My official betting thread

Remove this Banner Ad

Status
Not open for further replies.

Log in to remove this ad.

The same word where you also then say that you don't use full kelly, you use a watered down version, the simultaneous even calculator.
Click to expand...

This is again another moronic statement from this poster.

'Full kelly', 'true kelly' or whatever you want to call it ADJUSTS FOR SIMULTANEOUS EVENTS.

Another question for you to answer.

You have 7 x 60% plays going off simultaneously, what is your bankroll sizing?

It can't be 15% each (as this is 105% of bankroll), which is roughly what an individual play would require using kelly in true sequential order, which is how you are thinking about it.

It's 5.62% per play using 'real' or 'true' kelly (or whatever you want to call it). That's not so crazy now is it?
 
it involves combining models with imperfect inputs into a less flawed overall output
Click to expand...

The text highlighted is what you have got right.

It's got nothing to do with 'imperfect' inputs (whatever that means).
 
By imperfect inputs he means approximate winning probability (as generated by your model) and not true win probability.
 
brett128 said:
The text highlighted is what you have got right.

It's got nothing to do with 'imperfect' inputs (whatever that means).
Click to expand...
i guess I misread what you were saying, I took it as meaning you had less confidence in the NRL models' accuracy

so am I to assume then that you are confident the NRL model is as accurate as the AFL one?
 
brett128 said:
My current index figure is NOT arbitrary. This is my ACTUAL bankroll fluctuation. So OF COURSE it takes account all prices I got down at.
Click to expand...
Why so angry Brett?

This just proves that what I have posted is correct.

Did you not state at the start of the thread that you would be using a base figure of 1.91 to keep record?

Now you have changed your tune, and are producing an index figure based off of numbers that aren't provided anywhere in this thread.

They are what I called fantasy.

There is no need to get all high and mighty, use bolding or italics, you have been found out, it is not that big of a deal, but when you continually try to bolster your models output with fiddled numbers, that just makes you look desperate.

To me anyway.

I also think you seem angry at the fact you spent thousands upon thousands of dollars at University to learn exactly the same as what is plastered all over the internet, which only costs the price of a broadband connection.
 
brett128 said:
Hey everyone, I've only really scanned this board the last few days, but it seems multis are all the rage here. Yes, I use multis too, but they are only the small part of a kelly optimal solution in maximization of expected bankroll growth.

If you're a zero edge sports bettor (and tbh, if you haven't come across terms like the 'kelly criterion' in your sports betting forays before, you most likely have zero edge);

Your goal should be to:

To lose money as slowly as possible whilst maximising your expected utility (entertainment/contentment).

That's the only way to rationalize in my own mind why zero edged punters bet.
Click to expand...

Has your opinion changed after dealing with it first hand?
 
Before the Weekend results were 22-27 and a loss of 5.03 units
Weekend Results were- Went 6-2. I actually had a 1.5 unit bet on the Kangaroos Win straight out as well that I have not included. As per the line competition thread it was my best play of the weekend. Not including that I outlaid 9.05 units for a return of 13.42 units.
Leaves me at 28-29 leaving me still 0.66 units in the hole.
 
brett128 said:
Righto smart arse, answer me this (I've asked this qustion before, with no answer from anyone trying to refute me):

You make 100 bets ATS, you go 0-0-100. What is your win % ATS?

You haven't lost any money, but you haven't made any money either. So what's your answer going to be?

If anything, it would be slightly BETTER than half a win (0.5), because a 50% ATS record would imply a loss has occurred.
Click to expand...
You do understand simple mathematics don't you?

How can you have a win % when you are yet to have a win?

Do you know how a percentage is worked out?

If so, how do you divide zero by 100?

The answer, and it may even be too simple for you, is you don't have a win percentage until you actually have a win. Pushes do not count.
 
I dont really worry about all the statistics and theories in here. Im curious though, are you satisfied with your model? I am in no way looking to make a living off gambling but im curious, as when betting on a sport i watch regularly (afl) at figures around the 1.9 mark i would expect my observation and knowledge to get me to at least 50%.

If i was betting high stakes and my model had lost me 21% of bankroll i would be questioning it.
 

(Log in to remove this ad.)

The last 6-12 months has seen more (local) forums inundated with threads along the lines of "Corporate X shut me down", "Corporate Y limited my bets" etc.

I stumbled across this today on a forum I'd forgotten existed. It describes an account recently shut down by Bet365 and their current practices, which is more or less industry-wide:

"lasted two weeks, first week was restricted to win $1000, second week the account was closed because they consider it uneconomical. Checked the profit/loss on that account yesterday.... she finished DOWN $1650! They don't care about winners or losers, anyone that is price sensitive and takes best odds available is removed from their system".


Keep getting down. :thumbsu:
 
i guess I misread what you were saying, I took it as meaning you had less confidence in the NRL models' accuracy

so am I to assume then that you are confident the NRL model is as accurate as the AFL one?
Click to expand...

When combined, it's as good as anything out there.
 
By imperfect inputs he means approximate winning probability (as generated by your model) and not true win probability.
Click to expand...

When three orthogonal models are entered into Bayesian combination, the combined win % will be more robust (more closer to 'true' win %) than what they are individually.
 
This just proves that what I have posted is correct.

