PDA

View Full Version : Jack's Ratings system


jacfin
1st March 2004, 01:18 PM
Because today is a public holiday in WA I decided to do a bit of research on the Tabform ratings which are also the Unitab ratings. Are they in fact AAP ratings?
The results of the last 5 months for this system are 99 selections - 14 winners - avge divy $13 - LS profit 81 units.
The best winner was Roman Arch in Melb in the spring - $23.
You would need a big heart to follow a system with such a low strike rate but it achieved a good result if you had done so.
Will it continue to succeed? Probably not, but I've backed Sportsman ($17 winner)and Wishart ($14)from a handful of bets in the last couple of weeks so I'll continue on until the profits run out or ...... .
The rules -
Metro Sat only excluding Adelaide and Perth.
Consider any 100 pointer left after scratchings.
Back it if you can obtain $8.50 or more.
Ignore races where 3 100 pointers are >= $8.5.
I know that Sportz and others have looked at 100 pointers so if any of you have some extra rules which would improve the bottom line feel free to jump in.


<font size=-1>[ This Message was edited by: jacfin on 2004-03-02 11:45 ]</font>

<font size=-1>[ This Message was edited by: jacfin on 2004-03-03 12:06 ]</font>

<font size=-1>[ This Message was edited by: jacfin on 2004-03-04 12:18 ]</font>

sportznut
1st March 2004, 01:31 PM
Any idea how that system would have gone at country and provincial meetings? Desert Clearance a 100 rater just got up at Sale paying $8.60 here in Qld!!!

Anyway, I love the Unitab ratings and I've got one for you to look at:

1. At least 8 starters after scratchings.
2. Horse must be the only 100 rater left after scratchings.
3. Must be first up from a spell.
4. Must have won at this distance (c or d).

In the last 3 months, that simple system made around 85% profit and it also picked Sportsman.

Your system looks pretty good as it is, but if you want to improve things, I suggest you could look at winners at this distance (c or d) or last start beaten favourites.


<font size=-1>[ This Message was edited by: sportznut on 2004-03-02 08:39 ]</font>

jacfin
1st March 2004, 02:33 PM
Sportz
I knew that you'd have some good ideas to contribute. The distance winner seems like a good angle to look at. I'd like to get the hit rate up to 20% if possible even if it sacrificed some of the profit. I'll look into it when I continue my research.
I must confess that until this morning when I was doing the research I didn't know that unitab had ratings on C or P meetings. We get them for Sat meetings only in the TAB form guide. Something you wrote elsewhere made me think that they were perhaps the same as the unitab ratings and a spot check seems to confirm that.
Do you have the breakdown of win% etc for the first up system? It looks like a good one and shows your genius for finding 7 day a week systems. I have the utmost admiration for someone who can do this.

sportznut
1st March 2004, 03:01 PM
Around 30%.

I'm sorry, I haven't actually updated my figures, but up until last week, they were like this: 16 wins from 52 bets and a profit of 86%

I personally use a pre-post price range of $3.00-$21.00 and using that, the figures are like this: 14 wins from 46 bets and a profit of 101%

Hope that helps.

<font size=-1>[ This Message was edited by: sportznut on 2004-03-01 16:02 ]</font>

sportznut
2nd March 2004, 08:29 AM
Okay, not quite so good in the past week or two. Here are the updated figures:
Selections 59
Wins 18, Win return $102.30, Win profit 73%
Places 34, Place return $66.00, Place profit 12%

With $3-$21 pre-post price rule:
Selections 52
Wins 15, Win return $96.40, Win profit 85%
Places 31, Place return $63.90, Place profit 23%

As for your idea, I think it's an excellent one. As I was going through a few results yesterday, I noticed many good priced winners from the 100 raters on all days of the week. I'm going to have to look into that. :smile:

<font size=-1>[ This Message was edited by: sportznut on 2004-03-02 09:43 ]</font>

kenchar
2nd March 2004, 08:49 AM
Sportz,
Just curious do you check the selections 1st up strike rate or just bet them if the above rules apply.

