
Premium Member
my betfair account is down 6 figures over its lifetime, so have "won" a fortune at bookies and dont have many uk books left
i underlay highish odds a little but not by much, if i have £100 at 5/4.4 i may lay 106 and move on so id be +2/+40ish
and back when i had interwetten i fully underlaid all the monster arbs on offer
as the odds were so far off that even mugging youd be unlucky not to make money
i know the argument made in here is true
but i also know for a fact that i cant handle losses and couldnt deal with losing months and also couldnt mug something i didnt fancy even if the odds were value
if i made 100 a day mondaysaturday then lost 100 on the sunday id be in a foul mood even tho id be 500 up for the week
gambling isnt for me
willhill vernon laddies betfair casino/arcade freebets are more than enough gambling excitement for me
blowing bubbles has been mugging for a long time iirc so there must be something in it
good luck to anyone who tries it
if you have red mist tendancies then thinnk twice about it before committing

Post Thanks / Like

Premium Member
anyone giving it a go?
if so do you fancy sharing your staking plan...

Premium Member
I've been thinking about doing this for a while now, I already pretty much fully underlay everything but one of the main reasons for wanting to do this is time, I spend far too much time filling in spreadsheets and checking where the money ended up and if it is correct etc.
I'm pretty anal about recording every single bet I've done so it's easy to see what difference there is, I would have been better off by £54k if I hadn't hedged my bets, that's over 7 years and 18000 bets so a decent sample size for variance to even out.
I'll have to see how I can stomach the variance but hopefully shouldn't be too bad, I used to tilt pretty bad at casinos but that was just the emotion of getting bad beats on blackjack, with sports betting I think it should be easier to take all the emotion out of it and just look at it as numbers.

Post Thanks / Like

Premium Member
its tricky to see what kind of swings to expect and how much youd need to stake to be better off
at present il lump on as much as i can on any arb i come across rather than a fixed 500@evens 250@2/1 etc
also a lot of bets are qualifiers for promos that arent necessarily arbs should they be mugged or hedged...
also a lot of the freebets currently placed will be on something like a 10/10.5 and wouldnt be bet at all if mugging
also how many percent above betfair in order to lump on
say you were currently getting 1k on a 2.1/2 arb andmaking 25 or so regardless of outcome, if mugging would need to lump on 500 to have same expected return over a number of bets
it would be very difficut for me to stick a wad of cash on something i didnt fancy just because price looked to be value
think if i tried it i would basically be a gambler who lumped on at best prices

