loss of $-5953.89 or a percentage profit/loss of -15.61%

I'm afraid that is way way out.

UB,
First of all, excellent effort mate. very good of you to share.
Just a suggestion, can you use odds from one of the totes to add to your database?
That way if you find a system to be profitable on the database you can be confident of getting better results or equal by using betfair/top fluc etc.
If you can find a system to show a profit on UNITAB etc your laughing!!

Thanks to Chrome, I have a few..

Quote:

Originally Posted by Chrome Prince

Chrome,

I use worst betfair price on offer 10 secs to jump. So if your betting its the bet side and if your laying its the lay side. I only include prices from races which are 95% market percentage. The difference is mainly due to the longer shots that get up. You can have a discrepancy of $100 - $600. If you were backing the site uses the $100 if you were laying it uses the $600. Betfair SP might make it somewhere closer to the $100 from what I have seen. This can make a huge difference to the return. It also takes into account commission ( but I admit there is a flaw in that it doesn't take into account losses in the same race). So if you assume 5% of that loss is always commission, you are only losing 10% on the worst prices.

Looking at the back vs lay prices :

Backing returned : $35660.44
Laying returned : $42188.79

There is a 7K difference in the odds and if you layed at the lower odds you would actually be in profit. It all depends on the odds you get so I always take into account worst case for system tests.

Hi UB,
That's fair enough and I'm not having a go at you, just pointing out the big gap.
Perhaps a lot of the problem is low liquidity on outsiders with 5 seconds to jump, when many races can run well overtime.
At 95% markets, we can assume that the outsiders are well under their price and favourites are well over.

Quote:

Originally Posted by Chrome Prince

Chrome Prince,

I didn't think you were having a go as we have very similar ideas and thought processes. I always enjoy any ideas you come up with.

I'm doing very well in my test so far of the winning without form idea you brought up (and I put a slight twist on). I need another 3 weeks of data though before I incorporate it into a live test with small bets. If it goes well I'll share the results and twist.

5/5 Successes .

Cheers.

Quote:

Originally Posted by darkydog2002

5/5 Successes .

Cheers.

Very nice work,.

Thank you.

Cheers
darky