Did you not state at the start of the thread that you would be using a base figure of 1.91 to keep record?

Now you have changed your tune, and are producing an index figure based off of numbers that aren't provided anywhere in this thread.
Click to expand...


No, this proves I have little patience for half wits.

You can use whatever 'base figure' you want. $1.88, $1.91, $1.90, $1.95. It's up to you. My goal is to go 55+% ATS, it has been from day one.

Let me spell this out for you. If you want to grade me against an average of '$1.88', I would need 53.1% long term to break even. At '$1.95', I would need 51.2% long term winners. So you can see, even at the worse end of the spectrum, 55% ATS will beat even a $1.88 line long term.

The 'index figure' is ancillary to achieving my stated 55% ATS goal. It simple tracks my personal fluctuation, as an aside to my main goal.
 
brett128 said:
When three orthogonal models are entered into Bayesian combination, the combined win % will be more robust (more closer to 'true' win %) than what they are individually.
Click to expand...
Yeah I understand it. Just explaining what he meant by "imperfect inputs".
 
brett128 said:
No, this proves I have little patience for half wits.

You can use whatever 'base figure' you want. $1.88, $1.91, $1.90, $1.95. It's up to you. My goal is to go 55+% ATS, it has been from day one.

Let me spell this out for you. If you want to grade me against an average of '$1.88', I would need 53.1% long term to break even. At '$1.95', I would need 51.2% long term winners. So you can see, even at the worse end of the spectrum, 55% ATS will beat even a $1.88 line long term.

The 'index figure' is ancillary to achieving my stated 55% ATS goal. It simple tracks my personal fluctuation, as an aside to my main goal.
Click to expand...
Ever get the feeling you are confusing yourself?

It has nothing to do with your 'goal'. It has everything to do with where your current bank balance sits.

You cannot possibly, on the numbers provided, and at your recorded price of 1.91, be only 21% down. Impossible.

But we both know how you have come to that figure. There is the bet on Geelong you conveniently missed. There is the GWS first quarter cover that you conveniently didn't post. There is the prop bet that you and your mate got down on that softened your loss last week.

Afterall, these are included in your index figure of 79, if that is your ACTUAL bankroll fluctuation.
 
The answer, and it may even be too simple for you, is you don't have a win percentage until you actually have a win. Pushes do not count.
Click to expand...

You're completely wrong.

So in your mind, how do you calculate ROI% for a record of 1-0-100? Do you only count your turnover on the one bet that hasn't pushed?

Let's see If i'm following your warped line of logic correctly.

If I had 101 bets at 1 unit each, you would calculate your ROI% on these 101 bets to be 91% (1 winning ATS bet at $1.91), even though you have physically bet 101 units?

Accommodating the turnover of a push in sports betting is like following the precepts of accounting, it needs to be recorded somewhere. You can't just say 'they don't count'.

If BHP turned over $10 billion and made $0 profit on it, they wouldn't say the turnover didn't count now would they?
 
brett128 said:
You're completely wrong.

So in your mind, how do you calculate ROI% for a record of 1-0-100? Do you only count your turnover on the one bet that hasn't pushed?

Let's see If i'm following your warped line of logic correctly.

If I had 101 bets at 1 unit each, you would calculate your ROI% on these 101 bets to be 91% (1 winning ATS bet at $1.91), even though you have physically bet 101 units?

Accommodating the turnover of a push in sports betting is like following the precepts of accounting, it needs to be recorded somewhere. You can't just say 'they don't count'.

If BHP turned over $10 billion and made $0 profit on it, they wouldn't say the turnover didn't count now would they?
Click to expand...
No, I am completely right.

Do you often ask totally different questions in response to your original one being 'refuted'?
 
Afterall, these are included in your index figure of 79, if that is your ACTUAL bankroll fluctuation.
Click to expand...

Yes that's correct. I'm talking about my actual bankroll fluctuation. I'm not going to figure out for the armchair experts here where my bankroll would have been had I not taken those bets.

I'm here to be judged on achieving a 55%+ ATS record. That's what I said from day one.
 
No, I am completely right.

Do you often ask totally different questions in response to your original one being 'refuted'?
Click to expand...

You don't bet ATS much do you? I'm looking at ROI% just so I can get through to your brain in a different way, in order to help you understand.

Now since my ROI% question relating to a push is layed out there for you to see, why don't you answer it for me?
 
brett128 said:
Yes that's correct. I'm talking about my actual bankroll fluctuation. I'm not going to figure out for the armchair experts here where my bankroll would have been had I not taken those bets.

I'm here to be judged on achieving a 55%+ ATS record. That's what I said from day one.
Click to expand...
Just yesterday you disagreed when I said that your numbers don't stack up to a -21% bank hit.

brett128 said:
Every single bet is on this thread, week by week.
Go take a look, figure out timings for all the events, figure out what I have bet simultaneously, and then come to an approximation of how my betting run would have eventuated.

Then let me know if it's not -21%.
Click to expand...
Is this where I again say that I was correct.

It is not -21%.

You asked.
 
Status
Not open for further replies.

Remove this Banner Ad

Similar threads

H
HTP Purton Bets
Replies
11
Views
1K
HTPunter
H

Remove this Banner Ad

Back
Top