Four for the Pool Room
 

TWOBETS- “When backfitting systems it is impossible (almost then) to isolate aparticular input factor. You may think you are adding only "first up laststart" runners for example. But in fact within that group you have amillion other factors that may or may not be having an input on the result.Thus when you add or subtract any single input it actually affects the otherinput factors.....There, clear as a bell".

Wesmip1 - "The fewer but more complex filters, which are not easilyidentifiable by most punters, in a system, the more chance of it repeating.

AngryPixie– “There's dozens and dozens, even hundreds of potential variables you could measure, most of which aren't independant so in using them you're in fact doubling up on your measurements.”

Barny– “You’re paying a fairly hefty premium in terms of SP for the most popular filters /combination of filters – look where no-one else is looking”>

I included myself because ..... I can (it's my post !!!)

Where are they now?
 

When I first joined this forum their were quite a few long timers, a few years on most are still here but a few seem to have disappeared. I wonder why and where they might be now.

eg. Crash
Baghwan
Wesmip

I know their are more because I enjoyed the banter.

ps. Sorry, I meant to start a new thread, not hijack this one. Only 5 mts to play with.

Evenin' Barny,
Haven't had a rant for a while cause this forum makes me want to pull my hair out. Not that I've got any left.

This from Wesmip is the gist of my success and I wanted to add something that I perceive as fundamental but many on here obviously don't agree with.
All this talk about very specific yet obvious facts ( like placed last start, ridden by top jockey, or whatever.... these have all been factored in to the price and have thus been rendered useless to the average punter. The price will always be against you.

To my mind you have to look for the magic Wesmip filters for a nice edge in this game. I've only got one which is why I come across like a tight rrrs, and have nothing to share, but that's where I'm at.

Five for the Pool Room, methinks.

But Barmy, tsk tsk,

Barny Read up on William Benter.

Identified around 140 variables that could affect a race result. 24% POT at around a net value of $50 Million.

when wesmip talked about filters he would often identify 2 sets of ratings, the use of a wide barrier and Days last start. The ratings were very unique but in the background many other variables gave him this total.

If someone were to say use the top rater from Don Scott racing and sports and the unitab top rater and sky ratings you probably are doubling your points or allocation on one particular variable.

I suspect you need some serious life experiences in the examination of the punting caper to appreciate how pertinent the insights of TWOBETS, wesmip1 and AngryPixie are .....

My post was indicating that its not the specific amount of variables that is the key but more around which are not independent of each other.

Its not a matter of not appreciating there comments it comes down to doing your own testing and understanding how the results / ratings are produced.

I'm sure I'll be making adjustments for the next 10 years or so but at least I'll know whats in the background.

Very true Vortech,

Good rating sites like inracing ,r+s are always refining their handicapping procedures as new inovations are discovered though they also recognize the value of the basics.
Cheers
darky

So Vortech, if I've got your message clearly, your saying we need to identify variables that are clearly DEPENDANT, a bit like Rules, and Sub-Sets of Rules etc, all intemimgled with each other ?? That flies in the face of "going against the crowd".

It depends on the context in one uses the variables. An independent variables maybe a predictor variables or a controlled variable maybe even an maniupluated variable or exposure valiable.

I'm starting my base around bet selector base ratings and then incorporating weight ratings, class ratings, and speed ratings.

If this is "going against the crowd" then yes I am.

I don't understand the point you are trying to raise here?
I have clearly identified and stated I'm testing my ratings system here so I'm not saying it will work or not work. I am telling everyone my theory behind the rules and open for suggestions. Instead you tell me I'm going against the crowd.

Sorry maybe I should pay more attention to backing horses on a Sunday called "Frisky No More" or even more stating Unleashing the Vortech Beast!!

For the life of me I cannot understand what it is you're trying to communicate. I'm desperately interested because it's an area I'm close to getting to my ideal situation, and have a keen interest in !!!

Would you give examples of what you're referring to please ?

Predictor Variable, Controlled Variable, Manipulated Variable & Exposure Variable

Barny, I believe they are different names for what are essentially independant variables. Depends on the science or discipline (Vortech's context I think).

In a nutshell, a dependant variable is derived from one or more independant variables. In a racing context the independant variables barrier draw, days since last start, track condition etc. combine to make the dependant variable, finishing position.

Thanks, almost getting too scientific for me.

It never ends does it ?

You guys are out of my league, I might have to put Frisky in my Black Book. This forum is really heavy going.

But full marks to Vortech, Maythehorsebewithyou.

Star

I agree with this 100% and this is the reason i say if you have any more than 12 filters in your ratings you are wasting your time unless these filters are totally different from each other.

The main problem is deciding what filters contradict each other, most people use the main filters but as mentioned the unitab ratings takes in to account some of these filters as does price because this has been assessed using similar filters.