If you cannot calculate the risk of not hedging on the long run, and understand the results of your calculations, then hedge your bets.
(skip to the examples if you hate Maths  you shouldn't, if you're willing to play against bookies who know their Maths. If you're even lazier, go directly to the spreadsheet, that you can download)
Maths:
Calculate the Expected Value of not hedging after N bets (which will depend on the commission), then also calculate the Variance and the Standard deviation. Then, calculate the Zvalue: Z = EV / std_dev. This value will tell you what are your chances to lose more money than you gain, by not hedging.
The higher Z is, the safer it is to not hedge: with EV (expected gain of not hedging, over a hedging strategy) very superior to std_dev, you have a high probability to gain more without hedging (but still a probability to lose it all: it's not riskfree anymore). With a low Z, you are taking a big risk on the long term.
For high odds, Z is low (danger!). For low odds, Z is higher, so you might want to not hedge.
It depends what kind of risk you're willing to take...which also depends on your bankroll.
Let me introduce you to the notion of long term maximal risk.
With N (number of similar bets placed) getting higher,
EV will increase: EV(N) = EV(1) * N
std_dev will increase: std_dev(N) = std_dev(1) * sqrt(N)
But they don't increase at the same speed: std_dev increases more slowly.
If we plot them against N, the EV is like a diagonal line that goes up, and the Std_Dev is like a "cone" that gets wider and wider around the EV line. Do you have a mental picture of this?
Let us follow the lower part of the cone.
The Ztable tells us that we have 15.87% chances to end up under the EV  std_dev line (Z=1 in the Ztable), and 84.13% chances to end up over this line. It also tells us that we have 1% chances to end up under the EV  2.33 x std_dev line (Z = 2.33 in the Ztable).
If we're interested in what we have 1% chances to lose, we just have to follow this line. Initially it goes in the negatives, and it keeps on going down: as we're placing more and more bets, we're risking more and more money. But remember the diagonal and the cone: EV increases faster than std_dev decreases, so at some point, our "1% risk line" reaches a minimum, and starts to increase. That's what i call the "maximal risk", and you can calculate it.
On the long run, you have 1% chances to end up below this number. In the following examples, the values that i'm giving are this "maximal risk", so you can get an idea of the risk you are taking.
Example: Let's say i place bets of £50.
And i'm interested in checking how much money i have 0.1% chances to lose.
For an odd of 1.2,
after 992 bets (the risk is lower for either less or more bets),
the expected profit (50% chances to gain this or more) over a year is £652
i have 3.29% chances to lose more than i have gained
and 0.1% chances to lose £431 or more (i could possibly lose much more)
For the remaining, i rounded numbers to make it easier to read:
Odd: 1.5
after 1300 bets
x% chances to gain under £y :
50%  £1,300
5%  £0
0.1%   £1,100
(0.1% chances to gain under  £1100, also means: 0.1% chances to lose £1100 or more, compared to the hedgebetting scenario)
And you see already that the risk is almost as big as the "expected" reward.
Odd: 2
after 1700 bets
x% chances to gain under £y :
50%  £1,900
8%  £0
0.1%   £2,200
And the risk is now bigger than the reward (considering i accept to take a 0.1% risk). I wouldn't take this risk personally. If you have a huge bankroll, you might consider it.
Odd: 5
after 4400 bets
50%  £1,900
19%  £0
0.1%   £8,900
(The risk here is definitely too high for me)
Odd: 10
after 9000 bets
50%  £3,400
27%  £0
0.1%   £20,000
I wouldn't do an offer that contains some risk, without having done these calculation first. I started to make a program that would calculate and plot these, but didn't finish it, and now i'm onto a new (not betting related) project. So i might just never finish it.
But i made a mugpunting spreadsheet, and here it is:
https://dl.dropboxusercontent.com/u/...alculation.xls
you can check it with a website such as https://www.virustotal.com/uk/#url
So, you can change the values and see what happens. You can find a Zvalue table on Wikipedia, to get the risk estimate.
Conclusion: always hedge bets with highish odds! For low odds it's up to you. Is it worth taking a possibly high risk, in order to save just a little bit?
EDIT: check out blowingbubbles' posts below, and my answers, for a more interesting and practical Conclusion.
Conclusion  extended: if you're lazy and don't like to calculate how much to lay exactly (that's a pretty good reason for not laying a bet), you can do it like me: for low odds, just lay approximately. A few pounds more or less will even out over time (as long as it's on low odds!). For a £50 bet at odds under 3, i always lay it at £48, and that does the trick. Tiny risk, comfortable laziness I still pick a back odd which is close to Betfair's lay odd, though, as it's the best estimate for the outcome's probability that i have.
PS: One important thing to notice, is that these calculations don't take into account the fact that you could run out of money fast, and not be able to continue. So consider these estimates as optimistic: you are more likely to lose it all, than the results i've shown here!
I have assumed that back and lay odds were the same, and correspond to the probability of the outcome occurring. If i change these variables (eg: for a value bet), the risk doesn't change very much: it mostly depends on the odd: high odd = high risk; low odd = lower risk, but still a risk.
Also, i made the approximation that the probability distributions were normal, while they are skewed. I plotted it and the approximation was "good enough".
PPS: i might draw pictures to illustrate EV, std_dev, Z, the "risk lines" for various values of Z, and the "maximal risk".
Last edited by pamjou; 02/09/2013 at 21:26.
Reason: made the explanations clearer

Post Thanks / Like

I can attest to pamjou's entry, I kind of asked him to post his stuff, I know him from email conversations which have been very interesting re the mathematical basis for various endeavours relating to advantage play.
On the subject of mugging vs heding though, I know I found it quite interesting seeing what the actual maths behind value betting actually was from pamjou above, I know for a long time people had said 'why bother laying stuff off' and generally when I would look at the figures for my own betting, if I'd not laid anything off then I would have been 'up' a considerable amount over all the bets I'd placed (plus I would have had a lot less stress with not having to worry about bets being palped, saved a lot of time by not having to record the details of lay bets, and generally in a sense made my gubbed accounts 'more worthwhile'  though don't ask me to qualify that statement, I'm just getting at the fact that IF a bookie limits you or closes your account, then at least you know your account is (usually) profitable).
But it is still a leap of faith  even knowing the statistics  to believe that you would be better off mugging arbs. In theory, even given the maths (or perhaps because of the maths!), you could still easily come out down. Generally you'd need to have a decent bankroll to see you through the hard times of course...
In the end when I experimented with mugging value/arbs, I gave up fairly soon after a limited trial (during which I actually ended up down!). Partly I felt that it would be quite possible for me to start thinking that I had some kind of 'knack' at picking winners IF I went on a winning spree, and so then carry on without actually bothering to look for arbs any more, just thinking that my 'knack' would see me through and carry on making me money. Which is obviously borderline gambling addiction territory.
Although actually another part of it was that I just couldn't bear losing and when I did try it for a while (mugging arbs), I 'felt' the losses a lot more than the wins and so even though I knew fairly strongly that over time they would even out in favour of a profit, I just couldn't be bothered with it. That and plus I just like doing the maths involved in backing/laying / hedging off back bets so you know in advance what your p&l is in the different scenarios.

