Unitab 100 Pointers
 

I was looking at Unitab ratings where there is an outright 100 Pointer and was unplaced at its last start - the results from 1 Aug are quite promising. Maybe Unitab sees something in them. For example, the selection received interference at its last start, or was badly ridden, etc. The results are (Unitab divvies):

Tuesday 1 Aug.
6 selections.
2 winners for a return of $7.60
4 placegetters for a return of $8.10

Wednesday 2 Aug.
20 selections.
9 winners for a return of $33.30
13 placegetters for a return of $21.40

Thursday 3 Aug.
18 selections.
6 winners for a return of $33.10
12 placegetters for a return of $27.40

Overall, there have been 44 selections.
15 winners for a return of $74.00
29 placegetters for a return of $56.90.

If anyone's interested, today's selections are:

MACKAY
R1 no.2
R3 no.7
R4 no.2
R5 no.7
R6 no.5
R7 no.7

GUNNEDAH
R3 no.4
R5 no.1
R8 no.9

SWAN HILL
R3 no.5
R6 no.6
R7 no.8

QUEANBEYAN
R1 no.3
R2 no3
R4 no.4
R5 no.3
R6 no.3
R7 no.3

From yesterday's eighteen selections there were five winners paying $20.70, and eleven placegetters paying $19.20.
Twenty six selections today:

ROSEHILL
R6 no.7

CAULFIELD
R1 no.2
R4 no.3
R5 no.8
R7 no.1

DOMBEEN
R1 no.5
R3 no.4
R4 no.5

CHELTENHAM
R1 no.9
R2 no.6
R6 no.13
R7 no.16
R8 no.9

GOLD COAST
R2 no.10
R3 no.13
R5 no.9
R8 no.3

WODONGA
R3 no.2
R6 no.2
R7 no.1

TOOWOOMBA
R2 no.1
R4 no.4

BELMONT
R2 no.5
R5 no.9
R6 no.11
R7 no.9

michaelg,

If I get around to it today I will write a program to get the results for you from this year ( Jan -> now ).

I might get a chance as my wife is sick in bed.

A couple of questions though ? Do you include races where more than 1 horse is rated 100 ?

Good Luck.

Wes,

Michael does say in his first post "outright " 100 rater. Would be interesting to see the results of your search if you are able to do it.

Michael,

Have been watching what you have posted here. Thank you. I'm a bit surprised that it has not elicited more comment. I went through last Saturday's selections in the hope that it might give us a bit of a lead on today. I know there are many factors; but.... for what it is worth:

1) Both win and place runners according to your rules made a profit --- the win bets just scraping in; place returning some 40%.

2) I scored all meetings on which Unitab gave an outright 100 rater. Cheltenham, Darwin, N.Z., and W.A. did not have a winner (out of a combined 7 starters), and Eagle Farm only one winner in three. (I am counting as a winner any return --- place or win.)

Whereas the other venues all scored with all their selections --- W.F 1; M.V 3 (one NTD); Wang. 2; Gold C. 3; Newcastle 3.


More importantly, I think, is to get some understanding of just what the 100-rater means. I suspect that they are produced by some sort of computer programme (Does any one know more about this?); whereas other threads here seem to suggest that the Unitab picks at the top, as well as the named tips, are produced by "humans". (Again, does anyone know for certain?)

Of course, this brings up the question as to the current value of the ratings. Is it a 100 rater because some time ago it had a flash in the pan and won three in a row in the bush, etc., etc..

Anyway, my tuppence worth. Good luck today.

Both Unitab and Techform ratings are computer generated - no human input.

Unitab concentrates on the peak recent run. Techform concentrates on the career peak run.

Thanks, Chrome. More grist for the decision mill :) .

I was looking at trying to cut down the number of bets, so going from the "bet everyday" thread, i looked to see if any selctions were in the top three radio tab
Wesmip. are you able to give separate results for metro and non metro?

