Kanga Posted January 2, 2006 Share Posted January 2, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Hi All Newbie Poster, :ok although I've been looking around the forum and this particular thread with much interest for a while now - I've started a spreadsheet and have even invented my own version of this sytem which is working well. The question I have is how do you stop your heart ruling your head and keep strictly to the sytem / strategy ? Today I got my wallet burnt because I followed my own home team (although not strictly to my strategy but ultra close) they lost and it cancelled out all the other winners today :wall TIA Link to comment Share on other sites More sharing options...
slapdash Posted January 3, 2006 Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) No i don't, once its up it stays up. But only because i expect a positive yield. And thats a good point as it needs to be considered in the comparison. Hmm. If it stays up, then in the (very) long run you will go broke, even with a positive yield. Each time you recalculate the stake to 10% of the current bank you will go broke if the next 10 bets lose. With the strike rate of these bets, that may only be something like a one in a billion chance each time, but eventually that one in a billion chance will happen. (Actually there's a lot more than a one in a billion chance each time, as there are lots of other ways you could go broke before the next recalculation of the stake apart from a losing streak of 10 bets). If it's not too much trouble, I'd be interested to see what happens with last season's record if you repeat the percentage staking and plateau staking calculations, but with a 20% stake instead of 10%. There is a point to this line of questioning ... this season with going back to a lower stake: +124 For this, when do you go back to a lower stake? E.g., you start with 100 points, build it up to 125 points with 10 point stakes. Then you increase the stake to 12.5 points. Do you drop back to 10 points as soon as the bank drops below 125 points, or do you wait for 100 points? Link to comment Share on other sites More sharing options...
Datapunter Posted January 3, 2006 Author Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) :welcome Kanga, how do you stop your heart ruling your head and keep strictly to the sytem / strategy ? Extremely good question :lol I had to learn the hard way. The first years of betting i was all over the place betting on just about anything. And i was successfull with a lot of things but in the end i lost. And i lost because i bet and lost big stakes on only a handfull of bets. With hindsight the main reason was that i placed bets because i felt like betting. I was bored, or overconfident, or plain being stupid. It went fine many times but mainly in the area of in-running i was caught out a few times and lost big. First breakthru was when i said to myself "This is going nowhere" and actually realised it was going nowhere. Before i always had some reason or explanaition or justification. But at some point i thought the numbers don't lie. Been doing this for quite a while and its not working. The first step, and a major one, towards making a profit was to stop betting on anything and everything. I became very selective and for quite some time placed maybe 1 bet per week and learned as much as possible. Now i can say i've recovered all my losses from the early years and i'm showing a healthy profit. All because of this system, and the effort i put in that pays off. I still bet on other things but at quite small stakes. So to answer the question: i stick to what i know works because i remember the stupid losses from the early days. The rest comes down to discipline and being committed to making a profit. And i love winning, and showing a profit for me is winning. Link to comment Share on other sites More sharing options...
Datapunter Posted January 3, 2006 Author Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) slapdash i see your point that with plateau staking and not recalculating when the bank goes down there is an increased risk of bancruptcy. Lets face it : it is loss recovery. For me the increase in risk is acceptable because: - the system is continuing to show a positive yield - the strike rate is very high so the risk of consequitive losers is low - there has not really been a downward trend ever, there have been a few setbacks but no prolongued period of negative yield. After the setbacks the system has each time returned to an upward trend. - not recalculating after the bank went down has not (yet) resulted in a stake that is too high for the bank in terms of risk. Most was last season when the highest stake/lowest bank resulted in a stake at 15% of that bank. For this system i find that acceptable, but only just. If the system in past seasons ( as shown on Joe's site ) had shown more outspoken ups and downs i would probably not have gone with plateau staking but used something else. The numbers above are based on increasing the stake when a new plateau is reached, go back when the bank goes down to the previous plateau. But i'll try a few more variations because i also think there may be quite a difference if you change the level of the plateau from 25 to 10 or 50.... Link to comment Share on other sites More sharing options...
