Neural's Credibility

dturner
Hi Robert,

I have reviwed my Neural ratings for tomorrow's extremely difficult meeting at Caulfield and I question their credibility after reviewing the outcome of Caulfield Cup.

The neurals give Grand Armee a higher distance rating than Zagalia (it does not matter what level I set distance at !!)

Grand Armee has never even attempted the distance and Zagalia has won a G1 over 2400 ???

Please Explain...
You need to Login to Post a Answer in the Forum

Robert, My apologies for starting this "war". This was my first message on the Website and I am pretty certain it will be my last. I use the neurals as a tool to evaluate races and have never thought they they are infallible or would produce the result in every race. I was using them on the Caulfield program and just wondered how the Distance factor was calculated - all the variables are subjective, some more so than others which is why I asked the question. It was not my intent to start an argument as to whether Zagalia is a better horse than Grand Armee - my whole aim was to try to Win on the Caulfield Cup. FYI The top rater on my Neural was Mummify. rgds

incorrect robert. u can't put in bad luck, jockey decisions and horse's mental state on the day into a formula.

those factors have the final say on the outcome on a race, no matter what the form says!!!!!!!!



Mate, not withstanding the result, Grand Armee is a much better horse than Zagalia. If you check the forum search, this has been discussed before. A estimate is made on the horses ability to run the distance.

Danoz,

What a load of rubbish...

As demonstrated by the race Grand Armee is not even close to Zagalia at the distance of 2400m and this I believe was the point..

Grand Armee is out of a Marauding Mare ??? Does this not indicate to you a sprinting pedigree ie a former slipper winner who has produced numerous stakes winner's virtually all over sprint distances.

Zagalia is by Zabeel out of a Bluebird Mare enough said !!!!!!!!

This was not a GA vs Zagalia debate but a real query on the neural system.

dturner,

Have you received an answer from the powers to be ????

Spinning Knight,

A couple of points.

Before the race nobody KNEW for SURE how Grand Armee would perform over 2400m. Sure, everyone can have their own opinions before the race, but before having run over the distance how do you know?

The Neurals have to make some form of deduction on how a horse will go when racing over that particular distance. The further the computer has to go to get a distance rating the more it get penalised.

I've never said that the Neurals are infallible and sure there is always an element of "guessing" when it comes to horses untried over any aspect of the race. But then again how many races are there where ALL factors are know?

As a general statement Grand Armee is clearly a better horse than Zagalia - surely you don't disagree with that. So, the computer made a judgement that even with a penalty GM rated better than Z.

Surely, you look at the form of the horse as the major factor rather than breeding :). That's what the Neurals do. Given the number of decisions the Neurals make, sure it gets some "wrong", but every time you don't pick a winner does that mean that you've made the wrong assumption?

Having been involved in racing for a long time - I think that when doing form it is a dangerous practice to jump to conclusions too quickly. I always make the assumption that the horse will do the best it possibly can.

In this case, having seen the run I agree with you and believe that GM will struggle to get over any staying distances, but then again the time it ran for 2400m was still a hell of lot faster than most 2400m race winners.

Also, FWIW my view on the breeding aspect - After the fact breeding is great.

regards
Robert Vilkaitis
IT Manager

Robert,

A couple of replies.

"Before the race etc.."

Agreed no one knew for sure, but if a horse is having its first start over a distance then breeding surely is a relevant starting point.


"I've never said that the Neurals are infallible etc..."

Thank you for advising how the neurals are done.If ALL factors were known punting would be too easy.. I have/do use the neurals and assumed that their was a more scientific method used rather than an individual making assumption's..I will continue to use the neurals but allow some personal loading on certain factor's,they are a useful tool that save me time..Normally I would use Best Bets and use a similar rating category that the neurals calculates for me (I am by definition only a Mug Punter)..

BTW I didn't start the thread making a personal attack on your creation.


"As a general statement Grand Armee is clearly a better horse than Zagalia - surely you don't disagree with that. So, the computer made a judgement that even with a penalty GM rated better than Z."