Cheers

sportznut
2nd March 2004, 08:55 AM
Good point. No, these results are without checking for first-up strike rates. I'll have to look into that too. :smile:

Shaun
2nd March 2004, 09:03 AM
On 2004-03-02 09:29, sportznut wrote:
Okay, not quite so good in the past week or two. Here are the updated figures:
Selections 59
Wins 18, Win return $102.30, Win profit 73%
Places 34, Place return $66.00, Place profit 12%

With $3-$21 pre-post price rule:
Selections 52
Wins 15, Win return $96.40, Win profit 85%
Places 31, Place return $63.90, Place profit 23%

Gees sportz....any pro would give his first born to have a strike rate like that.....any longterm results

<font size=-1>[ This Message was edited by: sportznut on 2004-03-02 09:43 ]</font>

sportznut
2nd March 2004, 09:12 AM
Sorry, not at the moment. However, I am working on these 100 raters systems at the moment and by the time I'm finished, I'll have some figures for you. :smile:

Shaun
2nd March 2004, 09:40 AM
Sportz....in the above results what was the test period of those selections
how many selection a day would you say

sportznut
2nd March 2004, 10:36 AM
Shaun,
It's actually 100 days since the start of the test period, so about 4 bets a week.

Kenchar,
I've done a check and some of the winners hadn't even been placed first-up before. Having said that, horses with a good first-up strike rate were a bit more consistent, so I've used a rule that I often use for first-uppers to try to sort out the best bets:

At least 3 first-up starts, and at least 50% wins or 75% placings from those starts.
Nov20 CHbr 7-3 Papal Prince UNP
Nov29 Adel 4-5 She's My Baby WON 3.00/1.60
Dec13 Adel 8-4 Crest Of Gold WON 2.50/1.40
Jan23 Melb 4-5 Yell UNP
Jan25 MGam 2-2 Andalan WON 2.00/1.10
Jan27 Bena 8-1 Majestic Flush UNP
Jan28 Sydn 5-3 Beryl WON 5.50/1.90
Feb3 Melb 3-2 Penny Opera WON 3.70/1.40
Feb7 Melb 5-1 Lonhro WON 1.70/1.20
Feb11 Sydn 7-6 Derby Parade WON 7.50/2.30
Feb14 Sydn 6-1 Private Steer WON 3.10/1.50
Feb14 Sydn 7-3 Sportsman WON 17.40/3.70
Feb15 Colac 8-4 Ski Rebel PLA 1.50

Looks good!!! :smile:


<font size=-1>[ This Message was edited by: sportznut on 2004-03-02 13:25 ]</font>

pardon me
2nd March 2004, 10:40 AM
hi sportz, the unitab ratings are a good guide as u say and here is one that i have found seems to produce a place strike rate of over 50 % and ave divi around $2.50.
it also gets some good price winners but dont have stats on winners.
look at 3rd and 4th raters take the longest price of the two on unitab just prior to jump time providing it is$5.00+for the win.
if there are 2 on the same rating that counts as one.
i use it only on saturday but it may well work on other days as well.

sportznut
2nd March 2004, 10:44 AM
Sounds a bit like Kenchar's place strategy. Might be worth a look.

jacfin
2nd March 2004, 10:53 AM
I checked another month on the original system which produced two winners and a LS profit of 11 units.
That's 92 units profit over 6 months.
I'll continue on until I've done 12 months.
Both winners, Sportz, had won at the distance.
Please note that I've made a minor amendment to the original post eliminating races where 3 horses qualify. It's only happened 2 or 3 times in the six months.

sportznut
2nd March 2004, 10:59 AM
Good going. Yeah, I think if you deleted those that hadn't won at the distance, your overall strike rate would probably improve. The only thing is whether you would delete any really good priced winners by doing that.

michaelv
2nd March 2004, 11:08 AM
Sportz or Jacfin.....do either of you have a LINK ADDRESS for the unitab ratings? I've had a bit of a look around but can't seem to find it. :???: And, do you need to be a registered member or anything to gain access?

michaelv
2nd March 2004, 11:12 AM
Also, on the distance thing...what about horses that won or PLACED at the distance OR won or PLACED within 100m (+ or -) of todays race distance? I know it's a bit harder to research but it might grab those selections eliminated by the strictly c or d rule?

sportznut
2nd March 2004, 11:12 AM
Michael,

Totally free and VERY easy to use:
http://www.tabonline.com.au/default.stm

Simply click on Racing, then click on the race you want and the ratings are on the right side of the screen. :smile:

jacfin
2nd March 2004, 11:18 AM
It is pretty time consuming to check past results because each race has to accessed.
If we are thinking of checking each weekday perhaps we could have some volunteers and split the load.

sportznut
2nd March 2004, 12:44 PM
On 2004-03-02 12:12, michaelv wrote:
Also, on the distance thing...what about horses that won or PLACED at the distance OR won or PLACED within 100m (+ or -) of todays race distance?



