staking plan
dave
scenario:
system has 80% strike rate at average odds of 1/2 (1.47 to be exact)
increasing the stake after a loser seems sensible as the next one is far
more likely to be a winner than a loser, but this does not take advantage of
very long winning runs (23 longest so far).
any ideas on this wet and miserable friday afternoon?
michael adams
"dave" <t...@thegroup.com> wrote in message
news:2o3uq8F...@uni-berlin.de...
> can anyone suggest a staking plan to maximise profits in the following
> scenario:
>
> system has 80% strike rate at average odds of 1/2
> (1.47 to be exact)
>
> increasing the stake after a loser seems sensible as the next
> one is far more likely to be a winner than a loser, but this
> does not take advantage of very long winning runs (23 longest so far).
If what you claim is true than that's a very impressive 20% return
on turnover. So the best strategy there, is to say double your stake
once you've doubled your bank.
And keep on like that. And you'll be a millionaire in no time.
Or at least depending on the frequency of your bets. In other
words if there's only one bet a month, things will take a bit
longer.
The more cynical might suggest that you're simply going through
a lucky patch, and its all going to go back to the bookies. So
you should stop right now while you're ahead, and spend your
winnings.
Altering your stake from horse to horse is no good, as differnt
sequences of results will give a brilliant or relatively disastrous
outcomes, depending on what system you use. Wheareas by definition
you can't chose the way the results are going to fall beforehand.
You also say the next one is far more likely to be a winner
than a loser. Yes well they all are, if they're all genuine
1/2 shots. They should win twice as often as they lose.
So that's no reason for altering the stake in itself
anyway.
So level stakes, and double the stake once you've doubled the bank.
When you've managed it ten times in a row, post back to the Group
from your yacht in the Carribean. And rememnber who it was who put
you on the right lines.
michael adams
...
Les Corbett
"michael adams" <mjad...@oetel.net.uk> wrote in message
news:2o4183F...@uni-berlin.de...
Thank god for this ng. I have just flogged all my Ladbrokes and Corals
shares.
But...joking apart...I wonder????
LC
michael adams
"Les Corbett" <l.co...@btinternet.com> wrote in message
news:cfil9a$i60$1...@hercules.btinternet.com...
There was a time a while back there Les, where long odds-on
shots were showing a good return over quite an extended
period. This may have been more than once, I can't exactly
remember off hand. Maybe there have been such periods
over the decades even.
In terms of the market, long odds-on shots should give
punters the best returns, simply because the bookies
need to shave any margin to the bone and make them as
"attractive" as possible if they want to lay them at
all, in the face of stiff resistence. Which they'll need
to do unless they're going to bet without the favourite
at all. And the stats do tend to confirm this.
michael adams
michael adams
"Peter C" <peter...@europe.com> wrote in message
news:2o4l7iF...@uni-berlin.de...
> x-no-archive: yes
> You would really need an answer from Smartsig or the Smartsig
> website. The bozos on this ng generally don't believe in systems
> or in staking plans.
Whereas all the "experts" on the Smartsig website, spend good money on
race data, which they then need to justify in some way.
Anyone for GIGO ?
And yet while they're all winning so much money with these systems of
theirs, they still spare the time to keep posting messages on their
website, telling each other how to win even more !
Yup! Sure makes sense to me too !
Yet another gem there from Mr C I see.
michael adams
Les Corbett
> news:2o3uq8F...@uni-berlin.de...
> You would really need an answer from Smartsig or the Smartsig website. The
> bozos on this ng generally don't believe in systems or in staking plans.
But
> your system is making a 20pt profit to level stakes per 100 bets and any
> proven staking plan should improve that. Although, with an 80% strike
rate,
> it might simpler just to sell your granny and wack up the level stakes.
I am probably one of those sceptical bozos but, in truth, I have always been
open minded about staking plans just in case someone could explain the
logic which I miss. I have asked the question a few times here but never had
a reply which made much sense....to me at least. There seemed to be much
retrospective fitting but nothing to improve on the simple method of even
stakes
and increasing when the bank reaches a certain profit. But I have to admit
to
a suspicion of sytems which suggest that history shows the way.
Les
Derek F.
the "most tipped horse of the day" The staking system advocated for it was
to increase stakes by one point after each loser. By the end of the season
the stakes were usually up to 100 points.