Given the factors of the Caulfield Cup I disagree Z is far superior to GM over this distance. Was the Waterhouse factor an influence in the decision to rate GM so favourably, as discussed before I believe their is a Waterhouse bias associated within this site.

"Surely, you look at the form of the horse as the major factor rather than breeding :)."

Not over this distance and indeed the Melbourne Cup i believe they both play an equal part..


"That's what the Neurals do. Given the number of decisions the Neurals make, sure it gets some "wrong", but every time you don't pick a winner does that mean that you've made the wrong assumption?"

Yes, I assumed my horse would win!!!!

"Having been involved in racing for a long time - I think that when doing form it is a dangerous practice to jump to conclusions too quickly. I always make the assumption that the horse will do the best it possibly can."

Agreed keep up the good work.

"In this case, having seen the run I agree with you and believe that GM will struggle to get over any staying distances, but then again the time it ran for 2400m was still a hell of lot faster than most 2400m race winners."

Agreed as well, in a fairly early post (pre barrier draw) on 14th October I recommended Grand Armee as part of my flexi tri based on it's class and that I believed GAI had set it for this race all along. I also nominated Zagalia ( but without posting this was mainly based on her breeding).

GM yes ran a stronger 2400m than some/most but certainly was never going to run as strong as Z after the barrier draw. GM would have needed a very cushy run ala Mummify to excel.


"Also, FWIW my view on the breeding aspect - After the fact breeding is great."

My view was available before the race!!! as previously stated I did not start this thread questioning your credibility, I merely responded to a post and now I know your strategy in formulating neurals will make any necessary individual adjustment myself.

GM is not a much better horse than Z..as a general statement.

Being a Zabeel (again breeding research will show) she takes more time to mature and gets better with distance and age..Time will tell..

robert,

u might want to alter the distance formula for the neurals. horses that jump 800m like grand armee did usually fail. his rating should have been discounted because of this. that would be easy to adjust for because it's numerical.

as for "GA being a better horse then zagalia" -well that depends on distance surely!!!!

hareeba is a far greater sprinter than doriemus but doriemus would win easily over 2 miles. which is the better horse????? depends on distance doesn't it!

Spinning Top ,

Why didn't you just apologise to Robert for another one of your fly off the handle comments to serious users off this site rather than writing a book to make your point.

Spinner you can take your foot out of your mouth anytime your ready.

The Lukestar has spoken.

Lukewarm,

I need no apologies to be made..My point is indeed totally relevant and if Robert wishes to deliver a sermon to the non originating poster, I will invoke my right of reply..

You still haven't answered my question:

"Has Pentastic finished in the Caulfield Cup yet??"

And you talk about serious poster's and you post rubbish like that prior to the CC.

The Lukestar has been told.

How do u put names like zabeel into a mathematical formula? that's where the problem is..... unless a table is created to give values to certain sires and dams for certain distances u would have to use your own knowledge on that....

it is impossible to put together a mathematical formala to allow for all the factors involved in a horse race, which is why i dont use neurals.




Speedy,

You have left your self open to this...

"it is impossible to put together a mathematical formala to allow for all the factors involved in a horse race, which is why i dont use neurals."

You put ALL those factors in and still pick losers!!!

Sorry - couldn't resist,
Robert Vilkaitis

hey dturner im not sure bout ur question thers a fault somewhere but i was wondering what settings u put each one on like the cp and all that and do u add any points too any horses when there really low but have a good cp rating

Mizare,

Check ur inbox

lol

Talkback Forum

Racing And Sports offers our users the chance to get involved with spirited discussion about major racing issues on your Talkback Forum. They may include chatter about racing matters, issues relating to the website, sporting conversation, betting matters and more. Please be aware that Racing And Sports now moderates Talkback to ensure posting guidelines are adhered to. The views expressed on Talkback are those of the writer and not necessarily those of Racing And Sports. You must be a registered user to write postings or send messages to other users.

Latest Posts