Michael, I started checking my first-up system and I soon came to the conclusion that your idea would produce many more bets but not a lot more winners. I think if we stick simply to Distance winners, we should do okay. Alternatively, we could be very patient and just wait for the 'specials'. 9 wins from 13 bets looks promising!

Kenchar, thanks a lot for asking about those first-up strike rates. :smile:


<font size=-1>[ This Message was edited by: sportznut on 2004-03-02 14:04 ]</font>

michaelv
2nd March 2004, 03:31 PM
Thanks for the link Sportz and sorry about the distance suggestion but it was just a thought (and not much good either it would seem).

sportznut
2nd March 2004, 04:13 PM
That's okay, there's absolutely nothing wrong with making suggestions and offering opinions. :smile:

kenchar
2nd March 2004, 05:33 PM
Sportz,
I found 1 bet today which won well but too short for me, do you agree????

Cheers

sportznut
2nd March 2004, 05:41 PM
Are you talking about Strictly Fast??? It wasn't actually a selection because there were TWO horses in the race with 100pts. To be a selection, the horse must be the ONLY 100 rater in the field after scratchings.

By the way, that other 100 rater ran 2nd. There were also two 100 raters in Race 7 and they ran the quinella as well.


<font size=-1>[ This Message was edited by: sportznut on 2004-03-02 18:50 ]</font>

sportznut
3rd March 2004, 07:39 AM
Shaun,
I've researched another 2 months results and while the results have still been pretty good, they haven't been quite as good as the last 3 months. I'll keep looking a bit longer to see if I can find a definite pattern. At the moment I think metro racing might work best. What do you think is a reasonable period of time to test something like this, 6 months? 12 months??? By the way, in my original 3 month test, even betting on ALL sole 100 raters first up showed a profit, only around 20% but still a profit. While I haven't kept track of those figures, I think that trend has continued.

Jacfin,
While I was going through the results, I kept noticing all the good priced winners from the 100 raters 7 days a week, and I think you really could be onto something with your idea. :smile:


<font size=-1>[ This Message was edited by: sportznut on 2004-03-03 09:38 ]</font>

Shaun
3rd March 2004, 08:52 AM
Sportz i use this method when testing races because i don't have a data base....can't remeber who posted it they deserve credit for it

1. First sample = 150 recent races (or bets) If this shows miserable results, forget it. If it gets near to break even or pays POT continue. This sample will include at least one period of three consequitive days.

2. Second sample = 100 races from the same season in previous years (not recent races) but with NO consequitive days. If its near to break even or in POT, continue.

3. Third sample = 100 random races from all seasons trying only one or two races per day from here and there. If its near to break even or in POT, continue. A system MUST pass this test. (If you only sample consequitive days you can hit distorting cycles and mini-cycles. You have to sample against this phenomenon.)

4. Fourth Sample = carefully tabulate results over the last week's races, going back over some days covered in sample 1 but writing it up on a spreadsheet to examine results more closely.

