Testing

[michaelg]

Because the Place method is doing so well I decided to apply the D.S. market and Unitab to another method.

It has done surprisingly well over the past three days but then I thought the D.S. ratings could have changed after the race.

So I've identified today's selections to see if the results are near as impressive. I'm backing them each-way, even though there have been some outstanding quinellas.

The method/selection process is:

1) Single prices in D.S. ratings,
2) 96-plus points in Unitab ratings.

Today's selections are:

Swan Hill
1/2
2/1, 2
4/1, 7
5/2, 7
6/6, 8
7,2, 3, 4, 5
8/6

Wellington
2/4, 11
3/3
4/2, 5, 6
5/5, 11
6/4
7/12

Cessnock
1/3, 4
2/1, 2, 3
6/14
7/5, 6, 10

The first two races at Swan Hill have already been run. There's been three selections for one winner of $3.30 and all three have placed for a return of $4.00. The quinella in race 2 ($1.00 outlay) was snared for a return of $7.

I've just returned from a specialist who sent me for a brain scan...no, not to see if I've got a brain, but because I've got very mild vertigo. It's not serious but after two and a half years I've decided to do something about it. I go back to see him next week.

[evajb001]

Hi michaelg,

I'm not sure if your into staking ideas/plans as well. I wonder if this method would benefit from adjusting the staking based on if its got a higher rating or a lower DS price?

i.e. a low DS price combined with 100 rater might be a full unit bet, however a high DS price near $10 and a 96 rater might only be a half unit bet. Then anything in between can be adjusted with a simple equation, just something to consider.

Also I found in my backtesting so far that 100 raters win quite frequently, then there is a reasonably drop to 99 raters, then it increases a bit again for 98, 97 and 96 then starts to taper off again. That may be something to explore as well, can post the data for you if your interested.

[SpeedyBen]

I've just returned from a specialist who sent me for a brain scan...no, not to see if I've got a brain, but because I've got very mild vertigo. It's not serious but after two and a half years I've decided to do something about it. I go back to see him next week.Steady on, Michael.

[darkydog2002]

You always stimulate my mind Michaelg.
Thank you.

[michaelg]

Thanks for the suggestion, evajb001.

However at the moment I'll stick with just betting them level stakes, but if it continues to perform I'll certainly look at your suggestion.

At the moment there have been 8 races already run, for 15 selections producing 5 winners totaling $19.30 for the Win and $14.90 for the Place.

[Lord Greystoke]

Also I found in my backtesting so far that 100 raters win quite frequently, then there is a reasonably drop to 99 raters, then it increases a bit again for 98, 97 and 96 then starts to taper off again. That may be something to explore as well, can post the data for you if your interested.

Hi evajb001, pls post if you can - will be interesting to see how it pans out.

Cheers LG

[The Ocho]

Good luck with the vertigo michaelg. I didn't really like it as a movie but I hope they sort it out for you.

[michaelg]

Thanks, Ocho. It's very mild and I only get it when I stand up.

Pun intended, Speedyben? Quite humourous.

Today was not the best of days for the method. From 33 selections the Win return on NSW TAB was $28.90 and $27.10 for the Place.

The quinella outlay was $22 and the return was $50.

However if I had kept the maximum runners at 11 as I do with the Place system then the results would've been:

Outlay of $27
Win return of $28.90
Place return of $22.10
Quinella outlay of $21
Return of $50

From tomorrow I'll list the selections for fields of a max of 11 runners.

[ianian]

Under tab no 7
6 bets
1 win 2.40
1 place 1.30
Simple idea and simple to track may help with milti selections

[evajb001]

Here you go gents for interest of some stats, sorry to get in the road michael, if you wish can get a mod to delete the post and i can put in another thread.

The following is a chart of the Unitab Ratings for all winners in races that i've completed my ratings on since 1 April 2012. In general these are races with greater then Class 3 rating:

http://www.puntingprophets.com/racing/unirate.jpg

The following charts are firstly the stats for days since last start for all winners across the same time frame. It would seem 14 is the sweet spot as well as 6 days either side of this. The second chart is the days since second last start, and as the majority are 49+ it essentially shows that horses 2nd up seem to do the best.

http://puntingprophets.com/racing/daysls.jpg

Finally the following 2 charts are as follows, the first chart is the lengths the winner of the race was behind the winner of its previous race (if that makes sense). The second chart is the stats based on a horses winning % in the conditions, track and distance. -1% means it hasn't ran in that element before. Essentially this shows that horses that have ran multiple times and won in the conditions have greater chance of winning again while its stats at the distance or track are negligible with quite a few winners never even running at the track/distance before.

http://www.puntingprophets.com/racing/length_stats.jpg

Again sorry for clogging up thread, just figured these may be of interest for some. Obviously not a very large sample size (about 2066 races) but its a starting point.

[SpeedyBen]

The second up figures make a mockery of one of racings "truisms".
Fascinating figures all round.

