Test Last Start winners

[michaelg]

Unfortunately I haven't recorded the results but from what I can recall it seems there could be some merit with each-way betting, so I'll now record them...and hope.

Coonamble
5/1

Cranbourne
8/13

Kalgoorlie
3/1
4/6
6/9

Murwillumbah
5/8

Sun Coast
6/2
7/3

W'bool
1/1

[Shaun]

If you restricted that to 7 days and prices under $5 you might just make a profit.

[The Ocho]

Ha ha ha.

I have been lay testing this for the last week and a half as I thought as I looked at the form that there really aren't that many 11 form lines. Most times the horse doesn't win next start and so I thought that may make it a good laying system. It is also something that can be back tested since time immemorial (or at least since form guides have been produced).

I tested it both at level stakes and a $30 liability (so you could do the minimum bet on Betfair) and, up until yesterday, it was looking good at laying between $6-$15. Under $6 actually backing the horses was showing a profit. Yesterday (Saturday) turned everything on it's head with big level stakes and liability losses.

I hope you can come up with something with the each way thing.

By the way, I have a lot more horses than you for today that either came first last start or first before a spell (which I have also been trying). Yesterday there were 86 horses which fell into the above category and today there are 27 races with qualifying horses (some races there are more than 1 selection).

Good Luck.

[michaelg]

Shaun, I've previously had a look restricting the selection's last start to 7 day's, but from what I can remember it didn't make too much of a difference.

There are 3 selections today whose last start was seven days ago - they are the Kalgoorlie selections. We'll see how they go.

The Ocho, I don't select every last start winner. There are filters including if it won its last two starts then I omit it.

[darkydog2002]

Michaelg,
Have you considered the Don Scott "Price asessment" as a filter with this.
i.e
Sunshine Coast
Race 6 H 2 (asessed @.2.70
Race 7 H 3 $3.10
Murwillumbah
R 5 H 8 $3.00
Kalgoorlie
R 3 H 1 $2.45
R 4 H 6 $2.10
R 6 H 9 $1.50

CoonambleR 5 H 1 $2.75
Cranbourne
R 8 H 13 $4.20
Warnambool
R 1 H 1 $1.16 (steeple race)

Cheers
darky

[michaelg]

Darky, I haven't considered the prices.

I'm might not be at a computer during some of the arvo, however I could put a price limit on Betfair S.P.

Interestingly there have been two winners so far:

W'bool 1/1, rated $1.16 paying $1.40
S.Coast 7/3, rated $3.10 paying $5.80.

[Stix]

Value last start winners.....:cool:

Using R&S
1) Last Start Winner
2) Not a Track Winner
3) Not in UniTab top 3 selections
4) 100 UniTab 100 ptr rater
5) Good Track Only

Last Start winner = seems decent place to start
Not a Track Winner = Track winners are over bet, we are looiking for a better price
Not in Top 3 = again, not over bet and not a favoured runner
100 pt rater = no a hopeless conveyance *shrug*
Good track = no excuses!!

S/Rate 25%
Av Div $5.6
POT 39%

[michaelg]

Thanks, Stix.

I assume it applies to the top neural rater using the default settings?


From yesterday's 9 selections there were 3 winners for a return of $10.40 but a return of only $3.84 for the Place, two selections ran fourth.

Today's selections:

Goulburn
4/7
6/5
8/1

[michaelg]

I had a look at yesterday's qualifiers but were not the final selections because they did not win their last start.

The ones that ran second at last start did not fare too badly. There were 12 of these for 3 winners paying a total of $13.00, and 8 placegetters paying $13.10, giving a small each-way profit.

Out of interest, today's selections for those that ran second at their last start are:

Echuca
3/13
5/14

Goulburn
2/1
8/3

We'll see how they go.

[darkydog2002]

With the Don Scott price asessment I recall a gentleman telling me those rated at $1.50 or less had a very high strike rate.
What minimum price acceptable would be up to you but food for thought.
Cheers
darky

[michaelg]

Thanks, Darky,

some time ago I did check exactly that, i.e. the max price of $1.50 in his market. Unfortunately it did not perform as I had hoped it would. However I don't think I applied any filters to it.

[Try Try Again]

Hi Darkydog2002 and Michaelg,

As you know I only look at Don Scott's rated Favourite on NSW and Vic meetings on Good/Dead tracks and have recorded the results since 20th March 2011. For Don Scott Favourites rated at <= $1.50 there have been 131 selections for 48 winners (S/R 36.6%) for LOT% = 0.2%. Results are based on Citybest+ for city meetings and Country Best for country meetings.

For all DS odds-on rated Favourites there have been 286 selections for 82 winners (S/R 28.7%) and LOT% = 10.7%. Thus rated DS favourites >$1.50 but <$2.00 are performing very poorly with 155 selections for 34 winners (S/R 21.9%) for LOT% = 19.1%.

Admittedly the sample size is not large but at least it is growing each day!

[michaelg]

Thanks for that, Try.

Have you got records of where they finished at last start?

[Try Try Again]

Hi Michaelg,

No sorry I don't record that info.

I record Class of race, Distance, number of starters, Horse's name, Don Scott rating, SP, Finishing position (Won, 2nd, 3rd or unplaced) and Win & Place dividends.

[beton]

On my data =<$1.50 tote has a 60% strike rate.+<$2 tote fav has a 50% strike rate. All races all conditions. Hence these DS favs are performing well under. Beton

[darkydog2002]

Try Try Again and Beton,
Thanks for that.Appreciated.
Cheers
darky

[Stix]

Thanks, Stix.

I assume it applies to the top neural rater using the default settings?


using the old Trainer/Jockey system neural settings