Shaun
3rd March 2004, 08:52 AM
i have tested jacfin system before and the recent results are good the past results are not so good....the only thing i can say to this is that the ratings have changed in recent times and are much better than before here are some results with out going in to to much information

TEST ANY RUNNER RATED 100 ON QTAB OVER $8.00 IN BETTING

Saturday, 14 February 2004 42 RACES 9 SEL 1 WIN 2 PLACES
2.5
17.4-3.7
3.30

Saturday, 07 February 2004 39 RACES 6 SEL 2 WINS
11.60-3.60
8.80-2.70

Saturday, 31 January 2004 41 RACES 8 SEL 2 WINS 1 PLACE
16.20-4.70
12.50-3.60
2.90

Wednesday, 11 February 2004 35 RACES 3 SEL 1 PLACE
1.90

Thursday, 12 February 2004 33 RACES 3 SEL
Friday, 13 February 2004 37 RACES 3 SEL 1 WIN 1 PLACE
3.10
10.40-3.10

Thursday, 05 February 2004 40 RACES 11 SEL 2 WINS 2 PLACES
3.70
9.60-2.50
2.70
9.00-2.60

RACES 267 SELECTIONS 43 WINS 8 (18%) PLACES 15(34%)
AVE $11.93 $3.10
-------------------------------------------------------------------------------------------------
Wednesday, 29 January 2003 39 RACES 10 SEL 2 WIN 1 PLACE
8.80-2.60
11.80-3.70
3.20

Saturday, 01 February 2003 40 RACES 5 SEL 2 PLACES
1.90
2.80

Saturday, 08 February 2003 40 RACES 10 SEL 1 WIN 3 PLACES
16.00-4.20
3.10
2.60
2.80


RACES 119 SELECTIONS 25 WINS 3 (12%) PLACES 9 (36%)
AVE $12.12 $2.98

-----------------------------------------------------------------------------------------
MELBOURNE CUP CARNIVAL TEST
Saturday, 01 November 2003 RACES 43 SEL 9 PLACES1
2.50

Tuesday, 04 November 2003 RACES 42 SEL 8 WINS 2 PLACES1
9.10-3.80
17.00-5.30
3.20

Thursday, 06 November 2003 RACES 40 SEL 4 WINS 1
43.40-9.30

Saturday, 08 November 2003 RACES 42 SEL 6 PLACES1
2.30


TOTALS SEL 27 WINS 3 PLACES 6
AVE $23.16(11.1%) $4.40 (22.2%)
OUTLAY $100 UNITS $2700 RETURN $6950 PROFIT $4250 T.O.P 157%

------------------------------------------------------------------------------------------------------
RANDOM TESTS
Saturday, 02 March 2002 RACES 41 SEL 12 PLACES1
3.70

Monday, 25 March 2002 RACES 22 SEL 1

Wednesday, 26 June 2002 RACES 41 SEL 4
2.70

Sunday, 26 May 2002 RACES 58 SEL 6 WINS 1 PLACES 2
2.30
3.30
9.10-2.90

Wednesday, 10 July 2002 RACES 40 SEL 5

Saturday, 06 September 2003 RACES 41 SEL 3

Saturday, 13 September 2003 RACES 41 SEL 11 WIN 1 PLACE 4
2.60
2.60
17.30-4.20
4.90

TOTAL RESULTS
SELECTIONS 137
WNS 16-----11.67%
PLACES 39-----26.27
TOTAL$ $137
TOTAL R $228
PROFIT $91
TOP 66.4%

TOTAL $137
TOTALR $129.10
















<font size=-1>[ This Message was edited by: Shaun on 2004-03-03 10:04 ]</font>

sportznut
3rd March 2004, 09:13 AM
That system testing idea is a bit strange, but I guess it would work. Anyway, I've compiled my first-up system results back to Sep15 and this is what I've got so far:

Selections 102
Wins 32, Win return $167.10, Win profit 64%
Places 59, Place return $115.20, Place profit 13%


<font size=-1>[ This Message was edited by: sportznut on 2004-03-03 11:43 ]</font>

sportznut
3rd March 2004, 09:20 AM
Yeah, I really think there is merit in Jacfin's idea and I do agree that I think the ratings have probably improved in recent times. I'm currently working on a few things using these ratings and considering they're free, I think they work very well indeed.

ginger
3rd March 2004, 10:26 AM
sportznut
when you have checked back results did you notice if any 100 pointers were racing at their home track ?
cya

sportznut
3rd March 2004, 10:28 AM
No sorry, I didn't make any note of that.

jacfin
3rd March 2004, 11:05 AM
Shaun
Bear in mind that my rules apply only to Sat city meetings excluding Adelaide. Your results for Sat seem to include all meetings.
I've tested back to August last year and the LS profit is 86 units. Aug was a losing month with Bris contributing 7 losers straight. Interestingly, working backwards in the testing Bris has had 1 winner at $11 in the last 31 races. The recent Bris results have been the best - perhaps they have improved or maybe it's seasonal.
On the testing issue I find a good method to cover all seasons is to test the first Sat in each month for the last 12 months to get an idea of viability. Maybe the first week in each month if testing a daily system.

