We want to share an important update with our Betfair Community users.
Over the coming months, the Betfair Community platform will be winding down. The final active day for the platform will be December 21st. From December 22nd, users will no longer be able to access the platform.
Your discussions, insights, engagement, and, most importantly, feedback have helped shape the platform and guided innovation. Your feedback continues to be important to us, and we will continue to listen via our dedicated feedback channels, as well as through product testing and innovations.
To share your feedback and experience, visit: https://support.betfair.com/app/home
General Betting
There is currently 1 person viewing this thread.
30 Oct 10
20:40
Joined:
Date Joined:
12 Mar 07
I was looking at two series of numbers generated by separate selection techniques from a comparably similar betting idea, and the following occurs.
Both sets generate fairly approx. good profit results.
So as I couldn't really separate them for betting purposes, I decided to bet both of them.
That is just add them together with the results of course being just the sum of the two.
But then really just for fun, I played around with a few other things.
I tried mutiplying the two of them and taking the square root of the result.
I got a relatively poor performance out of that but I did notice a strange characteristic on the fluctuations of the results, which indicated to me that just throwing away this sqrt idea was not perhaps a good thing to do.
So, essentially just out of curiosity, I subtracted the square root amount from the added amount.
eg ( x+y)-(SQRT(x*y))
And lo and behold itsmoothed out in quite a material manner the peaks anfd troughs of the simple ( x+y) results.
As result I could now leverage up if using the new, more complex approach and increase profitability substantially.
I am talking literally about a series thousands of selections here.
Not being a statistician, I ask.
Have I accidentally stumbled onto some well known ststistical smoothing process in this square root thing.
Or is it all likely just be a complete fluke and not likely to continue as it has ?
Both sets generate fairly approx. good profit results.
So as I couldn't really separate them for betting purposes, I decided to bet both of them.
That is just add them together with the results of course being just the sum of the two.
But then really just for fun, I played around with a few other things.
I tried mutiplying the two of them and taking the square root of the result.
I got a relatively poor performance out of that but I did notice a strange characteristic on the fluctuations of the results, which indicated to me that just throwing away this sqrt idea was not perhaps a good thing to do.
So, essentially just out of curiosity, I subtracted the square root amount from the added amount.
eg ( x+y)-(SQRT(x*y))
And lo and behold itsmoothed out in quite a material manner the peaks anfd troughs of the simple ( x+y) results.
As result I could now leverage up if using the new, more complex approach and increase profitability substantially.
I am talking literally about a series thousands of selections here.
Not being a statistician, I ask.
Have I accidentally stumbled onto some well known ststistical smoothing process in this square root thing.
Or is it all likely just be a complete fluke and not likely to continue as it has ?
sort by:
Show
per page
Replies: 68
|
By:
The silence is deafening !
Seriously FAFH, I suspect 99% of the forum haven't a clue as to what you are doing, hence the silence. |
|
By:
I would have thought the formula I put in was pretty basic and self explanatory.
Isn't it ? Btw I don't think I know what I'm doing a lot of the time. Hence all the questions. |
|
By:
I think that your example is too abstract for most people to try to work out what you are doing
|
|
By:
I wouldn't think it was too abstract for any serious statistician though.
Perhaps he would need to know a few more factors probably. Ass away, and if I can supply the answers I will. But as I said, I'm not a statistician. I'm just playing around, in this case, with some numbers to see if I can extract a better performance from them. I'm really just trying to ascertain if I've stumbled onto something meaningful or just a short term freakish phenomenon. Not to worry. If there are no answers out there, so be it. The basic numbers/results are good without having to trick them around in any manner, and I can carry on going with that as I have done to date. Thanks for reading though. |
|
By:
" Ass away--- " ?
A freudian slip ? Anyway obviously I meant " Ask away--- " |
|
By:
Like the above poster I have no idea what you're doing and what the end game is but without even needing to calculate anything I can tell you several things, mainly that the larger a and b are the larger (a+b)-(a+b)^0.5 will be in relation to (axb)^0.5. [Probably] less importantly the larger the gap between a and b the larger the gap between the two expressions and if a and b are both below 2 then the first expression will actually be larger.
In short my answer to your initial question is 'I don't know the answer'. My statistical skills are limited at best. Although is the second expression really that 'more complex'? Perhaps statistically (who knows) but certainly not to someone who has even a basic grasp of algebra. Best of luck. |
|
By:
( x+y)-(SQRT(x*y)) will smooth out the variation between x and y.