The real difference is the way we asses say the last run how important do you regard the beaten margin and is it more important than the beaten margin of the run before that.

The same can be said about the other filters you use, some put a lot of importance in prize money won in the last 12 months where others will use a lifetime avg or maybe a race avg.

this is what makes your ratings different from others because with the amount and ease of access to information these days it is hard to find a new filter that has not already been used.

I like to use PPP and unitab ratings as a guide, it tells me that the rating i have given a runner is reasonably accurate as the best guide to a horses chances is it's price.

Star, if you can understand what Vortech was saying them you're much better than me old chum !!!

The best place to start reading is from the first page of this forum.
Takes nearly a year to read it all but you'll learn everything you need to know.

Ratings as I understood came about from Don Scott who one day was on his way to church and got on the wrong line. Ended up at the track.
From there he built a passion for the races and developed a method to measure horses that were winning in lower grades vs horses in other grades.

A little like Greater Western Sydney in the AFL playing the coming second last playing a top side in the VFL.

Further from that others have developed more ratings systems based on different variables.

Class ratings use variables to determine the best class horse in a certain race
Weight ratings like the ZipForm from the sportsman develop the horses with the measure of how the horse's weight compares to its current official rating.

And so on and so on.

Other ratings like newspapers and unitab I assume take all of these figures into consideration and develop one rating.

Barny by using your database you should run the entire database if your PC can handle it and start filtering your results into categories.
Do a search on William Benter and how he developed his variables.
Sometimes its easier to focus on one track.
I'm considering basing new rules and variable points around different venues/distances.

Good luck

To Star, it's EXACTLY what I've suggested in MANY of MY posts, I've actually done it ..... an enjoyable pursuit and you'll learn heaps. This site is brilliant for information, but you have to read, sort, catalogue, and keep many posts.

It's simply brilliant.

If you're not up to the task, then you're not up to exploring the vagaries of the punt, and IMHO you're destined to be a losing punter. Like anything in life .....

"The harder you work the luuckier you get" ..... Gary Player

No-one can explain it in one post .....

The greatest factor that is not in any rating is " how is the horse today?" Sunday week ago I was told Bombora was So-So, Wednesday I was told the horse was getting there, Saturday morning I was told that the trainer had cancelled plans for Sunday. The trainer and the owners knew early Saturday that nothing would beat him. This was well before he was even ready to go to the track.

Point being you can't win them all but have to trust in what you do.

Morphettville Race 8 MANGANESE, before the race i knew nothing about this horse only what i have read, my top 2 selections for the race.

4 Jeunealistic 3.48
5 Manganese 3.48
Results



1st 5 MANGANESE $29.00 $7.50
2nd 8 KERT $2.80
3rd 7 HANDSOME AS $2.00


I can explain why my ratings had this horse rated so high but can't explain why it was so high priced.

3rd best average price.
Not in top 5 for win & Place percentage
Best at the track and distance.
Best in the wet.
2nd for fitness.
2nd for class rating.
Not in top 5 for form
Not in top 5 for Qtab.
Not in top 5 PPP

"Not in top 5 for form" - That would pretty much do it!!

Sometimes I feel it disrespectful to the person who I am asking a question of. Sometimes what is said might appear to be against what the general theme of what quite a few long termers here who know the ropes and have more then likely seen it all.

With repect to the Shaun and the horse Manganese. Shaun had it rated second, yet the market did not agree.

Here is my Conundrum No 64. Somebody else said if they price a horse or their ratings do and the market has it well over those odds that he believes the Market is nearly always right. So he accepts that and believes he is at fault not the market. ( or something similar, meaning the same if I have it right )

Now, this interests me greatly because I suppose one common theme through all my meanderings is looking for value, looking for the angle that the crowd and market undervalue.

So, in respect to Shaun. What should he have done in this situation. I have conflicting opinions given to me. One, disregard and let the market guide as a filter or 2. Go with what you have got. If your ratings or system selects something, and you think you have a sound policy then go with it, you can not deviate and not bet, because of factors outside of what you have judged the race on.

?

Star

Shawn are you still using the following variables in your ratings

API
Win%
Place%
Distance/Track
Track Condition
Fitness
Class
Form
Qtab Rating

I think after hours of testing you guys are right all along. :(
I was wondering if you factor in Trainer / Jockey, WEight and Speed?

Regards

I believe the important criteria was that the above conditions only apply to a certain subset of races - Those being , the few races that the member finds playable.
And in the members case , I think he is trying to remove the unknown factor from his races as much as is possible.
The unknown in these seeming Overlay situations are that there must be something wrong with the horse in todays event because it is simply much too generous Odds and BookMakers , as we all know , are not a charity.

My 2 roubles on this...