Shaun
3rd March 2004, 11:16 AM
Jacfin....my test on saturday are for the 5 states city meetings on midweek races it covers all Australin races.....i do like the results not to bad and saturday does out perform midweek

sportznut
3rd March 2004, 01:52 PM
Jacfin,
I've been going through the results and I reckon $6+ would do okay. I've seen quite a few winners in the $6-$8 range.

jacfin
3rd March 2004, 02:15 PM
Sportz
As I continue my research I'll take note of the $6+ as well as the $8.5+ and report back.
I have noticed winners in all ranges and this could well be the basis for a different system.
I've always liked longshot methods but the ratings do seem like fertile ground for all sorts of ideas.

sportznut
3rd March 2004, 03:16 PM
You're not wrong!!!

From today's racing, Bris 5-1 Chinbok, the only 100 rater in the race wins at $13 and Murray Bridge 4-2 Cosipura, again the only 100 rater, wins at $7.70. A few other shorter priced winners as well. :smile:

<font size=-1>[ This Message was edited by: sportznut on 2004-03-03 17:11 ]</font>

jacfin
3rd March 2004, 03:55 PM
I've just completed another month and the results were good ( but not for Bris )
16 selns - 3 winners - return $34 - LS profit = $18.
Sportz
Would you mind emailing me at jacfin@mail.nu if convenient plse.

sportznut
3rd March 2004, 04:25 PM
Sure thing! :smile:

The results of your little system seem to be holding up very well. How many months is it now that you've looked at?


<font size=-1>[ This Message was edited by: sportznut on 2004-03-03 17:28 ]</font>

jacfin
3rd March 2004, 05:04 PM
I've looked at Feb 04 back to July 03 inclusive.
I'll go back to Mar 03 so that I've got the full year.

Shaun
3rd March 2004, 08:56 PM
hey guys...i have looked at the $6+ rabge and yes they do come in but that also gives you more loseing selections $8+ is a good starting point

jacfin
7th March 2004, 10:28 AM
Results for 6-3-04
................LStart
S9.3 - 3.2 28 days
M1.3 - 1.8 21 days
M5.6 - -
M9.8 - -
M9.16 - - 22 days

Day 1 -5.0 units
Cum -5.0 units

<font size=-1>[ This Message was edited by: jacfin on 2004-03-07 16:24 ]</font>

michaelv
16th July 2004, 08:27 PM
Jacfin, Thought I would resurrect this post as I was wondering how your research on the unitab 100 raters system was panning out.
I have had the occasional look at it over the past few months and the long shots just seem to 'keep on comin'.
This month [July] has been quite good so far with some 90 bets [at $8.00 or more (SP)] producing 12 winners returning 150.53 units [using MaxiDiv] at an average dividend of $12.50! That's a profit of over 60 units [66.67% POT]! for just 2 weeks! Not too shabby at all and I was wondering if your research showed similar results for the 12 months you were looking at?
Yes, obviously there were some extended ROOs (runs of outs) with only a 13.33% strike rate but the divies are truly marvellous and if you ran a bank of say 250 units or 0.4% of the bank, then the ROOs won't hurt so much.
I have decided to keep a closer eye on these from now on and have noticed the number of bets jump dramatically if you include the selections at $6.00 or more without improving the strike rate OR the profitability.
In the same 2 week period [as above] there were a total of 182 bets [at $6.00 or more) for 23 winners (a 12.6% strike rate) returning 235.34 units for a profit of 53.34 units or 29.3% POT. Still not too bad but the $8.00 bets were more profitable with a better strike rate and you didn't have to work so hard.
Now this could just be an aberation for this month which is why I am asking you about your research.
However, putting all this in dollar terms - if you could bet $100 per selection on a $25,000 bank and used the maxidiv offering then the 2 weeks in July would have given you a profit of some $6,000 for the 2weeks so far - which is quite exciting!