A dangerous staking system I used to use but not for such short priced
selections as yours was increasing stakes after each losing bet to recoup
previous losses in the sequence 1, 3, 5, 10, 20, 40 and staring a new
sequence after a winner Or six losers:-) I started it for the first time
one June and did not have a losing sequence until the last two weeks in the
season which wiped out the seasons profit.
Derek.
Ren
to
> increase stakes after a loser in order to recoup losses and make a profit.
> But when you are expecting more winners than losers, as the OP claims,
then
> it is often customary, although perhaps counter-intuitive, to reduce
stakes
> after a winner ( as in the Fibonacci plan ) because you are expecting
> several winners in a row and then a loser. Since the OP, with a 80% strike
> rate, is expecting a winner nto his best stake should go there so it also
> probably conforms to the Kelly Principle ( bet your greatest stake where
> your advantage is greatest ).
> The OP quotes a winning run of 23 which at his average odds of 1/2 would
> give +11.5 pts profit to level stakes. Using the reversed Fibonacci
> sequence of stakes ( reducing after a winner and starting again after a
> loser or at the end of 5 bets ) of 8,5,3,2,1 the OP would have a profit of
> +90 pts. from the 23 winner run.
> That is 4 full sequences of +19 pts each then wins of 8, 5 and 3 pts and
the
> run would come to and end on the 24th selection with a 2 pts loser.
> All theory of course - but running a system with an 80% strike rate at
level
> stakes might be quite boring. Jazzing it up and playing about with staking
> plans might add a bit of interest/fun.
> Which might be a hint for Les as to why some folks like staking plans,
> especially if they have successful methods and can't lose much by playing
> around. It must get quite boring for some people hunched over a computer
in
> the spare bedroom all day.
> And of course if his plan is a win bet with an 80% strike rate for his
> selection the OP should LAY a couple of shortest priced horses in the
races
> as well.
>
When i first got seriously interested in gambling (which doesn't mean i was
lumping on, i wasn't) i used a staking plan that worked on prices. The
higher the odds the bigger the bet, the lower the odds the smaller the bet.
Often there were lots of really low priced losers but you would strike
pretty decent with the mid range 3/1 - 6/1 and would get the occasional long
shot (NC Owen a tip from a Cumani stable lad who i met at Cheltenham) would
make the season easily, of course they didn't always come around but a
decent profit was being made from my low outlay wagers at the time.
Obviously as time progressed and work and college progressed the time to
study the form to produce the bets wained and i moved onto other methods.
Ren
michael adams
>
> When you are expecting more losers than winners human nature
> leads people to increase stakes after a loser in order to recoup
> losses and make a profit.
> But when you are expecting more winners than losers, as the OP claims,
> then it is often customary, although perhaps counter-intuitive, to
> reduce stakes after a winner ( as in the Fibonacci plan ) because you
> are expectin several winners in a row and then a loser. Since the OP,
> with a 80% strike rate, is expecting a winner nto his best stake should
> go there so it also
> probably conforms to the Kelly Principle ( bet your greatest stake where
> your advantage is greatest ).
> The OP quotes a winning run of 23 which at his average odds of 1/2
> would give +11.5 pts profit to level stakes. Using the reversed
Fibonacci
> sequence of stakes ( reducing after a winner and starting again after a
> loser or at the end of 5 bets ) of 8,5,3,2,1 the OP would have a profit
of
> +90 pts. from the 23 winner run.
> That is 4 full sequences of +19 pts each then wins of 8, 5 and 3 pts and
the
> run would come to and end on the 24th selection with a 2 pts loser.
> All theory of course - but running a system with an 80% strike rate
> at level stakes might be quite boring.
> Jazzing it up and playing about with staking plans might add a bit of
> interest/fun. Which might be a hint for Les as to why some folks like
> staking plans, especially if they have successful methods and can't
> lose much by playing around. It must get quite boring for some people
> hunched over a computer in the spare bedroom all day.
> And of course if his plan is a win bet with an 80% strike rate for his
> selection the OP should LAY a couple of shortest priced horses in the
> races as well.
Either you're a troll or you're an idiot.
The choice is yours.
With an 80% strike rate the stake to bank ratio can be
set pretty low. i.e such that the probability of doubling the
bank before there's any possibility of busting the bank is about
9.99. It isn't that hard to work this ratio out.
The stake to bank to ratio is -
the stake is set at x% of the bank.