Datapunter Posted January 3, 2006 Author Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Ive not updated the odds comparison on my site to include cup matches yet, these are from my local version: POTENTIAL MATCHES ( prices are an indication only ! ) 2006-01-07 FA Cup Arsenal v Cardiff City 1.14 Chelsea v Huddersfield 1.05 Man. City v Scunthorpe 1.19 Newcastle v Mansfield 1.13 Nuneaton v Middlesbrough 1.18 Stoke v Tamworth 1.29 Fulham v L Orient 1.280 2006-01-07 Greece Hellenic National League Panathinaikos v Larissa 1.25 2006-01-08 FA Cup Sunderland v Northwich 1.20 Burton v Man. United 1.09 2006-01-08 Italian First Division Milan v Parma 1.228 Link to comment Share on other sites More sharing options...
slapdash Posted January 3, 2006 Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) slapdash i see your point that with plateau staking and not recalculating when the bank goes down there is an increased risk of bancruptcy. A 100% risk of bankruptcy eventually. Though it might take a very long time. Lets face it : it is loss recovery. Yes, it's a kind of logarithmic Martingale. After a loss you increase not your stake but the percentage of your current bank that you are staking. And as with Martingale you will usually survive in the short term, but eventually you'll have that unlucky streak that leads to you increasing the percentage more and more until ... disaster. With percentage staking, there is an optimal percentage of your bank to use, depending on odds and yield, given by the Kelly formula. If you use either more or less than this percentage, then your bank won't grow as quickly. For odds of 1.20 and a yield of 4% (roughly your figures for last season) the optimal percentage is 20% of your bank. I'm pretty sure that the only reason your calculations showed plateau staking doing better than percentage staking is that with plateau staking you were "accidentally" staking closer to the optimal percentage when your bank dropped. That's why I asked what would happen if you changed the 10% to 20%: now when you "accidentally" stake more than 20% of your bank because your bank drops you will be moving away from the optimal amount, so I'd predict that plateau would no longer have the same advantage over percentage. You would get the same bank growth as you observed with (10%) plateau staking, but at lower risk, by using percentage staking with a slightly higher percentage. By the way, I'm not suggesting using 20% of your bank as a stake. There are practical reasons to use less than the "optimal" percentage. Link to comment Share on other sites More sharing options...
Datapunter Posted January 3, 2006 Author Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) A 100% risk of bankruptcy eventually. Though it might take a very long time. Ok, i got that, had to think about this for a bit because i was resisting. Here's the deep dark thought in the back of my head: yes yes thats all very fine theoretically but it will never happen in real life So just examining the way my mind works here. This system has for several seasons proved to be continuously profitable. So for this system to turn into a loss making system something would have to change. A change could be gradual. Bookmakers changing their pricing methods, exchanges shifting the market, governments imposing taxes, footballshoes becoming more efficient, etc... All these could result in a shift in price or strike rate. But that shift would be gradual and not sudden so i would have time to react. The system results would first level out and then start declining. A change could be sudden. But this would take a major change in the sport or betting industry. For example the off-side rule gets scrapped. You get a whole new game of football. This would not come as a surprise so anything happening that might cause a major shift will be announced and i would probably simply stop betting until the effect of the shift is known. So for me to actually going bancrupt not recalculating the stake when the bank goes down will not happen in real life. That will only happen in theory with a system running indefinetely. But what am i actually thinking and saying here ? Basically i'm saying i don't believe i will ever have a losing run. And there is the key word: believe. Now who in his right mind is going to base a betting strategy on belief :lol So plateau staking without ever decreasing the stake if the bank goes down is now out the door. A losing run is a real possibility and if it hits it will kill the bank if the stakes are too high. Thank you very much Mr. Slapdash Going to run some more numbers through the spreadsheet now :ok Link to comment Share on other sites More sharing options...
PAULM03 Posted January 3, 2006 Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) There speaks a man who knows how to make money:clap That has to be one of the best posts I've ever read on here and a must for anyone new to betting. Link to comment Share on other sites More sharing options...
Datapunter Posted January 3, 2006 Author Share Posted January 3, 2006 Link to comment Share on other sites More sharing options...
