PDA

View Full Version : Ratings Updates


UselessBettor
31st December 2011, 09:39 AM
I thought I would post a ratings update on how particular ratings are going since I uploaded a bit more data last night and this morning.

See below:

====

Back the Favourite


There were 45200 horse form lines in the database at this point in time. We update when we get time.
There were 4667 selections for the System
There were 1589 Winners for the System for a strike rate of 34.05%
There were $4622.34 returned for WIN (after 5% commission) which means a Profit or loss of $-44.66 or a percentage profit/loss of -0.96%

The Rules used were : priceRank < 90 and priceRank = 1 and priceLayRank < 90 and priceLayRank = 1
====
Back top Unitab Rating


There were 45200 horse form lines in the database at this point in time. We update when we get time.
There were 4633 selections for the System
There were 1064 Winners for the System for a strike rate of 22.97%
There were $4240.77 returned for WIN (after 5% commission) which means a Profit or loss of $-392.23 or a percentage profit/loss of -8.47%

The Rules used were : uniRank = 1


====
Back top Don Scott Rating


There were 45200 horse form lines in the database at this point in time. We update when we get time.
There were 4290 selections for the System
There were 1022 Winners for the System for a strike rate of 23.82%
There were $4239.7 returned for WIN (after 5% commission) which means a Profit or loss of $-50.3 or a percentage profit/loss of -1.17%

The Rules used were : donScottRank = 1

====
Top Neural (default setting)

There were 45200 horse form lines in the database at this point in time. We update when we get time.

There were 3943 selections for the System
There were 913 Winners for the System for a strike rate of 23.15%
There were $3629.91 returned for WIN (after 5% commission) which means a Profit or loss of $-313.09 or a percentage profit/loss of -7.94%


The Rules used were : nrRank = 1 ====
Top Place Strike Rate (> 0 placings)


There were 45200 horse form lines in the database at this point in time. We update when we get time.
There were 4466 selections for the System
There were 1030 Winners for the System for a strike rate of 23.06%
There were $4380.78 returned for WIN (after 5% commission) which means a Profit or loss of $-85.22 or a percentage profit/loss of -1.91%

The Rules used were : placeStrikeRate > 0 and placeSrRank = 1
====
Top WIN Strike Rate ( > 0 wins)



There were 45200 horse form lines in the database at this point in time. We update when we get time.
There were 3261 selections for the System
There were 683 Winners for the System for a strike rate of 20.94%
There were $3254 returned for WIN (after 5% commission) which means a Profit or loss of $-7 or a percentage profit/loss of -0.21%

The Rules used were : winStrikeRate > 0 and winSrRank = 1
====


Test Another System (http://testyoursystem.000space.com/TestSystem.php)


Try combining a few of these ratings and you might show a profit.

UselessBettor
31st December 2011, 09:46 AM
I combined 2 ratings (Win Strike Rate Rank + Don Scott Rank):



There were 45200 horse form lines in the database at this point in time. We update when we get time.

There were 1770 selections for the System
There were 542 Winners for the System for a strike rate of 30.62%
There were $1879.34 returned for WIN (after 5% commission) which means a Profit or loss of $109.34 or a percentage profit/loss of 6.18%

Test Another System (http://testyoursystem.000space.com/TestSystem.php) The Rules used were : donScottRank = 1 and winSrRank = 1




Could also try laying the poor ratings witha few filters:


There were 808 selections for the System
There were 524 which Lost (did not win race) for the System for a strike rate of 64.85%
You had to pay out $724.74 but brought in $767.6 after 5% commission. This means a Profit or loss of $42.86 or a percentage profit/loss of 5.3%


Test Another System (http://testyoursystem.000space.com/TestSystem.php) The Rules used were : barrier < 5 and rating = 100 and priceRank < 90 and priceRank = 1 and priceLayRank < 90 and priceLayRank = 1 and uniRank = 1