It's this stake to bank ratio that matters. As it's this which decides
whether you're likely to go bust or not. Fiddling around with the
size of the stake simply makes this more difficult if not impossible
to calculate and set with any precision.
Hence the point of sticking to level stakes.
Depending on the frequency of the bets, with a claimed 20%
return and a very low staking ratio, it should be posssible
to double the bank every week or every fortnight.
So that starting with a £1000 bank you'd be a millionaire either
within 10 weeks or twenty weeks. And a bilionaire within another
10 or twenty.
You Corbally, are a typical sufferer from Systematic Delusions.
The comparative success or failure of any of the staking
plans outlined above by Corbally, as compared with backing
the same horses any level stakes, depends crucially on the
order or sequence in which the wins and losses occur, and the
prices of those winners and losers. Which nobody can predict
beforehand of course.
In reality, stakes decided by staking plans could just as well
be being picked out of a hat. But then in the long run - over a
billion or more bets say - bets set at random within limits should
show exactly the same return as do level stakes. [Providing the
bank was robust enough that is ] However in the short term, there
can be a far greater likelihood of busting the bank.
This is also the real reason* why doubles etc are such a bad thing
as they break the principle of level staking. A bad thing that is if
you're serious, and not simply betting mainly for the excitement etc.
Other reasons* given are lack of real opportunity - which would be
true enough if you're trying to find eight selections for mixed
doubles and trebles on a Saturday - although not necessarily two -
and increased risk which is only true in the above sense. But not
in the sense usually intended. For people who simply have the one
Saturday bet, the risk return ratio on multiples is exactlly the
same as with singles. Just the returns are more spaced out.
michael adams
...
gordon.carr1
I'm having great difficulty trying to put this into words but basically, you
have to get 3 in a row before you collect any money. Once you have the first
three, every subsequent winner gives you a treble.
Bet 1: 1 point A
Bet 2: Returns from Bet 1 on to B
Bet 3: 1 point B
Bet 4: Returns from Bet 2 on to C (collect winnings)
Bet 5: Returns from Bet 3 on to C
Bet 6: 1 point C
etc.
I t sounds more complicated than it is. I wrote a spreadsheet in Excel to
make it easier to sort the stakes out.
Gordon
Michael Russell
to grow your winnings is to use BB +SR. The originator was some American
engineer, I think, but it didn't catch on in the States because they don't
know what square root is .... :-)
BB = Base bet; say 100. SR represnts the square root of your winnings
to-date. If you've won 400 to-date, then the bet's 100 + 20.
What your base bet is as a proprtion of your bank (do an Archie to predict
LLR, should help) is a separate decision. When you revise the base bet is
another question ....
Michael
"dave" <t...@thegroup.com> wrote in message
news:2o3uq8F...@uni-berlin.de...
Les Corbett
"gordon.carr1" <gordon...@ntlworld.com> wrote in message
news:QflTc.538$TS....@newsfe3-gui.ntli.net...
Yes, it isn't difficult,but it is typical as another post from a staking
plan
affecionado was. If x then do y. If y then increase/decrease until z.
But the question remains unanswered.....Why?
Les
michael adams
"Michael Russell" <michael...@spamex.dsl.pipex.com> wrote in message
news:411ddce7$0$20253$cc9e...@news-text.dial.pipex.com...
> Well, one of the denigrated Smartsig group demonstrated that
> the fastest way to grow your winnings is to use BB +SR. The
> originator was some American engineer, I think, but it didn't
> catch on in the States because they don't
> know what square root is .... :-)
His name was John L. Kelly, he was an Information Systems
Engineer and he published his paper, primarily concerning
information loss during data transmisson in the Bell Technical
Journal for July 1947. The Kelly System is fully explained on
p.72 of "The Theory of Gambling and Statistical Logic".
Epstein 1967
Kelly systems are otherwise known as proportional staking
or fractional staking systems, their common feature being that
the stake is constantly being adjusted in proportiion to
the current bank.
Kelly type systems and algorithms were first studied over
30 years ago and nothing has changed in the interim. Their
performance is primarly dependant on the rate of return,
which needs to be over 60% min. which in combination with
the strike rate will determine the average price of the
winners. Otherwise they can actually perform worse than
level staking.
...
>
> BB = Base bet; say 100. SR represnts the square root of
> your winnings to-date. If you've won 400 to-date, then the
bet's 100 + 20.
