Accumulators superior to singles

[Shy10ck]

Re: Accumulators superior to singles So it is really, a match between variance and value. And nothing, at all to do with singles vs accus, Let me ask in another way. Which single would you prefer? Odds 2,2 and a 50% winning chance (lower variance, lower value) or Odds 4,84 and a 25% winning chance (higher variance, higher value)

[moggis]

Re: Accumulators superior to singles well Shy10ck,I dont think theres much doubt that 90% will answer the second one. but let me give you a quote from a book called "secrets of successful betting",by michael adams, "in effect all multiple betting can be seen as some kind of crazy system ,in which all the most important questions are being determined on a more-or-less random basis. And while this may be an ideal arrangement when the main objectives are entertainment,and the excitement which can be generated by such uncertainty,in any other circumstances such deficiencies are difficult to overlook."

[Machine]

Re: Accumulators superior to singles

odds 4,84 and a 25% winning chance (higher variance, higher value)

If your betting as you say...Doubles 1&2..3&4...5&6..7&8 Where are you getting 25% winning chance from ? For a start your only playing 4 possible doubles from 28 That gives you a 14.28% (4/28) of catching a winning double not a 25% chance. If you have python on your comp place this code in its IDLE and run

import random
print 'assuming 4 correct from 8'
lines = int (raw_input('Enter number of attempts ,' ) )
for i in range(lines):
    x = (random.sample(range(1,9),4))
    print '%9s%3d%3s%3d%12s%3d%3s%3d' % ('double A',x[0],'&',x[1],'double B',x[2],'&',x[3])

raw_input('Press enter to close', )

You will get a random doubles output like this IDLE 2.6 ==== No Subprocess ==== >>> assuming 4 correct from 8 Enter number of attempts ,12 double A 2 & 3 double B 1 & 8 double A 3 & 6 double B 8 & 2 double A 3 & 2 double B 6 & 5

[Shy10ck]

Re: Accumulators superior to singles You are complicating it. A double with 2 fifty percent games, will have a win rate of 25%. 2, 3, 4 or 5 doubles dont change that.

[moggis]

Re: Accumulators superior to singles you know I have just had a 15 minute conversation with the most successful gambler I know(apart from tony ansell,who I dont think wouldve appreciated me bothering him),and I am STILL utterly confused. And I think Shy10ck will be as well. One thing that I feel confident to point out though is that if monkeys nest is right......then one of the most successful soccer betting books ever written.....must be seriously flawed!

[Machine]

Re: Accumulators superior to singles Im changing it am I Well lets see what you said then Eh........

I dont agree. 8 matches, 2,2 odds and 50% winning chance, stake = 100 pr match 8 singles = 8 x 100 x 2,2 = 1760. You win 50% = 880, = 80 profit (880-800) 4 doubles = 4 x 200 x 2,2 x 2,2 = 3872. You win 25% = 968, = 168 profit (968-800) Betting the same 8 matches, you gain a profit 88 if you bet them as doubles.

I am not talking permutation. I think that is where you go "wrong". 8 singles, 1,2,3,4,5,6,7 and 8. 4 doubles, (1,2), (3,4),(4,5) and (7,8)

You are right Froment' date=' [/quote'] Every time your proved wrong you change your tune. Have fun losing your money.

[Shy10ck]

Re: Accumulators superior to singles

Im changing it am I Well lets see what you said then Eh........ Every time your proved wrong you change your tune your like a little fckin schoolkid. Moron of the highest order. Have fun losing your money.

Tired of punters lounge? I change nothing and you proved nothing.

[Machine]

Re: Accumulators superior to singles :rollin :rollin :rollin Ive proved you haven't got a clue what your talking about . Or are you going to deny you ever typed those statements Was your account hacked by someone. Why should I be tired of punters lounge Oh no, you going to get me barred. Good job I learned how to back 4 doubles from 8 and win 25% of the time before my swords snapped in two and all my buttons cut off Adios

[moggis]

Re: Accumulators superior to singles Shy10ck,good luck with this thread.I think this will be my last post on it and I genuinely want to help.. So,I think it would be a good idea if you were to read the chapter in the book by michael adams from which I quoted. After that,I am 90% sure that it really is all about your personal choice and what you are trying to achieve from gambling. .And your own view of risk... anyway,good luck (I cant help wondering what monkeys nest would make of what I have been doing for the last 13 years!:unsure?)