MACKAY
3r x 2w $9.00, 3pl

GUNNEDAH
2r x 0

SWAN HILL
2r x 2w $4.90

QUEANBEYAN
4r x 0w, 3pl

11r x 4w $13.90
$13.20 place return

Note that Doomben R1 No. 5 is not a sole 100 rater.


The general impression that I got in doing the above research was that the 100 raters with last start "0" were sus., and that it would pay to be wary about 100 raters coming back from a spell.

I am finding this thread most interesting
 

Hi all, I havent been to this site for a little while now, and decided to look again, with a great deal of interest in this 100 point theory. Can anyone tell me in general terms how does on go about having a data base whereby you can just go back to say january which was mentioned and come up with all kinds of results,..???,.......I would love to be able to do this.

Good luck with this, I'll keep my eye on it, and look forward to seeing these past results.

Paul

Thanks for the input, everyone. I am under the impression that 100 pointers with Techform are not as accurate as Unitab, but that may not be the case even though I applied the rules to it from 1 Aug and it showed a loss for win and place.

I did not check last Saturday with Unitab but am pleased to see that they showed a profit, especially because my Saturday neurals can be hot and cold - I am under the impression the Unitab ratings are to a degree connnected to the neurals.

Merriguy, you are right - Dombeen R1 no.5 is not the sole 100 pointer and therefore it is not a selection. I hope it does well because I've already backed it. I have now checked those that finished last start with '0' and also first-uppers. The results are:

'0' last start.
7 selections for 3 winners paying $16.30. 4 placegetters paying $9.70.

First-up
6 selections for 2 winners paying $8.10. 3 placegetters paying $4.80

First-up with '0' last start.
3 selections for 1 winner paying $4.30. 2 placegetters paying $3.60.

So far it appears that they are worthwhile backing.

Wesmip1, I'm also interested to see what your research throws up.

Stay tuned Paul, I'm currently working on adding a REVISED version of Techform ratings to the database, whereby top rated horses will show a level stakes profit.

There won't be as many 100 raters, as currently are on the TAB website, as I've identified which horses should be 100 and which ones shouldn't.

Possibly next month it might be fully ready.

Due to copyright restrictions, I cannot reproduce the ratings exactly, but I can tweak them to provide my own ratings.

Of course, my amendments will be disclosed to purchasers, so they can use this information for upcoming races.

All,

Ok I have up to the 24th Feb loaded into the database. Thats enough for some preliminary data.

100 Raters (All of them)
Selections : 2712
WIN : $ 2416.30
PLACE : $ 2431.30

100 Raters ( Only 1 Selection in the race )
Selections : 2018
WIN : $ 1781.30
PLACE : $ 1813.50

All 100 Raters ( Won last Start )
Selections : 480
WIN : $ 411.90
PLACE : $ 419.20

All 100 Raters ( Placed last start meaning 1st, 2nd, 3rd )
Selections : 1196
WIN : $ 976.40
PLACE : $ 1071.40

All 100 Raters ( Not Placed and not resuming )
Selections :965
WIN : $ 910.00
PLACE : $ 871.20

All 100 Raters ( Not Placed and not resuming and Only 1 Selection at 100)
Selections : 684
WIN : $ 647.50
PLACE : $ 637.40

All 100 Raters ( Resuming )
Selections : 225
WIN : $ 227.10
PLACE : $ 211.60

All 100 Raters ( Resuming but last start Unplaced)
Selections : 187
WIN : $ 192.80
PLACE : $ 172.60

All 100 Raters ( Resuming but last start Unplaced and only 1 selection in race)
Selections : 143
WIN : $ 131.80
PLACE : $ 126.10

I'll get a few more months just to confirm the results....

Good Luck.

HI WESMIP,

CAN YOU PLEASE TRY UNI-TAB OUTRIGHT 100 RATERS WITH A T-D-C, ONLY NEED TO HAVE ONE OFF THEM. NO 1ST UPS THOUGH.NO HEAVY TRACKS TOO.
WOULD APPRECIATE IT, IF IT IS POSSIBLE.
THANKS
LEROY

leroy,

It will take me tonight to get up to date with just form, rating and who won/placed.