Datapunter Posted January 3, 2006 Author Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) :lol Link to comment Share on other sites More sharing options...
Mr Intensity Posted January 3, 2006 Share Posted January 3, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) :lol :lol Keep thinking DP this thread is an absolute classic:clap :clap :clap :notworthy :notworthy :notworthy Link to comment Share on other sites More sharing options...
Datapunter Posted January 4, 2006 Author Share Posted January 4, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Now included the Scottisch FA POTENTIAL MATCHES ( prices are an indication only ! ) 2006-01-07 FA Cup Arsenal v Cardiff City 1.14 Alloa v Livingston 1.28 Chelsea v Huddersfield 1.05 Falkirk v Brechin 1.23 Hibernian v Arbroath 1.11 Inverness v Ayr 1.24 Man. City v Scunthorpe 1.19 Newcastle v Mansfield 1.13 Nuneaton v Middlesbrough 1.18 Rangers v Peterhead 1.07 Ross Co v Forfar 1.34 Stoke v Tamworth 1.29 2006-01-07 Greece Hellenic National League Panathinaikos v Larissa 1.25 2006-01-08 FA Cup Clyde v Celtic 1.12 Fulham v L Orient 1.280 Sunderland v Northwich 1.20 Burton v Man. United 1.09 2006-01-08 Italian First Division Milan v Parma 1.228 Joe has advised on 11 of these, so should get busy this weekend :clap Link to comment Share on other sites More sharing options...
Norfolk Enchance Posted January 4, 2006 Share Posted January 4, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) I wouldn't recommend Alloa at that price DP ;) Link to comment Share on other sites More sharing options...
Datapunter Posted January 4, 2006 Author Share Posted January 4, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Aye, those are just a 'snapshot' prices, doing the rounds of the odds comparison sites now to make an early selection. Link to comment Share on other sites More sharing options...
slapdash Posted January 4, 2006 Share Posted January 4, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) I think he just meant that you've put the wrong team in bold type for that match! Link to comment Share on other sites More sharing options...
Norfolk Enchance Posted January 4, 2006 Share Posted January 4, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Sorry, yeah, I just didn't want you going over to betfair and putting in an offer to back Alloa instead of Livi :) Link to comment Share on other sites More sharing options...
Datapunter Posted January 4, 2006 Author Share Posted January 4, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Ah, Alloa, i would have noticed you know.... eventually .... :$ Anyhow, here we go ( matches with * are also advised by Joe ) * 2006-01-07 14:00 Arsenal v Cardiff 10 @ 1.16 * 2006-01-07 16:00 Chelsea v Huddersfield 10 @ 1.09 - 2006-01-07 16:00 Falkirk v Brechin 10 @ 1.30 * 2006-01-07 16:00 Hibernian v Arbroath 10 @ 1.14 * 2006-01-07 16:00 Inverness CT v Ayr 10 @ 1.29 * 2006-01-07 16:00 Man City v Scunthorpe 10 @ 1.22 * 2006-01-07 16:00 Newcastle v Mansfield 10 @ 1.16 * 2006-01-07 16:00 Nuneaton v Middlesbrough 10 @ 1.21 - 2006-01-07 16:00 Stoke v Tamworth 10 @ 1.33 - 2006-01-07 18:15 Panathinaikos v Larissa 10 @ 1.28 the ones that didn't make it are: Livingston, no market/volume available on Betfair, may happen later... ( I may cancel Panatinaicos temporarily to be able to bet on Livingston, i'll see ) Rangers, don't trust Rangers and the price just isn't worth it... Ross co, price range is now in the 1.30's so less price bias... And as you may have noticed i've used the entire bank :loon So these bets i intend to place saturday, ( bar an amazing losing run that is :lol ) * Celtic @ 1.15 - Fulham @ 1.30+ * Sunderland @ 1.23 * Man. United @ 1.12 - Milan @ 1.26 There you go, nice and early :ok Link to comment Share on other sites More sharing options...
slapdash Posted January 4, 2006 Share Posted January 4, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) And as you may have noticed i've used the entire bank :loon So these bets i intend to place saturday, ( bar an amazing losing run that is :lol ) If none of the ten wins, you can have a few drinks on me to cheer you up. :ok Link to comment Share on other sites More sharing options...
