Trifecta and F4 betting - Racing Talk - Racehorse TALK harm-plan

Racehorse TALK



Trifecta and F4 betting - Racing Talk - Racehorse TALK

Author Topic: Trifecta and F4 betting  (Read 1299 times)

0 Members and 1 Guest are viewing this topic.

Offline bascoe

  • Open
  • User 2568
  • Posts: 146
O.P. « 2020-Jun-03, 08:53 AM »
In an effort to try to get this forum back to some quality contributions, I would like to revive a thread I posted a few years back on Trifecta betting.

TAB provide estimates (should pays) on Quinella and Exacta combinations, and I retrieve these in my programs.

For a variety of reasons calculating a should pay estimate for trifectas and F4 is imprecise.  As I price each runner I was thinking a simpler way to approach Tri and F4 pricing was this:

1) If a quin AND exacta combo looks to be paying "overs" (comparing my price to the should pay)
2) For trifecta bets take the above combinations and add any other runner in a trifecta IF the runner is an over on the win market
3) For F4 bets take the above combinations in 1 and add any other runner in a F4 with any 2 runners that are overs in the win market

Thoughts?

« Last Edit: 2020-Jun-03, 08:55 AM by bascoe »

Offline fours

  • Group 1
  • User 704
  • Posts: 6722
« 2020-Jun-03, 08:59 AM Reply #1 »
bascoe,

People have probably heard enough from me on these subjects so I will just say be careful with your assumptions.

Even people that have spent years in the high value rooms make bad assumptions. They tell you you are wrong and find out that they have been operating under a false belief and or beliefs.

Fours

Offline bascoe

  • Open
  • User 2568
  • Posts: 146
« 2020-Jun-03, 01:27 PM Reply #2 »
Quote
be careful with your assumptions

Sure - all I can assume is that my prices are the 'best' I can make them, and that they have proven to be profitable on level stakes win betting.

JFC went to considerable effort a while back on the extrapolation of a tri price from exactas, but they were still just rough estimates.

In order to simplify the process I though about the process I outlined above.  I though some of you may have had similar experiences.

I will give it a try on some races today and see how it goes.

Offline fours

  • Group 1
  • User 704
  • Posts: 6722
« 2020-Jun-03, 02:00 PM Reply #3 »
bascoe,

There is much more to it than prices.

Fours

Offline bascoe

  • Open
  • User 2568
  • Posts: 146
« 2020-Jun-03, 02:04 PM Reply #4 »
More than prices>

Ok I don't have a divining rod nor do I have esp - what else are you suggesting needs to be considered?

Staking? Races are run honestly?

No need to be coy, if you have a view feel free to share.

bascoe

Offline fours

  • Group 1
  • User 704
  • Posts: 6722
« 2020-Jun-03, 02:56 PM Reply #5 »
bascoe,

You are looking at prices but do others get to see prices you don't?

Do others get set after you have sent your bets away?

Are others relying on a rebate to get into profits thus crushing your potential collects that your maths suggest are possible after you send your bets?

Are jackpot divs available?

Are jackpot div calculations different in this locality and or with this TAB? They are by the way.

Have normal conditions changed recently?  eg recently NSW tab divs paid more than VICTAB divs for first fours on a Sat despite having smaller pools.

Is the change due to new players or merely a short term erroneous phenomenon?

And that is just for starters - not going to do it all for you.

Fours
« Last Edit: 2020-Jun-03, 03:03 PM by fours »

Offline PoisonPen7

  • Group 1
  • User 55
  • Posts: 21836
« 2020-Jun-03, 03:42 PM Reply #6 »

For a variety of reasons calculating a should pay estimate for trifectas and F4 is imprecise.  As I price each runner I was thinking a simpler way to approach Tri and F4 pricing was this:

1) If a quin AND exacta combo looks to be paying "overs" (comparing my price to the should pay)
2) For trifecta bets take the above combinations and add any other runner in a trifecta IF the runner is an over on the win market
3) For F4 bets take the above combinations in 1 and add any other runner in a F4 with any 2 runners that are overs in the win market

Thoughts?

Bascoe that sounds like sound reasoning to me.

It simplifies what is a very complex task made even more complex because you are comparing prices to your own assessment (compared to those who just look at the odds and assume this is the "true" price - as jfc does).