[evajb001]

It certainly surprised me Speedy, Those charts basically show that running every 2 weeks is good for fitness but you don't necessarily have to be 2-3 races into your prep, as long as you've had 1 recent start your good to go.

I was also really surprised that a horses track and distance W% had little bearing on its ability to win. To me it just shows that conditions have a huge bearing on the outcome of a race but I think most here know that swimmers win on wet and others like it firm and they don't generally mix between the two.

Anyway sorry for getting off topic of your testing michael, I hope the Tatts Rating chart helps somewhat.

[michaelg]

Thanks for your suggestions.

However at the moment I'll keep it simple even though I'll keep in mind the suggestions. The only filter I'll currently apply is that the maximum number of runners per qualifying race is 11.

Today's races and selections are:

Townsville
2/1, 3
3/1, 3
4/4, 5, 9
5/6, 9

Seymour
4/3, 6, 7
5/1, 5
7/9, 10, 11
8/1, 3

Taree
1/1, 8
4/1, 4
7/4

I'll be betting them to Win and box quinellas.

Win outlay is $24.
Quinella outlay is $15.

[ianian]

So far the races with 3 or more chances dont seem to do as well as races with 2 or less chances .But very good so far given all chances are still in.

[michaelg]

Ianian, you're right. Today there were 3 races of 3 selections, for an outlay of $9 the return was only $2.00, and yesterday there were 13 selections for a return of $10.80. It's still too early to tell but it's something for me to look at.

Today there were 24 selections for a NSWTAB return of $23.50 for a loss of $0.50, but on Betfair S.P the return was (including their commission) $26.45 for a profit of $2.45 or a 10% POT.

The method struck 7 winners from the 11 races, and one of the selections ran second in three of them. And there were 3 quinellas for a loss of $0.50. Even though it was a losing day it was not a damaging one.

I think my brain scan has done something worrying to my brain because I made an appointment with my specialist for next Tuesday without at the time realising it is probably the most important day of racing.

[jose]

That is a catastrophe MG,

[SpeedyBen]

Michael
I'm sure you would agree that it would be hard to make a profit backing 3 or 4 fairly short priced horses for the win. However, quinellas and exactas may stand a chance of making a profit. I'm keeping a record of various permutations of exactas and will post the results when you have posted more selections. So far it is very encouraging in the 3/4 seln races.

[michaelg]

SpeedyBen, yes, it's difficult to show a Win profit from short priced horses, but hopefully this method will ultimately be profitable due to Betfair's superior prices even though I'm currently betting with the TAB.

I will also continue betting quinellas, and again hope they will be successful but am not too confident because of the TAB's high take-out.

I'ts encouraging to know you are doing well with quinellas/exactas using small combinations as I am currently doing, and will be very interested to learn of your results.

[michaelg]

Today's selections are (if there are more than 50% selections in a race it becomes a no-race).

Ascot
2/8, 9

Eagle Farm
1/3, 6
3/2, 6
4/2,11
5/1, 3
6/10
7/1, 4, 6

Kembla
2/2, 3, 4
3/5, 7
4/10
5/1, 2, 6
6/2, 10, 13

Bendigo
1/3
2/8
3/3
4/11, 12
5/2
7/3, 10, 13
9/5, 6, 10

Balaklava
1/2
2/3
4/3
5/2
7/1, 3, 6

Launceston
4/5
6/1
7/1

Win outlay $50
Quinella Outlay $29

I'm checking the Win results of those races where there are 3-plus selections.

[ianian]

Kembla scr 5.1

Kembla Race no 6 SHOULD BE RACE NO 7 ??

[michaelg]

Thanks, ianian.

It should be Kembla R7 not R6. And Kembla R5 is not a qualifying race because there are more then 11 starters.

[ianian]

Hi dont mean to interupt so often, but have you noticed that it still the single top, 100 rated that have been the winners for the last couple of days.

tues rough
2.80
L
L
2
L
2
5.5
wed
21.00
l
l
2.25
2.20
l
l
l
4.40
l
1.24
3.40
l
6.10
l
l
l
l
l
l
l
so very close and kept most winners???

[michaelg]

Thanks for that, ianian.

I'm somewhat rushed at the moment so I can't check it out yet.

Yesterday was a poor day overall. For the outlay of $45 the return was $39.40 on NSWTAB and $55.19 on Betfair S.P.. But when only looking at fields of a max of 9 runners the outlay was $25 for a NSWTAB return of $25.10 and a Betfair S.P. return of $44.10 for a POT of 76%.

So I'm listing today's selection in fields of 9 runners and less.

Ballarat
2/1
3/4
4/5
5/2
6/1, 3, 5
7/6

Gosford
1/3
2/4
3/1, 2, 3, 4
4/1,2
5/2, 3
7/1, 2, 4

Rocky
1/1
2/5
3/2
4/1
5/3, 8
6/1, 3

Northam
2/1, 5
3/1
5/2, 8
6/4
7/3, 5