[Machine]

Re: Accumulators superior to singles Keep up the sales pitch Muggis Keep dropping a few more 'names' and Im sure you'll get plenty of takers. Dont tap on the glass though ;)

[moggis]

Re: Accumulators superior to singles monkeys nest I am grateful for the relative politeness of your post. I sincerely hope you can make the get-together I mention on the other thread..

[Shy10ck]

Re: Accumulators superior to singles

Shy10ck,good luck with this thread.I think this will be my last post on it and I genuinely want to help.. So,I think it would be a good idea if you were to read the chapter in the book by michael adams from which I quoted. After that,I am 90% sure that it really is all about your personal choice and what you are trying to achieve from gambling. .And your own view of risk... anyway,good luck (I cant help wondering what monkeys nest would make of what I have been doing for the last 13 years!:unsure?)

Thank you moggis. I will have a look at it, if I get the book in hand.

[Drawsandmore]

Re: Accumulators superior to singles

If your betting as you say...Doubles 1&2..3&4...5&6..7&8 Where are you getting 25% winning chance from ? For a start your only playing 4 possible doubles from 28 That gives you a 14.28% (4/28) of catching a winning double not a 25% chance. If you have python on your comp place this code in its IDLE and run

import random
print 'assuming 4 correct from 8'
lines = int (raw_input('Enter number of attempts ,' ) )
for i in range(lines):
    x = (random.sample(range(1,9),4))
    print '%9s%3d%3s%3d%12s%3d%3s%3d' % ('double A',x[0],'&',x[1],'double B',x[2],'&',x[3])
 
raw_input('Press enter to close', )

You will get a random doubles output like this IDLE 2.6 ==== No Subprocess ==== >>> assuming 4 correct from 8 Enter number of attempts ,12 double A 2 & 3 double B 1 & 8 double A 3 & 6 double B 8 & 2 double A 3 & 2 double B 6 & 5 Hi monkeys nest, Can the code be changed to run the following criteria? 8 selections, permed in all possible double combinations to give 28 bets...(each number to appear in 7 bets) and then randomly generate 4 winning numbers and see what happens? To my way of thinking, this would confirm what Shy10ck has said that doubles/trebles could be the way to go...Which is truly alarming to me, as I am a confirmed singles bettor. I want to try and understand both points of view, and my maths skills are not good enough to understand your argument properly. I'm not sure that you're both comparing eggs with eggs, but I'm buggered if I can see the wood for the trees at the moment.

[Shy10ck]

Re: Accumulators superior to singles

If your betting as you say...Doubles 1&2..3&4...5&6..7&8 Where are you getting 25% winning chance from ? For a start your only playing 4 possible doubles from 28 That gives you a 14.28% (4/28) of catching a winning double not a 25% chance. If you have python on your comp place this code in its IDLE and run

import random
print 'assuming 4 correct from 8'
lines = int (raw_input('Enter number of attempts ,' ) )
for i in range(lines):
    x = (random.sample(range(1,9),4))
    print '%9s%3d%3s%3d%12s%3d%3s%3d' % ('double A',x[0],'&',x[1],'double B',x[2],'&',x[3])
 
raw_input('Press enter to close', )

You will get a random doubles output like this IDLE 2.6 ==== No Subprocess ==== >>> assuming 4 correct from 8 Enter number of attempts ,12 double A 2 & 3 double B 1 & 8 double A 3 & 6 double B 8 & 2 double A 3 & 2 double B 6 & 5 If your AI system is based on such total misunderstanding of numbers and words, then you are more lost than I thought. A double with two matches with 50% winning chance each, will always have a 25% winning chance, NO MATTER how you permute them. If you get confused over such simpel math, then I would really consider NOT to put any real money in your AI system. I would also strongly consider to behave in forums. You have an attitude problem.