So will lots of other functions. What are x and y? |
|
By:
I didn't mean complex in an algebraic sense.
I agree it's extremely simple. I meant complex in the sense of is there something statistically significant that perhaps I'm doing, but am unaware of,being a layman. Most probably not. I can accept that. But I thought no harm asking, just in case there was. |
|
By:
x and y are just bet amounts, given out to me by an algorithm designed to select and calculate certain bet selections and relevant amounts on soccer games.
The x number comes from one algorithm and the y from another. But they both refer to same game at the same time and bet in the same direction on the same thing. So as I said I can combine them in essentially any manner I want. I'm interested in your statement that the formula I have accidentally stumbled is in fact regarded as a smoothing type operation. Could you perhaps expand a bit on that for me, in layman's terms. Thanks in advance. |
|
By:
i like this one, it smooths things out also
E = mc2 *and i also agree, it's extremely simple |
|
By:
I don't have enough energy to try that one out.
But thanks for the welcome input Lusi. On form as per usual. |
|
By:
just kidding FAFH, don't take it wrong
i like your threads, they stir the pot [;)] now, on a more serious note, your threads do raise questions also give some answers/clues thanks to the contributions of many that do it in good faith, there are a lot of different POV's and also shows that you are moving those brain cells around...you have been opening some interesting threads in the GBF, so keep them coming *btw as im not a statistician i cant really give you any meaninful input, im more of a gestalt kind of guy [smiley:crazy] |
|
By:
Kind words Lusi. Tks.
I'm not used to such feedback. In fact, I'm trying really hard to one day create a thread that doesn't stir the pot. But at the moment, I'm finding it a bit beyond me, for some unaccountable reason. Maybe this is the one. It's so boring and abstract, I really can't see how it can raise the hackles in anybody. Probably the ignorance and naivety of it might p!ss somebody off. Something will that's for sure. |
|
By:
Are we talking about a staking system?
|
|
By:
If I understood, you bet on soccer, generating "random numbers"..around here, on soccer plays the overround is about 0,1 per cent..It means that random betting anything, whatever the generation source of the bet/stake, your expected result is to lose 0,1 per cent, on tournaround, as a gross margin, less the betfair tax on winnings..not so bad, a real resistent system, but, definetely, a system with the aim to lose money...
I'm in the game by many years and can confirm that the approach to "random numbers" is profitable, really so much, but you must have a pre set strategy on how to match them and how to use the results...hope this is of some utility... |
|
By:
first thing that struck me was that you are generating random numbers, using one of the two main methods.
however, it does sound like you are now guessing i.e. no mention of the law of large numbers; the size of your dataset; regression analysis [linear, in this case]; P-Values; Chi-Squared analysis-of-fit, etc. It is 7 years since I completed my stats degree, and although rusty in places, nowhere do I instantly recognise the "square-root of a product" as a distribution i.e. binomial/poisson/Std. Normal, etc. And I use a distribution to price my football bot. |
|
By:
Have I accidentally stumbled onto some well known ststistical smoothing process in this square root thing.
what you have stumbled upon is a way of smoothing some historical results. i would strongly suggest you don't read anything into it. |
|
By:
Giotto
I don't know where you got the idea that I bet using " random" numbers". Not true at all. I would agree that any strategy using such an approach will ultimately lose. I fluctuate my stakes on a formula type basis most certainly, in that there is no subjectivity involved, but there is a mathematical idea/foundation to the fluctuations. But what I do in this area is not really the point of the question.. Trust me when I say, as I did in my few earlier posts, the two series of bet amounts do produce consistent profits on a stand alone basis. I am only looking at various creative ways of combining them, apart from simply adding them together. Maybe I am wasting my time, and there is no other way to use them and, if so, I am perfectly content with that. I just thought I would point out the quite interesting results I got when doing what I did with the SQRT thing, in the hope, not necessarily the expectation, that it would trigger off some knowledge in others, particularly those skilled in statistics, which I'm not. Thanks for your comments though. |
|
By:
Look at the risk of insulting the intelligence of any or all who are reading this, it is pretty simple to see what I am doing here.
If the two bet amounts are the same, then I'm just halving the bet size in combinong them with this fancy formula. However if they're not the same I'm not halving them. And this is, for some unknown reason to me, smoothing out the peaks and troughs in performance. And I'm asking if is there perhaps some underlying statistical explanation for the effect I'm getting. The effect is quite marked and long running. Trust all is a bit clearer now. It's not all that important. More of a curiosity thing really. |
|
By:
So I take it that this is a staking system.
