News
Current BetFair Sports API status: online. Current BetFair Games API status: online.

WellDone Betting Bots -- May 21, 2012, 23:27
Pages: « 1 2
Print
Author Topic: backing by kelly plan question  (Read 4774 times)
bamthwok
Élite
Full Member
*****

Reputation: +4/-0
Thanks!
Posts: 113


« Reply #15 on: September 04, 2010, 16:06 »

Hi,

The first 4, is the betfair odd (average matched price).

I suggest to calculate the stake on average price.

In real betting, your trigger will fire within a certain odds range. If the price range is quite narrow then the average odds might work guess.

The more historical data you have to base your percentage on the better. Most important: your trigger must be already in profit at level stakes!

Of course you could try to calculate your percentage advantage for individual odds. But I believe it's a lot of guessing. But if you know for sure, that your trigger  has a bigger advantage let's say for lower odds compared to higher odds then you can try to adjust the stake accordingly.

Hope that makes sense.

Cheers,
bamthwok


Logged


mcbee
Élite
Hero Member
*****

Reputation: +49/-2
Agree
Gender: Male
Posts: 1250



« Reply #16 on: September 04, 2010, 16:38 »

hi
is the first 4 the current odds or the average odds taken from the history.



mcbee
Logged



If you don't try you will never know ?
bamthwok
Élite
Full Member
*****

Reputation: +4/-0
Thanks!
Posts: 113


« Reply #17 on: September 05, 2010, 04:42 »

Hi,

The first 4 (average odds) is taken from the history.

Cheers,
bamthwok
Logged


mcbee
Élite
Hero Member
*****

Reputation: +49/-2
Agree
Gender: Male
Posts: 1250



« Reply #18 on: September 05, 2010, 08:07 »

hi
NOW then, that was the answer that i expected.
tsm and mfp have interpreted the first odds as current odds, so the result is that the lower the odds the less the stake and the higher the odds the more the stake, i have changed this to the average history odds and it works great, but i needed somebody esle to confirm my thoughts.
could you check the formula for the backing kelly and confirm the findings, PLEASE.



mcbee
Logged



If you don't try you will never know ?
bamthwok
Élite
Full Member
*****

Reputation: +4/-0
Thanks!
Posts: 113


« Reply #19 on: September 06, 2010, 07:46 »

Hi,

I finally found some time to have a look at the Kelly Trigger and also at TSM formula.

Indeed, according to my understanding both of them are different from what I think is right.

My understanding is that there are only two key variables: average historical odds,  and the estimated strike rate of future events based on historical strike rate of your betting strategy.

The variable "expectation" does not make any sense to me, or there is something that I don't understand.

Also, there is no need to consider commission. Commission will be automatically deducted from the betting bank. But one has to understand that one will lose money if average odds and 100/strike rate are the same.

The formula can be simplified as per attachment.

Divisor has been set to 1 (original Kelly) and the trigger will look up the current betting bank (here: test funds). For a long term strategy it might be enough to adjust the bank once a day.

Cheers,
bamthwok
Logged


mcbee
Élite
Hero Member
*****

Reputation: +49/-2
Agree
Gender: Male
Posts: 1250



« Reply #20 on: September 06, 2010, 12:00 »

hi
thank you for the time you are spending on this problem, i am also looking at the lay kelly and have emailed tsm a week ago, they are now looking into the formula.

mcbee
Logged



If you don't try you will never know ?
Pages: « 1 2
Print
 
Jump to:  

Anone designed by pixelslot.
Powered by SMF 1.1.16 | SMF © 2011, Simple Machines