[Shy10ck]

Re: Accumulators superior to singles

Hi monkeys nest, Can the code be changed to run the following criteria? 8 selections, permed in all possible double combinations to give 28 bets...(each number to appear in 7 bets) and then randomly generate 4 winning numbers and see what happens? To my way of thinking, this would confirm what Shy10ck has said that doubles/trebles could be the way to go...Which is truly alarming to me, as I am a confirmed singles bettor. I want to try and understand both points of view, and my maths skills are not good enough to understand your argument properly. I'm not sure that you're both comparing eggs with eggs, but I'm buggered if I can see the wood for the trees at the moment.

As I look at it, then there is plenty of opportunities in accus. Its a question about you are willing to accept higher variance (win less often) to gain value (win more pr bet). You dont even have to play them as fx doubles. You could bet a single, and reinvest the winning on the next bet. That way you can play doubles at Betfair. Match 1, just have to finish before match 2 starts.

[Drawsandmore]

Re: Accumulators superior to singles

As I look at it, then there is plenty of opportunities in accus. Its a question about you are willing to accept higher variance (win less often) to gain value (win more pr bet). You dont even have to play them as fx doubles. You could bet a single, and reinvest the winning on the next bet. That way you can play doubles at Betfair. Match 1, just have to finish before match 2 starts.

Hi Shy10ck, at the moment, I am with you on this. (theoretically at least). Practically though, most of my bets are on the same day and at the same time, so I don't really have the opportunity to invest any winnings from the first match onto the second match. Another issue for me is that the best odds in any two matches are not always available from the same bookmaker, and psychologically, I really need to feel that I am achieving the very best odds I can on each match. Finally, given that at best, I would only expect to predict one match in three correctly, the variance would be just too great for me to handle...I'd need a much larger bankroll :sad For the time being, I suspect that singles will remain the most profitable way of betting for me, nevertheless, I am grateful to you, monkeys nest, and all the other contributors to this thread. I think I've learned something from it, and isn't that what this forum is all about?

[froment]

Re: Accumulators superior to singles Blah, pitty to see what this thread has become; it was my favourite reading yesterday evening. ;) Shy10ck, I'm afraid I disagree with you. I said in my first post in this thread that your math looked odd to me, but I couldn't figure it out myself; I just felt it's not possible 12fold to be that more profitable than 12 singles. Monkeys Nest has explained it thoroughly, in both simple words and through math.

A double with two matches with 50% winning chance each, will always have a 25% winning chance, NO MATTER how you permute them.

This is where you are wrong. In fact, the approach is wrong - your statement is correct, but there is no way you can apply it in reality. Let me try to explain: if you place singles, there is indeed no matter in which order you place them, each winning one brings profit. However, with doubles, it very, very matters how winning picks are distributed! Overall, there will be 25% of winning doubles, but problem is that in reality you can choose ONE SINGLE combination of doubles, and there is no guarantee that it will be winning double, it's pure luck. Let's say, you intend to bet on 8 matches this evening, A, B, C, D, E, F, G and H. Nevermind if they start at the same time, or they are consecutive, as you say. How will you combine them in doubles? You can combine this way: AB CD EF GH And you can also combine this way: AC BD EF GH After hour and a half of running up and down the pitch, you find out that winning matches were, let's say, A, B, F and G. If you placed your bets as in line 1, you indeed have 25% strike rate; if you placed line 2, you lost them all, despite strike rate being 50%! Look at post #12 in this thread, that's what Monekys Nest warned you of!

What if all your doubles are win x lose you still are getting your 50% sr but not as doubles.... POP bank gone.