Datapunter Posted January 4, 2006 Author Share Posted January 4, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Ahhhh, slapdash, thats so sweet :cheers With percentage staking, there is an optimal percentage of your bank to use, depending on odds and yield, given by the Kelly formula. If you use either more or less than this percentage, then your bank won't grow as quickly. For odds of 1.20 and a yield of 4% (roughly your figures for last season) the optimal percentage is 20% of your bank. How did you work out the 20% ? :tongue2 Link to comment Share on other sites More sharing options...
AJ Posted January 5, 2006 Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) A small query re the betfair prices, when do you go to the market looking for a match ? Is it early in the week or later, do you find you have a better chance of getting matched at your price if you go earlier?? Do prices tend to move in as time goes by ?? Link to comment Share on other sites More sharing options...
slapdash Posted January 5, 2006 Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) How did you work out the 20% ? :tongue2 The smiley makes me think I'm probably missing something, but ... 4% (yield) divided by 0.2 (decimal odds minus 1) is 20%. Link to comment Share on other sites More sharing options...
Datapunter Posted January 5, 2006 Author Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Foolsgold You never really know which way a market is going to go. So i simply don't even try to guess. The impression i have is that low volume matches tend to rise in price whereas high volume matches tend to rise a bit then fall. But it is no more than an impression. My method is this: - look at my own odds comparison and make a shortlist of possible bets, - check the average price at other comparison sites like betbrain and oddschecker - decide what price i want to go for ( ususally wait for Joe's selections before making a final decision, just in case ) - put up an offer at Betfair at the price i want and hope to get matched ( all above bets are on offer at Betfair with those prices right now ) - check how the price goes every now and then and adjust accordingly, usually dropping a bit in order to get matched. - cancel bets where the price drops too far down, ( or never rises high enough ) - NEW, look for alternative prices at the bookies Sometimes i need to lower my price because the market doesn't rise. Sometimes i get matched. Sometimes the price rises above my set price and i could have gotten a better price had i waited. In the end it all levels itself out. Key is to make sure you only have a bet when the price is reasonably above average. Define reasonably ? i hear you ask. That depends on each individual case and i really don't have any fixed rules i follow. I tend to take a price close to average for the big teams like Chelsea and i tend to let a bet on smaller teams go by if the price isn't really good, like the Dutch or some Scottisch teams. So just checking now: Arsenal v Cardiff 10 @ 1.16................matched @ 1.16 , trading at 1.15 Chelsea v Huddersfield 10 @ 1.09........trading at 1.07 lowering price to 1.08 Falkirk v Brechin 10 @ 1.30................no volume, can't tell Hibernian v Arbroath 10 @ 1.14..........low volume around 1.14, no change Inverness CT v Ayr 10 @ 1.29............no volume, can't tell Man City v Scunthorpe 10 @ 1.22........matched @ 1.22 , trading at 1.21 Newcastle v Mansfield 10 @ 1.16........matched @ 1.16 , trading at 1.16 Nuneaton v Middlesbrough 10 @ 1.21..trading at 1.19 , no change Stoke v Tamworth 10 @ 1.33..............low volume around 1.31 , no change Panathinaikos v Larissa 10 @ 1.28......low volume around 1.26 , no change But remember these are very early markets, over 2 days to go.... Link to comment Share on other sites More sharing options...
