Race Rating Help
 

Hi all,
I'm looking at rating current and past races (not horses), and give them a value of between 1 and 100 (or more).
My predicament is that is a Maiden of first starters at say a country track, be given the same rating as that of a Metro track? One would think that trainers would enter horses of better quality at a Metro track, than that of some country track.
The same would go for a BM64 at Geelong would not be the same strength as that of one at a Metro track and not all Group 1 races should be allocated the same number, for example a Golden Slipper (in my opinion), should not be rated the same as a Cox Plate, and of course I personally think a Cox Plate is higher than a Melbourne Cup. With a few exceptions of course Cox Plate runners go on to win other Group 1 races, whereas Melbourne Cup runners usually come back for the Melbourne Cup again and have a pretty poor record in other Group 1 races, again with some exceptions.
Any input greatly received :D

No input or even ideas?

Not sure this will help you but the way i use to rate races was like this, each class from my perceived lowest to highest was give a number like this.
G1 3
GROUP1 3
GP1 3
G2 5
GROUP2 5
GP2 5
GP2 5
G3 7
GROUP3 7
GP3 7
Listed 9
LR 9
Quality 10
Open 12
HCP 12
HANDICAP 12
100 12
96 13
95 13
94 13
93 13
92 13
91 13
90 13
Class 6 13
Cl6 13
C6 13
3MW 15
2MW 15
WELTER 16
Flying 16
90 18
89 18
88 18
87 18
86 18
85 18
1MW 18
0MW 18
One Metro Win 18
No Metro Wins 18
Set Weights plus Penalties 20
Class 5 21
Cl5 21
C5 21
84 21
83 21
82 21
81 21
80 21
79 21
78 24
77 24
76 24
75 24
74 24
73 24
72 24
71 24
Class 4 25
Cl4 25
C4 25
Class 3 27
Cl3 27
C3 27
70 30
69 30
68 30
67 30
66 30
65 30
64 32
63 32
62 32
61 32
60 32
59 32
Class 2 34
Cl2 34
C2 34
Class 1 36
Cl1 36
C1 36
58 39
57 39
56 39
55 39
54 39
53 39
52 39
51 39
50 39
MAIDEN 42
MDN 42

I also included these special penalties.

3Y 5
3YO 5
3F 5
2Y 8
2YO 8
2F 8
FM 3
MARES 3
F&M 3
C&G 3
CG&E 3
WFA -2

I also used this race value component.
IF(AU3="","",IF(AU3<10000,14,IF(AU3<25000,12,IF(AU3<40000,10,IF(AU3<70000,6,0)))))

So if the last race was 50k Open it would get a score of 18
If todays race was a 75k Open the score would be 12
I would calculate 18-12 to get 6, the lower the better

In general I would rate Metro slightly higher

Agree

Agree

All 2yo/3yo G1s would be inferior imo

Hi Chrome,
I found this chart for Victorian class/weight ratings which may help .

Rather than type out the whole thing here is a link.

http://aussieraces.com/victorian_ra...lasses_134.html

Hope it's not against forum rules. If you can't access it because of the rules I'll type it all out.

I think NSW racing has the equivalent class/weight ratings as Victoria.

For Qld,SA and WA subtract 3.0 kg from the Vic and NSW figures.

Tasmanian ratings are generally between 1.5 kg and 3.0 less than QLD,SA and WA.

NT ratings are generally 1.5 to 3.0 kg less than Tasmania

The Victorian chart is from one source and the adjustment by subtracting for different States is from another x-pert.

Also according to the x-pert races 18oom and over don't contain the same strength as shorter races and 2.0kg should be deducted from the relevant State class/weight rating. I'm presuming that he is not including G1,G2,G3 and Cox Plate Melbourne Cup etc

Hi Shaun,

I am very new to this forum, I am a full time Horse Racing Analysis. I'm retired and can spend more time on my projects and Systems.

I was given your Consensus speadsheet from a friend. I would love to discuss the spreadsheet and it's workings a little more in detail, I think it is a great spreadsheet and you shure have a gift.

I was a member of cyberhorse forum many years ago, it was manly for Bet Selector Users, as usual, a lot of bad and insulting post spoilt the forum and it stopped.

I did a Class Rating some years ago, members may be interested in.

It was based on Benchmark ratings, for example, Benchmark 58 would be Rated a 58, it was to gauge the actual official Handicap of the horse.

If you looked at a horse's official Hcp and deducted the Benchmark or Class as listed below, you could get an indication of the Race strength.



You will see in this table, Class 1 =BM58 to Group I 125 as well.

Just another way of looking at Class Ratings.

My original ratings were started from the cyberhorse VFG, buck then i would copy and paste the data to excel before i started to use vba to import data.

I now use python to import the same data from the TAB.
The ratings i listed may look a little strange as they are used to look up this data from imported form, i have always had issues with comparing country, provincial and metro form, incorporating the race prizemoney helps.

I agree Shaun, prize$ is a good indicator to Race Class. My point is this, a horse 14 days ago, ran in a BM64 Ballarat, $35,000.Today the horse is at Hamilton, BM64 $35,000. Has the horse gone up in class or not?

The Ballarat race had an average HCP of 63.9 (SUM all horses HCP/Field size).

The Hamilton race had an average HCP of 61.1.

Looking at those figures, the horse is going down in Class but still in a BM64 race.

I would think a Country Bm64 would have a Average of around 59 in most cases.

Thanks for all the input guys, it gives me some indication of where to start.


My way around this is to look at the average career prizemoney of all runners (add them up and divide by the number of runners to give you a figure) and compare them, because many times race prizemoney will steer you wrong, in other words,

Python
 

Hi Shaun,

I read you are using python, can you recommend a free web site that is good to learn from?

I have done small bits and pieces of python but I'm not advanced by any means, any help would be much appreciated.

Unstable

This channel has some good stuff in small amounts so easy to follow.
https://www.youtube.com/c/Codemycom

Pyton
 

Thanks Shaun,

Also, I have your wonderful Impact Ratings SS, I have added my New Ratings to macro, however, the Prices sheet, give a error #REF! New Ratings will not load to the RacePage.



I know it's been a while since you created the SS, are you still using it?

It would rely on the old Qtab site, that no longer works, with python you can access the JSON data from the TAB

Pyton
 

Okay thanks Shaun

Unstable, a question, maybe a dumb one.

Do you mean that you add up all the different weights of the horses in a handicap race and then divide by the number of horses?

Also is that before or after scratchings?

Thanks

Hi Peter, I could be wrong here, but I think he means the handicap rating, not the weights carried.

free turorial python
 

try
tutorialspoint,com.python/index.html
that gives examples of the language elements.
i have not tried:
learnpython.org,
freecodecamp.org
i might have used:
pythonguides.com
The best books are from O'Reilly
the cookbooks are good as is Python in a nutshell
Zed Shaw Learn Python The Hard Way third ed gives you a good intro with exercises
HTH

Hcp
 

Hi Peter,

CP is correct, Official Handicap, you can find them on this web site, https://www.racingaustralia.horse/home.aspx

Python
 

Thanks Chook, I will check them out.

Chrome Prince,
Yes, that makes sense.

Unstable,
Thanks for the reply and explanation. I like your idea to determine the strength/quality of the race

HCP
 

Here is a Good example of Race Strength, Sat 9/4/2022, Werribee R4

I use a gap of 4. If the average hcp in a BM58 is 54, then I back all horses that are 4+ above the Average up to 4 horses.



Tab number 1,2 and 6 are 4 above average, I would Dutch those 3 horses.