Is it possible that you will place line 2, and not line 1 each and every time? Yes, it is - you get correct 50% of matches, yet there is no any guarantee that you will ever place winning combo! If you place consecutive bets, one after another, is there any guarantee that your streak won't look like this: Won - Lost - Won - Lost - Won - Lost - Won - Lost - Won - Lost... and so on, untill infinity? No, there's no guarantee, it's unlikely, but possible! And then again, you hit 50 % of the matches correct, yet you haven't completed any winning double! Of course, there is a way to guarantee some winning doubles: you would have to bet on them all! And that's topic Monkey Nest covered in post #15:

Personally I thought that there were 28 possible doubles in 8 singles (I may be wrong but I dont think so) Wouldnt that make it 28.57 (800/28) per double (not 200) 4 correct would give you 6 doubles 2.2*2.2*28.57*6 = 829.67 - stake 800 = 29.67 Singles = 80 profit Doubles = 29.67 profit

Indeed, there is 8 over 2 = 8 * 7 / 2 * 1 = 28 doubles, out of which 4 over 2 = 4 * 3 / 2 * 1 = 6 doubles are winning; you can choose only 4 out of those 6, so your probability of hitting a winning double is ( 4 / 6 ) * ( 6 / 28 ) = 14.28%... again as explained in post #33! And note, it's only in case you back them all, all 28 doubles; in case you continue with random doubles, there is no probability - you may hit winning one each fourth time... but you may pass a thousand doubles without any winning one!

[Machine]

Re: Accumulators superior to singles Draws I can do anything I want with the code ,it was only thrown together in a few mins when I was bored this morning. But Ill leave Shylock to educate you in the finer points as he seems to be the 'expert' in these matters,and I wouldnt want my 'attitude problem' getting in the way. :lol Nice post Froment He'll still deny it though You can lead a horse to water etc etc ;)

[Drawsandmore]

Re: Accumulators superior to singles

Draws I can do anything I want with the code ,it was only thrown together in a few mins when I was bored this morning. But Ill leave Shylock to educate you in the finer points as he seems to be the 'expert' in these matters,and I wouldnt want my 'attitude problem' getting in the way. :lol Nice post Froment He'll still deny it though You can lead a horse to water etc etc ;)

Whoa! At no point have I mentioned anything about an "attitude problem" from you! I asked politely for an explanation that I could understand. As already stated, I'm a confirmed singles backer...Y'know, the argument you're actually proposing!! Unfortunately, I'm not as good with maths as you undoubtedly are, and whilst I could understand Shy10cks reasoning, I couldn't understand yours...Not a criticism of you or your argument, the problem is mine...I just asked if you could help me to understand. Now I've read Froments response, it is somewhat clearer. But, while I fully accept that, given a 50% strike rate a sequence of W,L,W,L,W,L,W,L ad infinitum is possible no matter how unlikely, you must surely accept that W,W,W,W,W,W etc for the first 50% of your matches and L,L,L,L,L,L,L etc for the remainder is also possible albeit just as unlikely...Giving you a 50% strike rate with your doubles. So, if you can write something that can run every single permutation of winners and losers (with a 50% strike rate) and then run every possible outcome, what happens? I'll keep the numbers manageable for me to type! four matches, 2 doubles, double A+B and double C+D...no perms...allocating random winners. A B C D W.L.W.L. = 100% losing bets L,W,L,W, = 100% losing bets W,W,L,L, = 50% winning bets W,L,L,W = 100% losing bets L,W,W,L, = 50% winning bets L,L,W,W, = 50% winning bets The four 1 point single bets return 4.40 points in all 6 scenarios. The two 2 point double bets return 9.68 points on three occasions. As everything, (apart from the strike rate) is random, the average return from the doubles is 4.84 points. The result would be the same if my doubles were AD & BC or AC & BD...Where am I going wrong?

[Machine]

Re: Accumulators superior to singles line 5 is not a winning double as your only backing A+B and C+D line 3 and 6 are the 2 double lines line 5 is equal to B+C ;)

[slapdash]

Re: Accumulators superior to singles Let's try to keep this civil, guys.