I will then run an update to grab the C-D-W out and run it for you.

If I get a chance I should ahve it done by tommorrow afternoon. I'll let you know how I am going.

Thanks.

Wesmip1, thanks for the info. The last-start unplaced selections seemed not to lose as much as those that placed at their last start.

As stated previously, I am under the impression the Unitab ratings could be connected to the neurals, which I don't think are generally too successful on Saturdays. Today was a wipeout, in fact the only "non-Saturday meeting" was Wodonga where there were three selections for one winner of $3.00 and all three placed paying $5.30.

Is it possible to delete Saturday and Public Hols from your results?

Thanks.
Michael.

Thanks wesmip,
Just when ever you ge a chance will be fine.
I forgot to mention , mainly interested in Monday -Saturday non-metro.
Vic,Qld,NSW and S.A
Thanks.

Leroy

Why oh Why would you leave some venues out??? They all pay with the same currency...

Today: Darwin Race 2 #6 $28.10 and $8.00 the drum... was the only one with a D besides its name as well for those interested..

All,

I have up to the end of June.

Everything is showing a loss.

All 100 Raters ( Not Placed and not resuming and Only 1 Selection at 100)
Selections : 2253
WIN : $ 1910.3
PLACE : $ 2000.4

100 Raters (All of them)
Selections : 8575
WIN : $ 7230.7
PLACE : $ 7491.3

All 100 Raters ( Won last Start )
Selections : 1466
WIN : $ 1322.3
PLACE : $ 1324.1

All 100 Raters ( Placed last start meaning 1st, 2nd, 3rd )
Selections : 3688
WIN : $ 3222.6
PLACE : $ 3322.1


All 100 Raters ( Resuming )
Selections : 814
WIN : $ 565.7
PLACE : $ 694.5

All 100 Raters ( Resuming but last start Unplaced)
Selections : 623
WIN : $ 440.3
PLACE : $ 529.5

All 100 Raters ( Resuming but last start Unplaced and only 1 selection in race)
Selections : 494
WIN : $ 346.7
PLACE : $ 420.2

Good Luck....

michaelg, leroy,

I can remove Saturdays as I store the dates.

I will run the query without Saturdays ( I assume you still want Sundays included ).

Leroy,

I will get the CDW data tommorrow hopefully. Then I can check your criteria.

Good Luck.

Without Saturdays

All 100 Raters ( Not Placed and not resuming and Only 1 Selection at 100)
Selections : 1736
WIN : $ 1479.5
PLACE : $ 1551.5

Still a losing system.

wesmip1,

Can I suggest that the filters you are providing are not sufficient.

Maybe think about why they are all losing money and what factors need to improve the performance of the horses as opposed to the performance of the ratings.

Bear in mind that the Unitab ratings are ratings based on peak recent performance, therefore they will always be unders unless you find an angle different to form , as they are form based already.

What about class rise/drop, gear changes, country horses that come to town and are 100 rated, days since last run - fitness (not just first up or not), weight, not just placings but margins etc etc.

The edge is there, you haven't wasted your time, you might need to look elsewhere within the data.

Think about it this way, lets say that a 100 rated horse is the class horse of the race with the best form, of course it will always be unders, what else should improve the strike rate yet improve the average dividend?

hey wes do you have strike rate and run of outs?
and i reckon you shouldnt be using unitab div,s bound to be overs elsewhere

Chrome,

Very true. I ran this quickly to test the system proposed. I didn't do it to try and find extra filters.

System,

Of coure you can get better than unitab divi's but if you can make a profit at unitab divis then you will usually only improve that elsewhere.

The strike rate is available but I don't have my database open at the moment. Going out this morning so I will post it later today.

Thanks for all the effort put into posting the results above, Wes. Much appreciated --- not only by me, I'm sure.

System,

Is there any particular one you want the strike rate for ?