Therefore I look forward to your response Jacfin.

Merriguy
17th July 2004, 09:46 AM
Interesting post, Michaelv.

In obtaining these last results are you speaking about covering every meeting every day, or just the Saturday metrop. meetings as was first suggested?

Referring to your other post today, it is always said that the ratings are actually AAP ratings. Can't see anything on their site to prove or deny that. They may be able to offer the service you are seeking --- though I guess it would cost!

jimmyc_dogs
17th July 2004, 01:09 PM
ok guys, maybe a very stupid question, but are these ratings the same as the ratings that are on the tab walls near the jockys name (IN NSW)

thanks/(or mayb sorry 4 wasting your time)

jimmy

sportznut
17th July 2004, 01:47 PM
Nope.

You find them at
http://www.tabonline.com.au/cgi-bin/main.pl?page=/today.htm

Simply click on a race and the Ratings out of 100 are on the right side of the screen.

jimmyc_dogs
19th July 2004, 08:47 PM
hey guys, can you please list all the rules to choosing the selections???

thanks
jimmy

sportznut
19th July 2004, 09:06 PM
Unitab Ratings.
Horse rated on 100.
$8.50 or more at jump time.
Delete any races with 3 or more 100 raters at $8.50+.
(because of the way prices change in the last few seconds, I think Jacfin ended up suggesting a cutoff of $8.30+ and you could even think of using $8.00+)

Jacfin was using it on Bris, Syd & Melb on Saturdays only, but I think it can work every day. Today, using Unitab figures I think there were just 4 bets for 2 winners at $14.90 and $8.50. Not bad huh, but I must warn you that not every day's like that. With this idea you could possibly run into long losing runs.


<font size=-1>[ This Message was edited by: sportznut on 2004-07-19 22:21 ]</font>

jimmyc_dogs
20th July 2004, 06:11 AM
and also must have won at distance???

jimmy

sportznut
20th July 2004, 06:44 AM
No, that was just a filter that I suggested to Jacfin and not part of the overall rules. Neither of yesterday's winners had won at the distance, so you would have missed them if you used that rule.

jimmyc_dogs
20th July 2004, 11:15 AM
thanks sports, as usual done well

jimmy

sportznut
20th July 2004, 03:51 PM
Today: 3 selections for 1 winner at $8.90 :smile:

jacfin
20th July 2004, 05:01 PM
Hi Michaelv and Sportz
This is a test to see if I remember my password.

jacfin
20th July 2004, 05:17 PM
Michaelv
I have a spreadsheet which covers all 100 pointers >= $8.30 in Syd and Mel on Saturdays for a 12 month period with columns showing whether they'd won at the distance, how many starters in the race and a few other columns which I can't recall at the moment.
If anybody wants a copy just email me at jacfin@mail.nu.
I'm impressed with your figures for every day Michael. With the number of selections it may pay to raise the minimum price to $10.
Over many years of system following I have found that a hit rate of as low as 10% can be workable and the longest ROO for even that rate would not be expected to exceed 60. A bank of 100 units would probably suffice if the hit rate is maintained.

sportznut
21st July 2004, 05:39 PM
A frustrating day if you followed this plan to the letter today. A couple of winners at less than $8 and also a $10 winner in a race with 3 qualifiers.

jacfin
22nd July 2004, 09:53 AM
In my research on Sats I found that applying the following filters resulted in a slight improvement in profit with the elimination of a lot of bets.
Field size must be 10 or more
No 2yo races or jumps races
No second uppers ( not one won in the year )
Must have a c or d next to it in the form guide
No races of 2400m or more.

racingnovice
19th August 2004, 09:28 PM
Hi,

Would anyone be able to send me the spreadsheet please to racingnovie@yahoo.com.au . I have emailed jacfin but has not replied in the last couple days. I have been doing some more tests with this system and a few filters. The results are below sofar. This is only been tested over 6weeks from 17th july to now and a few weeks in janurary. The spreadsheet would save me loads of time :smile:.

SELS 39
WINS 9
RETURN $92.90
POT 158%

In the mean time i'll be doing more testing :smile:. I'm goin on $8+.

Thanks in advance