You are right that the variance is higher for multiples' date=' but please show me how you will increase your stakes on singles and have the same profit as for multiples.[/quote'] Let's take the example you gave in the first post. Two betting opportunities, each with a 50% chance and odds of 2.2. You looked at two strategies: (A) Two 100 unit singles. This gives an expected profit of 20 units. (B) One 200 unit double. This gives an expected profit of 42 units. Certainly (B) gives a higher expected profit than (A), which is good. But equally certainly it has a higher risk/variance, which is bad. At this point you could just say that one strategy has better profit, and the other has better (i.e., lower) risk, so it's just a matter of individual preference, and how risk-averse you are, which is better. But let's look at a third strategy: © Two 210 unit singles. This gives an expected profit of 42 units. So (B) and © have exactly the same expected profit. But, even though your total outlay is higher with ©, it's actually less risky in the medium or long term, for any reasonable measure of risk. Let's look at three different measures. (i) Risk of ruin. Suppose the same scenario occurs repeatedly, and each time you choose the same strategy, either (B) or ©. And let's suppose you start with a bank of 4200 units (the exact size doesn't matter much, but I've chosen a figure which is an exact multiple of the stake for both strategies, just for simplicity). With strategy (B), you have about an 11.1% chance of going broke, but with © only about a 3.6% chance. (ii) Variance. Both strategies have an expected profit of 42 units per go. But the profit from (B) has a standard deviation of about 419 units, whereas the profit from © has a standard deviation of only about 297 units. (iii) Kelly staking. Rather than fixing the stake, let's look at what the Kelly criterion would suggest, given that you've already decided whether you're going to do two singles or one double. The Kelly stake for singles with these odds and probabilities is about 8.33% of your bank (this should be reduced very slightly if you're placing the bets simultaneously), and then your expected profit would be about 1.67% of your bank for the two singles together. The Kelly stake for the double is about 5.47% of your bank, and then your expected profit would be about 1.15% of your bank for the double. So © is less risky than (B), but gives the same expected profit, so is "better" in any reasonable sense.

[slapdash]

Re: Accumulators superior to singles

Paul Steele wrote the classic soccer betting book"profitable football betting".......... and he advocates trebles,4-way,5-way and even the occaisional 6-way bet!

In the early days of fixed odds football betting coupons, I believe that it was normal for bookies to take trebles as minimum, and even then only if you included no home wins? I think you normally had to do at least five-folds (or something like that) if you included home wins? I may be misremembering the exact conditions. I haven't read Steele's book, but I know it's quite old, so could it be that he's doing trebles, etc, because that's what he had to do in those days? Or does he actually write about the choice between singles and multiples?

[staffy]

Re: Accumulators superior to singles I am guessing a good punter over a long period would have a similay yield if he was to be on either singles or multiples. Any punter can get lucky once in a while I guess and win big but how often really does it happen.

[Drawsandmore]

Re: Accumulators superior to singles

line 5 is not a winning double as your only backing A+B and C+D line 3 and 6 are the 2 double lines line 5 is equal to B+C ;)

Doh! Thank you MN :notworthy

[moggis]

Re: Accumulators superior to singles re paul steeles book I have the 2004 edition of the book in my hand. on page 15 he mentions the time you could only do 5 ways. in the last chapter however he is doing multiples yet again. just one thing,which is quite important-he actually does perms with 1 or 2 bankers

[Machine]

Re: Accumulators superior to singles Excellently put Slapdash I believe from posts Ive read that your the maths Guru round these parts. Is there a simplified version of the RoR formula. Ive found various ones on trading sites but Im finding them hard to code in a project Im doing. If there is a 'general purpose' one, could/would it be possible for you post it up please Cheers :hope

[moggis]

Re: Accumulators superior to singles on page 139 he talks about 3-match perms.....and I quote... "the lowest no. of matches I cover in any week is 3.it is,however,rare that I would go for a 3-team option........."

[moggis]

Re: Accumulators superior to singles HOWEVER,on page 18 it does indeed appear that he advocates singles after all!!!!!!!!!!

[moggis]

Re: Accumulators superior to singles why then would he not choose 3-ways over 4 or 5-ways!!!! I give up!

[slapdash]

Re: Accumulators superior to singles

Is there a simplified version of the RoR formula. Ive found various ones on trading sites but Im finding them hard to code in a project Im doing. If there is a 'general purpose' one, could/would it be possible for you post it up please Cheers :hope

I think I made a fairly detailed post about this some time ago, in response to somebody else's question. I'll try to find it. I doubt it's any simpler than what you've found elsewhere, though.