UselessBettor
31st December 2011, 09:49 AM
Maybe just back favourites on a good track:



There were 45200 horse form lines in the database at this point in time. We update when we get time.There were 2762 selections for the System
There were 963 Winners for the System for a strike rate of 34.87%
There were $2809.2 returned for WIN (after 5% commission) which means a Profit or loss of $47.2 or a percentage profit/loss of 1.71%

Test Another System (http://testyoursystem.000space.com/TestSystem.php)
The Rules used were : priceRank < 90 and priceRank = 1 and priceLayRank < 90 and priceLayRank = 1 and cond = 'G'

UselessBettor
31st December 2011, 09:55 AM
Or back favs going up in weight:

There were 1056 selections for the System
There were 368 Winners for the System for a strike rate of 34.85%
There were $1133.3 returned for WIN (after 5% commission) which means a Profit or loss of $77.3 or a percentage profit/loss of 7.32%

Test Another System (http://testyoursystem.000space.com/TestSystem.php)

The Rules used were : weightChange > 0 and priceRank < 90 and priceRank = 1 and priceLayRank < 90 and priceLayRank = 1

UselessBettor
31st December 2011, 09:57 AM
Another simple one.

Any favourite who has not won at the track and distance:

There were 2770 selections for the System
There were 989 Winners for the System for a strike rate of 35.7%
There were $2870.85 returned for WIN (after 5% commission) which means a Profit or loss of $100.85 or a percentage profit/loss of 3.64%

Test Another System (http://testyoursystem.000space.com/TestSystem.php)
The Rules used were : distWinner = 'N' and courseWinner = 'N' and priceRank < 90 and priceRank = 1 and priceLayRank < 90 and priceLayRank = 1

The Ocho
31st December 2011, 10:04 AM
Another simple one.

Any favourite who has not won at the track and distance:

There were 2770 selections for the System
There were 989 Winners for the System for a strike rate of 35.7%
There were $2870.85 returned for WIN (after 5% commission) which means a Profit or loss of $100.85 or a percentage profit/loss of 3.64%

Test Another System (http://testyoursystem.000space.com/TestSystem.php)
The Rules used were : distWinner = 'N' and courseWinner = 'N' and priceRank < 90 and priceRank = 1 and priceLayRank < 90 and priceLayRank = 1
That's a good one UB although you would have to be on hand to know who the fave is and if it hasn't won at the track and distance.

BY THE WAY - Does anyone know when the late mail selections get posted?

UselessBettor
31st December 2011, 10:08 AM
That's a good one UB although you would have to be on hand to know who the fave is and if it hasn't won at the track and distance.

BY THE WAY - Does anyone know when the late mail selections get posted?
Two words - Betfair Bot

Have no idea on the late mail selections.

The Ocho
31st December 2011, 10:28 AM
But the bot wouldn't know if the fave in the next race has or hasn't won at the track and distance.

UselessBettor
31st December 2011, 10:38 AM
But the bot wouldn't know if the fave in the next race has or hasn't won at the track and distance.
You would need to feed it a list which I think most bots can do.
Or do as I have done and write your own bot to do it all for you.

James
7th January 2012, 05:32 AM
You would need to feed it a list which I think most bots can do.
Or do as I have done and write your own bot to do it all for you.
Gday UselessBettor

You have written your own Bot! Is this hard to do? Do you need to be a programmer or can anyone do it?

Thanks

James

UselessBettor
7th January 2012, 07:17 AM
Gday UselessBettor

You have written your own Bot! Is this hard to do? Do you need to be a programmer or can anyone do it?

Thanks

James
There are enough examples on the developer betfair site that most people could at least get the examples working. Modifying them is a bit harder and you probably need at least basic programming knowledge but you cn learn that off the web in a few days/weeks. Just get "Learn Visual basic in 21 days" or something similar and you will learn enough to be able to modify the examples to do what you want.