Offline bascoe

  • Open
  • User 2568
  • Posts: 146
« 2020-Jun-04, 12:43 AM Reply #7 »
Well it was a short experiment - I could see quite easily the flaws in my proposition.  Value is not a linear exposition based on simple numbers, I can see that tri combos even past my exacta should pays were overs even though they were unders on win prices..

back to the white board -

the thing is...we know we can win, have done it before... just need to get in front of the curve again.

One thing I know, and it plays to the jfc theory...I made a lot more profit before 2008 when rebates became rampant.  Not a lot of rebates around now but the pools have been stripped so I am working on a whole new data set.

bascoe

Offline PoisonPen7

  • Group 1
  • User 55
  • Posts: 21836
« 2020-Jun-04, 06:56 AM Reply #8 »
Well it was a short experiment - I could see quite easily the flaws in my proposition. 

Geez that didn't take long. One day's worth of data?

If you could see so easily then why did you post in the first place  :what:

Offline fours

  • Group 1
  • User 704
  • Posts: 6722
« 2020-Jun-04, 09:26 AM Reply #9 »
bascoe,

The level of competition changes over time and is an assumption to check.... it's one reason why jackpot pools can disappoint at times.

You mentioned rebates changing and maybe we can thank the POC for a bit of that and the ATO too!

Corporates can change their behaviour over time as well and are reported to have done so after the POC ie much less betting back into the TAB pools now and maybe their rebates got removed too.

Corona and the above all made pools tiny for quite a time but some recovery occurring now..... only some.

So just looking at prices is rather simplistic but still has a small place or base.

Fours


Offline innerwiz

  • Class3
  • User 972
  • Posts: 36
« 2020-Jun-05, 05:15 PM Reply #10 »
I need to comment here regarding Trifecta calculations and prediction.

TABcorp now provide the full combination Trifecta pool dividends.  Not Trends, but the full 3dimensional combos (1-2-3 = xxxx.xx div) via their TABcorp Studio Premium service.

It's via a JSON file encoded binary via AVRO a technology to shrink filesizes by putting them into binary.. link: https://avro.apache.org/

If you're trying to estimate Trifecta Dividends you have to realise at least a good number of punters have access to the full combo list.

And it makes you wonder what do they provide to the Isle of Man?
« Last Edit: 2020-Jun-05, 05:54 PM by innerwiz »

Offline fours

  • Group 1
  • User 704
  • Posts: 6722
« 2020-Jun-05, 05:57 PM Reply #11 »
innerwiz,

Nice to see you touch on some of the things I was mentioning.

Pretty sure the z man gets last look for automated bets where pool size plus combos are considered - another assumption there.

Fours

Offline innerwiz

  • Class3
  • User 972
  • Posts: 36
« 2020-Jun-05, 06:15 PM Reply #12 »
fours,

I'm all for the best winning.  But it should be a level playfield, with everyone given the same information.

Offline sobig

  • Group 2
  • User 583
  • Posts: 3722
« 2020-Jun-05, 06:22 PM Reply #13 »
Anyone can get trifecta on the website.

I know it is not the full listing but when you enter numbers prior to placing a bet the approx shows

Offline wily ole dog

  • Group 1
  • User 218
  • Posts: 27394
« 2020-Jun-05, 06:33 PM Reply #14 »
Fours, once again thanks for pushing me to F4 betting.
I was dead keen on the winner in Toowoomba 7. Backed it straight out for $40 & $10 on the F4
The F4 divvy was far superior. Both returned the same $$$

One of many examples I could give you :thumbsup:

Offline fours

  • Group 1
  • User 704
  • Posts: 6722
« 2020-Jun-05, 06:44 PM Reply #15 »
Wily one.

When you do the work and

bet smart as well

 the returns come

and

they are far beyond what is commonly thought possible.

Congrats

Fours
ps I just missed a big day when Revenire won and my 150/1 pop finsihed well for 5th but in the photo ( called a photo but I did not think so ) at Cessnock where barriers help you a lot..... if you let them.   You already know that is part of the game but I get my 4ths when needed too. For the unitiated I just missed a collect of 8 grand plus on 6 combinations only ( bet spread across tabs )
« Last Edit: 2020-Jun-05, 06:56 PM by fours »

Offline jfc

  • Group 1
  • User 723
  • Posts: 7263
« 2020-Jun-07, 07:00 PM Reply #16 »
I need to comment here regarding Trifecta calculations and prediction.

