Well done to the winners today and the overall leaders in the comp.

Perhaps some people might like to do a post mortem of why they did so well or poorly.

I'll take the lead here and I did poorly....

In the first leg Race 4 Niconero's best run was last start over 1200m at Caulfield in an Open class race. The horse was jumping 200m in distance and heaps of class, I didn't think it could go on to improve that much. Econsul had never done much first up so I left it unrated pretty much. I liked Tirade a lot going on the run in the Grp 3 Guineas Prelude over the same distance and track, but that was last preparation, and should he have lived up to that form should have brained 'em. Truly Wicked had won a Group 2 at MV and Classiconi had won a Listed Race at Flemington last prep. I think I need to look harder at recent form.

In the second leg, although it was spruiked that they had Elvstroem set for this race, trainers often spruik their own horses often to prove fruitless at the end of the day. I did miss looking at the beaten margins first up and looked at the placings. First up Elvis had run 0.4 lengths 3rd in the Bletchingly and 1.4 lengths 5th in the Zeditave. So the fifth was really equivalent to a second or third.

Some lessons to be learned.

__________________
RaceCensus - powerful system testing software.
Now with over 409,000 Metropolitan, Provincial and Country races!
http://www.propun.com.au/horse_raci...ng_systems.html
*RaceCensus now updated to 31/10/2024
Video overview of RaceCensus here:
http://www.youtube.com/watch?v=W821YP_b0Pg

On the point of recent form I agree with you 100% chrome. Since the start of the challenge i have missed several winners that had high recent ratings that I ignored as flukes, aand then went back further to get a more realistic rating which cost me a couple of winners. I guess the if we learn by our mistakes then the challenge has proved useful.

Well Chrome, I'm surprised (and a bit embarassed) that I'm leading this thing. You see, I've really only been using ratings on a consistent basis for over a month. What I do is so damn simple, but it seems to have been working pretty well so far. (Using my ratings, my best bet today was Miss Gloryhunter in the last in Brisbane)

I must admit that I use the ratings from both the Tas Tab website and the Unitab website to help me get my weight ratings. Considering the info on those two sites is totally free, you just can't beat them. Then I add a few bonuses or penalties regarding days since last start, jockeys, barriers and state of the track. For me, it's still a learning process though. There are a few things I have to look at to try to improve my ratings even further.

Concerning today's races, I had Savabeel down far too low and I've gone back and looked at the ratings and saw where I went wrong there. I definitely think I rated the other 3 year olds too high by comparison. On pure class, he should have been ahead of them, even first-up. As for the other race, quite happy to get the quinella in my 5 picks. Could possibly have had Wildly higher up given the state of the track.

As for recent form, I always use the horse's last run in my ratings (unless it's first-up or the rating for that run was WAY below average). I don't use just one race to find a base rating though. I use two and average them. Normally it's the horse's last start and the best rating from the previous two starts. So I don't usually go back further than a horse's last 3 runs, unless it is 1st-up, 2nd-up or racing over a totally different distance or track condition.

Interesting post Sportz, thanks for that. Using LS and best of previous two at first glance seems flawed, but what it will do is throw up the types we should really be backing - the consistent ones.

We'll chat more about this soon.

Duritz.

Well considering i have had a faulty spread sheet things are not to bad....my ratings are not really designed to handle 1st and 2nd ups i prefer a feww runs but anyways i re-rated yesterdays races to see how it turned out with the spread sheet fixed....i hope.

MR4
7 Niconero (1) $5.32
14 Talons Shown (12) $6.04
5 Wildly (2) $7.19
8 Truly Wicked (13) $9.54
3 Classiconi (9) $9.95

MR5
7 Casual Pass (3) $7.28
8 Lad of the Manor (NZ) (9) $7.42
15 Danehill Express (2) $7.64
10 Pretorius (NZ) (11) $8.32
14 Not a Single Doubt (14) $8.32

Hi all

Sportz how do you find the ratings on the two tab sites. How do you interpret them. Do you look that they agree on both sites etc.

I dont look at form, past runs,days,weight, jockey etc. For example yesterday Makybe diva was high up in the ratings however one would know that at 1400m/caufield/first up that it would be less likely of a positive result.

When spreading over 5 selections light analysis can be forgiving. If i was to narrow it down to say top 2 then i would study the ratings a bit more indepth or use an external source for comparison. I would think.

I haven't used or viewed TAS TAB i have unitab.

cheers
ubetido

__________________
Life should NOT be a journey to the grave with the intention of arriving safely in an attractive and well preserved body,but rather to skid in sideways, BEER in one hand- PIZZA in the other, body thoroughly used up, totally worn out and screaming WOO HOO!

Here's the Tas Tab form guide:
http://www.tastab.com.au/racebet/fo...num/default.asp
And here's the Unitab website:
http://www.tabonline.com.au/

On the Tas Tab form guide, click on a horse's name and it'll give you a complete form rundown for that horse including a rating for each run. The rating will be in the form of a number like 645 or 725 or something. To convert them to weight ratings, I simply slip a decimal point in there and make it 64.5 or 72.5. Technically speaking, doing that is probably not 100% accurate, but it's worked out okay for me. So, I get the two ratings I want and then average them out. In that instance, it would be 68.5.

Next, I add the horse's Unitab rating. These ratings are a score out of 100 on the right side of the screen beside each runner. Before using it in my ratings, I divide it by 2 though so it's a score out of 50 instead of out of 100. So, if the horse has a Unitab rtg of 93, I turn that into 46.5 and add that to the rating I have from the Tas Tab site. Then I add small bonuses and penalties depending on the days since last start, the jockey, the barrier and the state of the track. Once I've done that, I'll have a total from which I subtract the horse's handicap weight and what's left is the horse's rating for today's race.

I think your treatment of the tassie tote ratings is pretty close to the mark sportz, on a quick comparison with the ozeform ratings, the average divisor between the two figures appears to be about 10.5 although it ranges from there up to 12 but I would not trust those ones as ozeform have made some errors in the past as pointed out on this forum previously.

I must say it's very generous of you Sportz to outline your method in such detail, especially when you're leading this competition.
I've actually looked at the Unitab ratings and think it's really a great starting point, but does need something extra.
I'm not looking at this from a ratings point of view, rather what I need is an accurate prediction of the market price to some degree at least.

I've looked at wa tab prepost and it is a very rough guide only.

Thanks again.

__________________
RaceCensus - powerful system testing software.
Now with over 409,000 Metropolitan, Provincial and Country races!
http://www.propun.com.au/horse_raci...ng_systems.html
*RaceCensus now updated to 31/10/2024
Video overview of RaceCensus here:
http://www.youtube.com/watch?v=W821YP_b0Pg

Hi

Thanks for that explanation Sportz. Much appreciated.

ubetido

__________________
Life should NOT be a journey to the grave with the intention of arriving safely in an attractive and well preserved body,but rather to skid in sideways, BEER in one hand- PIZZA in the other, body thoroughly used up, totally worn out and screaming WOO HOO!