Rule no 1
Form an opinion of your own because,
Without one, we are nothing,
As punters (or men?)

Rule no 2
Seek all the advice/info/filters you can get
But at the end of the day.
Be prepared to back your own judgement!

LG

Hi Mo.

Agree 100% with your post.

Cheers.

True, Bookies are not charities. But I think that statement becomes blurred when they are now Tote watches, The Corporates certainly are and if they follow the Tabs who definately are market driven less their take then they fall in line.

Top Fluctuations and SP are so close in all races that tells me the market defines and it is the weight of money that decides.

Now, this brings up another interesting question. Where does the bulk of the money come from. Is it a few professional syndicates, or is it the sheer number of small punters or is it from something more sinister like ( Laundered Money ) I do no know , had to throw that in.

But, if we had some idea of the money trail and its connections then that might be something to factor in or to eliminate.

Maybe it is posssible to try to pick which races are of interest to this ( Money ) and either follow or avoid.

?

Star

Shaun would be able to answer this based on his methodology.
1- Is it profitable to bet horses that are way over his rated odds.
2- Is it not profitable to bet horses that are way over his rated odds.

With mine occasionally circumstances happen that a horse marked,
At say $5.0 will win at $21.0 etc.

"LONG TERM FOR ME BACKING HORSES OUTSIDE THE MARKET,
iS UNPROFITABLE SO AFTER THE CURSING AND SWEARING,
i GET BACK TO THE JOB AT HAND BASED ON 1,000S AND 1,000S,
OF PREVIOUS RESULTS".

Shauns ratings may find the bigger the better.
Everybody uses different ideas to how they arrive at and bet,
Their particular selections.

You can't put it into A OR B it's what method is "PROFITABLE",
For the selections you are betting.

Cheers.

Are they?
I didn't realize that.
In fact , i assumed that they were amongst the major players of the tote pools.
But I have not been to a Racetrack for over 10 years now , so things sure have changed.

Unless I am missing something, the exact thing one should be looking for with their ratings is the situation where, for example, our rated $5.0 chance is actually paying double, triple or more.

I really don't get spending the time & effort producing a set of ratings which all but mirror what the market believes is the true chances for each runner - I mean why bother, this has already been done for you, & IMHO will not be successful longterm.

norisk, thats essentially how i'm using my ratings now.

I use my ratings in two ways:

One way is to bet every race that completes ratings on, using 3 of my top 5 ratings and tilting towards value - this is a more stable and consistent method.

Second is to bet only selective races and where there is significant value in my top rated horse - this is a more volatile but highly profitable (so far) method.

I definitely wouldn't even bother with putting the time and effort into my ratings if it wasn't giving me some kind of value-edge compared to what the market is already perceiving.

Norisk, while I agree the situation you describe above is the ideal, I doubt many of us - me included - price their rated selections with any great degree of accuracy at all. The days of the massive overs left us in the 90's and these days with cheap high speed computing available to all, once hard to come by information now much easier to come by, and the influx of international money into the local markets, the market is extremely accurate.

Variance will of course throw us outliers like Manganese but as garyf suggests we'd go broke if we just concentrated on these.

Agree somewhat AP however strike rates for price brackets remain fairly static, & I when I see suggestions to ignore runners because they appear overvalued, I shake my head, defeats the whole purpose.

My ratings are far from perfect but they throw up good priced winners on a regular basis for long enough now to know it's not a fluke or statistical anomaly or the like & thanks to BF good overs are still available.

The additional factor i use are PPP i found this to help when the ratings had a couple runners with close scores.

In the end the best option for me is to stick with my percentages, no matter what the prices i still have better results for the top 3 selections than i would for say 4th to 6th selections.

Yes i get it wrong at times and the top selection at higher prices lose and lower selections at better prices do win, but in the long run my top 3 selection average over 50% winners with the top selection at around the 20% to 30% winners.

Since this is on par with most ratings services i am happy to stick with that, but as always the better class and form races produce the best results long term because i feel that most horses are trying to win these races.

If the best class race on the day is an Open then unless the trainer is just running to get his horse fit he would be trying to win.

This is where better prices can come from because people think that a particular horse is not able to win when that horse is trained to win that race.

UH OH???.

Based on this the last 5 years for me betting full time has only been,
A fluke so obviously i need to log on to the "seek job site" and get,
Me a 9-5 job quick smart before the money runs out.

Cheers.

Not sure what that little hissy fits about - just giving my opinion...

Exactly what i am doing.

Yes you have posted a lot of times GaryF.
And no I haven't read all your posts as deeply and as thoroughly as I perhaps should have.
But I was under the impression that you do not rate horses yourself , but in fact purchase ready made Ratings by Others.
If wrong , my bad , if not , then what you are doing and what many others are doing is apples and oranges.