TABcorp now provide the full combination Trifecta pool dividends.  Not Trends, but the full 3dimensional combos (1-2-3 = xxxx.xx div) via their TABcorp Studio Premium service.
....

Was I taken aback by this assertion?

That would be an understatement.

Anyway, after extensive and arduous investigation I can confirm that this is actually true.

Up to a point.

You see, not only has Tabcorp kept this a secret from the masses, but it has also kept this secret from registered Premium Players like moi.

The consequence is that any of you non-Premium Players attempting a Trifecta or F4 have no realistic expectation of evading ruin.

So why are you again acting like stunned mullets!

Offline fours

  • Group 1
  • User 704
  • Posts: 6722
« 2020-Jun-07, 07:14 PM Reply #17 »
All,

jfc is probably correct if you bet every race.

But don't do that.

I pass most races AND  bet very different strategies for different conditions.

Trying to impose ''one size fits all' approaches means you will definitely donate over all.

Fours
ps told you that high value room players get their assumptions wrong! So imagine how many some one starting out has wrong....

Offline bascoe

  • Open
  • User 2568
  • Posts: 146
« 2020-Jun-12, 01:17 PM Reply #18 »
"TABcorp now provide the full combination Trifecta pool dividends.  Not Trends, but the full 3dimensional combos (1-2-3 = xxxx.xx div) via their TABcorp Studio Premium service."

I can call Q and Ex estimates but the Trifecta pool returns an empty array.

So is the call available to ALL tabstudio users - or only Premium players?

Offline jfc

  • Group 1
  • User 723
  • Posts: 7263
« 2020-Jun-12, 04:44 PM Reply #19 »
Considering Tabcorp told me that there is no documentation it's hard to know what to say.

I believe the feature is only available only to Premium Players who have read this thread.

Tabcorp could easily make this available to everybody like HKJC does,

But such an action would be high out of character for them.

I repeat that anyone not in on this clandestine information should avoid those Exotics like a pandemic.

Offline bascoe

  • Open
  • User 2568
  • Posts: 146
« 2020-Jun-13, 09:50 PM Reply #20 »
Re hkjc - is it possible to get all perms for  tierce rather than only top 10, banker and investment via  http call?

Offline stiffarm

  • Class2
  • User 2721
  • Posts: 27
« 2020-Jun-22, 06:59 PM Reply #21 »
Based on what happens on site, if you send the appropriate payload to this endpoint: https://api.beta.tab.com.au/v1/pricing-service/accounts/{account_no}/enquiry

Then you will receive the approx dividends you request.

Offline stiffarm

  • Class2
  • User 2721
  • Posts: 27
« 2020-Jun-22, 07:26 PM Reply #22 »
Turns out you don't even need an account to get the approx divs. You can use this link: https://api.beta.tab.com.au/v1/pricing-service/enquiry

And an example payload would be:

Code: [Select]
{
    "clientDetails": {
        "channel": "TABCOMAU",
        "jurisdiction": "VIC"
    },
    "bets": [
        {
            "stake": "$0.00",
            "type": "PARIMUTUEL",
            "meetingCode": "M",
            "scheduledType": "G",
            "raceNumber": 2,
            "meetingDate": "2020-06-22",
            "betType": "TRIFECTA",
            "flexi": true,
            "legs": [
                {
                    "positions": [
                        {
                            "selections": [
                                1
                            ]
                        },
                        {
                            "selections": [
                                2
                            ]
                        },
                        {
                            "selections": [
                                3
                            ]
                        }
                    ]
                }
            ],
            "enableMultiplier": false,
            "source": "racing.race-nav.race.bet-type"
        }
    ]
}

Offline wily ole dog

  • Group 1
  • User 218
  • Posts: 27394
« 2020-Jun-22, 09:25 PM Reply #23 »
Keeps telling me itís a page error

Offline stiffarm

  • Class2
  • User 2721
  • Posts: 27
« 2020-Jun-22, 09:36 PM Reply #24 »
Do you have an example of the code you're using for the POST request? It's returning the correct JSON output for me.


BACK TO ALL TOPICS
Sitemap