Here is the info you requested for the last one I posted :

All 100 Raters ( Not Placed and not resuming and Only 1 Selection at 100 )
Selections : 3219
WIN : $ 2745.90
PLACE : $ 2869.80
No of Winners : 699 (21%)
No Of Places : 1687 (52%)
Longest Run of Outs WIN : 28
Longest Run of Outs PLACE : 9

Hope that helps you out... Thats a fairly long losing streak....

Note I am getting 2005 data as well so this comprises Jan 2005 - Mar 2005 and Jan 2006 - Aug 2006 data.

Good Luck.

Chrome, have you done the excercise: when Unitab and Techform agree with clear top rater??

Partypooper,

Yes I have, shows more of a loss.

What we have is a situation where the class horse of the race is also the one with the best form. Of course, it gives a very high strike rate but a loss on turnover.

Therefore, the Techform ratings actually give a better return with EXTRA filters, rather than Unitab with extra filters. This is only from the research I've done to suit my systems, but it seems to reflect the same across the board.

So then you have Techform 100 and Unitab 100 equalling poor odds or overbet horses.

- Unitab 100 rater BY 4 POINTS OR MORE
- Unplaced last start
- Won 2nd last start
- $2.00-$10.00

i think this should produce a profit

100 raters by 4 points or more on its own is a great starter IMO

Chuck,

Only thing I couldn't do was $2.00 - $10.00
Price is only good if you have the time to sit there and watch it and then who says whether you should take it if its $10.10 or $1.90 with 30 secs to jump. Thats why I don't keep it ( only keep prices of winners and placings ).

Results were :

Selections : 136
WIN : $116.9
PLACE : $145.5

This is over 16 months worth of data.

thanks mate how about simply:

100 pointers by 4 or more
100 pointers by 5 or more
100 pointers by 6 or more

cheers for any info

Chuck,

I need to setup the indexes on this database ... I started running the
100 pointers by 4 or more and it has been going for 12mins already.

Ok Index was created and here are the results :

100 pointers by 4 or more
Selection 3037
Win : $2729.30
Place : $2775.80

100 pointers by 5 or more
Selection :945
Win : $837.80
Place : $878.00

100 pointers by 6 or more
Selection :512
Win : $448.50
Place : $483.20

100 pointers by 7 or more
Selection :333
Win : $313.4
Place : $316.1

100 pointers by 8 or more
Selection :215
Win : $197.20
Place : $207.90

100 pointers by 9 or more
Selection :55
Win : $53.40
Place : $48.80

100 pointers by 10 or more
Selection :27
Win : $34.60
Place : $27.70

Hope that helps....

thats great

its very interesting...all narrow losses (except one profit) sure i'm sure with the right filters we can come up with a profit

IMO you have left out one main filter that i have always used, pick your races now my type of race may not be yours but i only bet in races that are OPEN and higher.

No 3yo only races
No Fillies & Mares races
No Restricted of any kind

It has to be Open, Quality, Group races even the country events have open races.

This may not be everyones idea of race filtering but it has worked or me.

Take it that 2yo and jumps are out to.

Leroy,

You requested the following :


Here is the results :
Selections : 3156
WIN : $2640.00 (-16%)
PLACE : $2795.50 (-11%)

Anymore ideas want testing ?

Good Luck.

Tab no. 1-6
Barrier 1-6
Unitab highest rater by 4 or more
not run a place last start

Wesmip1,

As you asked (hee hee) how about the following:

Delete- Jumps, Amatuer, Fillies&Mares only, Apprentice /Ladies only/& races of 2000m+

Unitab + techform 100 pointer
Last start METRO winner, 7,14, or 21 days only
Tab-Form pre-post fav, providing it is not carrying more than 60 kg's (joint favs ok)
Place S/R 50% +
Placed at least once in todays distance (i.e. 1st 2nd or 3rd)

Bets placed at TOP FLUCTUATION only:

Races with 2 or more qualifiers = NO = BET

How about this...
100Pter
Finished 5-9th LS
No Resumers.

Thanks Wesmip.
No ideas at the moment,but thanks anyway.
Thanks also for all the work your doing on this thread.
I'm sure between everyone we can find something that works.

Leroy