Are x and y bets on the same event? |
|
By:
Frog Hair, all this nonsense about X=Y add Z multiply by your birthday and the winner is a stand out
zipper dont do that nonsense let have a bit of fun and win a least £500.pounds now £500. pounds is a good days wage . You put up your selection in the following races zip will put up mine and i will beat any formula out of sight, hey you have nothing to lose only your pride . The races are Lingfield 1.10 2.50 3.55 4.25 . Huntingdon 2.30 4.05 Carlisle 2.05 3.45 .. That makes 8 races zip will 8 bets today Hope to see you later starting 1.10..zzzzz |
|
By:
zipper ...he she knows very little re horse race betting....so you will have to get someone else to "play" with
 ....not me ...I'm too "busy"! |
|
By:
DFC the Challenge is open to every one all are welcome hotshots , smart arses , aftertimers ...every one and Frog Hair
|
|
By:
Look at the risk of insulting the intelligence of the OP , don't delude yourself with a staking plan you have happened across.
|
|
By:
DFC hey i like that " Im too busy ".. zip has followed your thread and have got to say your good, very good .. not has good as me but your very good zipper .zzzzzzzzzzz
|
|
By:
your asking what is the relationship between (x+y)-sqrt(x*y) and what?
|
|
By:
DFC looks like this Hairy Frog is a no show ..
Well zip aint surprised you know how it is girls 3 pints of beer and a smell of the barmaids apron and them guys are falling all over the street |
|
By:
Look folks, the essence of my question is for some unknown reason evading most if not all of you ( zipper of course excluded).
There is no point, on this thread at least, in going into unrelated side issues, such as the merits or demerits of staking plans per se. So let's leave it all there then. Nothing gained. Nothing lost. Thanks for your general interest in replying as you have thus far. |
|
By:
Frog ~ you have started many threads, and asked many questions. You have written at length, sometimes cryptically and sometimes enigmatically. But finally I think I have realised what it is all about. And that is, you're just not that smart.
|
|
By:
So, if it isn't a staking system, what is it? And what exactly are x and y? And what do you do with the answer?
We are curious about this, but your opening post does seem confusing. |
|
By:
Whatever.
You're all right. I'm all wrong. Sorry Marksman ( you're a earnest, good guy) but I just can't be bothered with it all any more, not when all the usual sad, sorry lot of sh!tkickers come out time and time again. Unfckinbelievable as they say. |
|
By:
FAFH,
Well then please stop starting this fishing expeditions. Having re-read the thread in full [and indeed, other threads], you seem to exist on this forum purely to unearth the edges of other players. I'm sorry I bothered trying to help you. Do your own research - you're worse than the robots that try & manipulate prices in football games immediately after a goal has been scored & the market suspended... |
|
By:
A pissant, harmless thread like this turns into another slanging match with the usual suspects, and even a few unusual new ones.
I'm really sorry to say it but there is a relatively small bunch of @rsewipes on here who are totally negative and soul destroying. The hate BF, they hate the bookies, they hate my posting style, they hate the commission they have to pay, they hate everything it appears. They just suck the blood out of anybody who comes on here who apparently does not roll over and show due deference to their views and opinions. You can't debate with them, because they hate debating. You can't question them, because they hate questions. You can't be polite to them, because they hate being polite. You can however be angry with them, because they love to be angry. I sure hope their gambling lives are choc a bloc with excitement, because the rest of their lives sound as dull as dishwater. Well I'm off down to the yacht club to meet up with a few of my exciting, stimulating, successful real life mates. No envy or jealousy there. They've all got big dcks. Just put in the last few sentences only to really p!ss off those who know they are who I'm talking about. To the rest of you, all the very best in your gambling exploits. And yes I'll pick up my dummy as I go out the front door. |
|
By:
seek help
|
|
By:
cracking threads these |
|
By:
here we go
|
|
By:
Well I'm off down to the yacht club to meet up with a few of my exciting, stimulating, successful real life mates.
No envy or jealousy there. They've all got big dcks. Tea all over the screen. Meltdown post of the year. |
|
By:
Actually I think I have got it: Suppose system X says have £10 on a team, and system Y tells you to have £10 on the other team, you calculate your bets as follows:
£10 on team X and £10 on team Y minus the square root of 10 times 10= £10 So the total stake will now be £10, so we have £5 on each team and hope that its not a draw. |
|
By:
I can see how this approach will cause smoothing, but that is not the same as being profitable.
|