>
> What your base bet is as a proprtion of your bank (do an
> Archie to predict LLR, should help) is a separate decision.
> When you revise the base bet is another question ....
....
In any Kelly system...after every bet.
This is all pointless tinkering, if there's no substantial
level stake return to start with.
The only advantage Kelly Systems have over any others is
that its impossible in theory to bust the bank. Providing that
is, you can find a bookie who'll accpet bets of for example
0.01 of a penny.
If you don't believe any of this then write yourself a few
million monte carlo simulations. Or ask this Smatsig "expert"
to do it for you. And then prepare to start wiping the egg off
your face.
>
> Michael
michael adams
John L
"michael adams" <mjad...@oetel.net.uk> wrote in message news:2o6ivjF...@uni-berlin.de...
>
> "Michael Russell" <michael...@spamex.dsl.pipex.com> wrote in message
> news:411ddce7$0$20253$cc9e...@news-text.dial.pipex.com...
>
>
> > Well, one of the denigrated Smartsig group demonstrated that
> > the fastest way to grow your winnings is to use BB +SR. The
> > originator was some American engineer, I think, but it didn't
> > catch on in the States because they don't
> > know what square root is .... :-)
>
>
>
> His name was John L. Kelly, he was an Information Systems
> Engineer and he published his paper, primarily concerning
> information loss during data transmisson in the Bell Technical
> Journal for July 1947. The Kelly System is fully explained on
> p.72 of "The Theory of Gambling and Statistical Logic".
> Epstein 1967
>
BB+SR and Kelly are different things. Whether Kelly as practised
has much to do with Kelly as originally proposed is another
question. I suspect as many American as British punters know
what square root means and that mroe have heard of Kelly.
--
John.
John L
"michael adams" <mjad...@oetel.net.uk> wrote in message news:2o5qtaF...@uni-berlin.de...
>
>
> This is also the real reason* why doubles etc are such a bad thing
The great advantage of multiple bets used to be that you paid
tax only once. Now, unless you need to lock in a price that
might disappear after the first race, there seems little point.
--
John.
michael adams
"John L" <j...@lammtarra.fslife.co.uk> wrote in message
news:cfnhgg$a6l$1...@news5.svr.pol.co.uk...
You missed a trick there nevertheless. The paper was published
in 1956 not in 1947 as I stated.
1947 was Von Neumann and Games Theory.
Anyway non of it stands up to a few million simulations,
now matter how many variables you fiddle with.
Aren't computers wonderful?
michael adams
michael adams
>
> "michael adams" <mjad...@oetel.net.uk> wrote in message
news:2o5qtaF...@uni-berlin.de...
> >
> >
> > This is also the real reason* why doubles etc are such a bad thing
>
> The great advantage of multiple bets used to be that you paid
> tax only once.
Good point. And one the Treasury never got around to addressing.
> Now, unless you need to lock in a price that
> might disappear after the first race, there seems little point.
Given that they really were separate bets, with the winnings
from each leg forming part of the betting bank in between bets -
although untouchable and held by the bookie. So that the backer
had no real control of the size of the stake in the subsequent
legs which thus contravenes the principle of level staking.
Which with an infinite bank wouldn't matter in the the long
term over millions of bets. But in the short time might.
Actually that's another big disadvantage you put your finger
on there, that I overlooked. The ability to choose to bet at
a particular price in "running" muliples*. Unlike say fixed
multiples such as the Spring Double etc.
* At least on the assumption this is indeed always a disadvantage.
Which is arguably not always the case for those with some kind of
edge on the market.
michael adams
...
>
> --
> John.
>
>
Good point.
The reason they
Les Corbett
Maths theory gone mad. As far as betting is concerned, square
root is a rather unusual potato.
Les
michael adams
>
>
> Maths theory gone mad. As far as betting is concerned, square
> root is a rather unusual potato.
>
> Les
>
Specially bred for making chips I believe Les.
michael adams
...
dave
many thanks for all replies
this is actually a football system that i'm developing. the task now is to
check it against the past 5 seasons of data which i've now managed to obtain
:-)
let's hope it's not just a lucky patch - i'll know for sure within the next
few days, one way or the other
if profitable, i'll post the system after i've doubled my bank 5 or 10 times
;-)
sorry thats it's OT but i only lurk in this NG and knew i'd get a good
response
thanks again