Nice post pamjou though isn't it pretty obvious the higher the odds the more variance? With a correct staking plan then there is little risk, you will come out ahead eventually. If you put £500 on an evens shot then you need to adjust accordingly as the odds get higher. You are assuming in your example using the same stake size on every bet regardless of the odds. Mug everything IMO regardless of odds but be sensible, match off anything above your staking plan. Commission saved alone makes it worth it let alone time, ease and having more money available.

Post Thanks / Like

Originally Posted by blowingbubbles
you need to adjust [your stake] accordingly, as the odds get higher.
Very good point! With higher odds, if you want to mug you could choose to place a smaller stake. But then this also lower the amount of money that you're making. It's a compromise: mugging with a lower stake, or hedging for a higher return (and no risk)? Mugging can be good in some cases, but it's not good in all cases.
And sometimes, you don't get the choice of the stake: when it's a special offer for example.
Originally Posted by blowingbubbles
you will come out ahead eventually.
Even with the best chances in the world (a high EV and a high Zvalue), you could come out behind.
The question is: what risk am i willing to take?
Am i happy to take 1% chances to lose £2000 or more (can i afford this if it happens?), and 10% chances to have not made any money after 2000 bets, in order to have 50% chances to gain £2000 or more? That question can only be answered individually. And the answer to that question will tell you what is the maximal stake that you can mug for each odd.
What i'm saying is: you can make an informed choice about the risk.
With the spreadsheet that i shared you can see how much you can expect to save on the long run for a particular odd and stake, and what is the risk you are taking. It seems reasonable to aim at a similar level of risk (the Zvalue can be one good indicator), whether the odd is low or high. So, this spreadsheet can help you make a staking plan which corresponds to the risk you are willing to take.
Would you share what your staking plan is? Just out of curiosity. I understand if not
Last edited by pamjou; 01/09/2013 at 15:56.

Originally Posted by pamjou
It's a compromise: mugging with a lower stake, or hedging for a higher return (and no risk)? Mugging can be good in some cases, but it's not good in all cases.
And sometimes, you don't get the choice of the stake: when it's a special offer for example.
No compromise is needed, mug up to your staking plan then hedge the rest of the bet. Even if you are only mugging a fiver its still better then hedging the whole bet. To this day I still hedge when I'm over exposed on an event happening.
Originally Posted by pamjou
Even with the best chances in the world (a high EV and a high Zvalue), you could come out behind.
The question is: what risk am i willing to take?
Am i happy to take 1% chances to lose £2000 or more (can i afford this if it happens?), and 10% chances to have not made any money after 2000 bets, in order to have 50% chances to gain £2000 or more? That question can only be answered individually. And the answer to that question will tell you what is the maximal stake that you can mug for each odd.
What i'm saying is: you can make an informed choice about the risk.
With the spreadsheet that i shared you can see how much you can expect to save on the long run for a particular odd and stake, and what is the risk you are taking. It seems reasonable to aim at a similar level of risk (the Zvalue can be one good indicator), whether the odd is low or high. So, this spreadsheet can help you make a staking plan which corresponds to the risk you are willing to take.
Would you share what your staking plan is? Just out of curiosity. I understand if not
Yes I agree your SS is very good for setting a staking plan, I just wanted to point out don't be afraid of taking the longer odds on just make sure to have a correct staking plan in place depending on the odds you are taking.
3 years ago ish when I started mugging I figured that staking no more then 0.5% of my float should be safe and adjusted that as the odds went up using common sense mainly Never had a strict plan in place but I've been sensible and rarely ever had a big bet on at longer odds anyway. (sticking £25£50 on longer odds is never going to be a problem for me personally as I have a large float ) I still stick to the same rules and I'm well, well up despite not even making any money mugging last season!

Post Thanks / Like

Originally Posted by blowingbubbles
No compromise is needed, mug up to your staking plan then hedge the rest of the bet. Even if you are only mugging a fiver its still better then hedging the whole bet. To this day I still hedge when I'm over exposed on an event happening.
Good point again, thanks
Originally Posted by blowingbubbles
(sticking £25£50 on longer odds is never going to be a problem for me personally as I have a large float)
I have a small float, so i'm very riskaverse at the moment. If i had a huge float i would probably mug much more bets.
Similar Threads

By ThisCharmingMan in forum Player Support / Gaming Site Feedback / General Complaints
Replies: 1
Last Post: 08/12/2010, 20:57
Tags for this Thread
Posting Permissions
 You may not post new threads
 You may not post replies
 You may not post attachments
 You may not edit your posts

Forum Rules