Datapunter Posted January 5, 2006 Author Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) slapdash I had a quick look around for the Kelly formula and it is applied here but couldn't make heads or tails of it. ( the :tongue2 means: eargerly anticipating the reply ) There seem to be quite a few definitions around :unsure Most of them talk about value in the odds, on each individual bet, and i'm having trouble making the bridge to applying it here to a whole system. Most confusing. I found this one: The Kelly's formula is : Kelly % = W - (1-W)/R where: Kelly % = percentage of capital to be put into a single trade. W = Historical winning percentage of a trading system. R = Historical Average Win/Loss ratio.Which makes little sense to me, and i found this one:Bet Amount = Overlay / (bookies price – 1) Which appears to be what you have applied. So..... Any chance of a translation into plain english ? Please. I'm lost. Link to comment Share on other sites More sharing options...
slapdash Posted January 5, 2006 Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) The Kelly's formula is : Kelly % = W - (1-W)/R where:Kelly % = percentage of capital to be put into a single trade. W = Historical winning percentage of a trading system. R = Historical Average Win/Loss ratio.Which makes little sense to me, ... and even less to me! So..... Any chance of a translation into plain english ? Please. I'm lost. OK, here's how I think about the Kelly formula. First of all, let's make the unrealistic assumption that you have a sequence of bets that all have the same odds (call the decimal odds N) and the same expected positive yield (call the yield x%). Since the yield is positive then of course you can make money, but how much should you stake on each bet to maximize your winnings? The wrong question to ask is "What stake will maximize my expected winnings after a given number of bets?". The answer to that is that you should stake your entire bank every time. Unfortunately, although this will give the highest possible average profit after, say, 100 bets, this average is made up from winning an astronomical amount with a microscopic probability and losing everything most of the time. So in practice you will lose everything if you follow this strategy. But what happens if you stake a fixed percentage of your current bank on each bet, and the expected number of bets win? An example, where the numbers are easy: Suppose the bets are at odds of 3.00 and have a 50% chance of winning, so the expected yield is 50% (i.e., N=3 and x=50). Suppose you bet 25% of your current bank on each bet. When you win, you win 50% of your current bank, so multiply your bank by 1.5. When you lose, you lose 25% of your current bank, so multiply your bank by 0.75. So one win and one loss, which is what you expect from two bets since the probability of winning is 50%, will multiply the bank by 1.5 x 0.75 = 1.125. Suppose instead you bet 20% of your current bank. A win multiplies your bank by 1.4 and a loss by 0.8. So a win and a loss multiplies the bank by 1.4 x 0.8 = 1.12. Or 30% of your current bank? A win multiplies the bank by 1.6 and a loss by 0.7. So a win and a loss multiplies the bank by 1.6 x 0.7 = 1.12. So if you bet a bit more or a bit less than 25%, your bank grows less quickly, assuming average luck. What if you bet a lot more than 25%? 50%? A win multiplies the bank by 2 and a loss by 0.5, so a win and a loss multiplies the bank by 2 x 0.5 = 1, so with the expected proportion of winners your bank won't grow at all. 60%? A win multiplies the bank by 2.2 and a loss by 0.4, so a win and a loss multiplies the bank by 2.2 x 0.4 = 0.88, so with the expected proportion of winners your bank will actually decrease. It's still true that after a given number of bets you will on average be in profit (with a greater average profit than if you always bet 25% of the bank), but this average is made up of a huge profit if your luck is much better than expected, but a loss of most of your bank if your luck is about average. This is typical of what happens in general. There's a particular percentage of your bank to use that will make your bank grow fastest if you have average luck, and it turns out to be given by a simple ("Kelly") formula x% divided by N-1 in terms of the expected yield and odds. (If you want, I could explain how this formula is derived: the quickest way I know uses logarithms and some fairly basic calculus.) If you bet less than the Kelly amount, your bank will grow slower, but it will still grow. If you bet more, your bank will grow slower, and if you bet too much more then your bank will decrease: there's no simple formula for when this starts to happen, but it's roughly twice the Kelly amount. How does this apply when your bets have varying odds and yields, and what does it say about a single bet? If you use level stakes then with some very mild assumptions on the bets (all odds being less than a million is plenty good enough) it's well known that after a large number of bets, your average profit per bet is almost sure to be close to the expected average profit. This is the "Law of Large Numbers". It doesn't apply if you use percentage staking or other staking plans where the stake increases with the bank. However, if you use logarithms ... [Remember that the key property of logarithms is that they change multiplication into addition, so log(A x B) = log(A) + log(B).] If you place 25% of your bank on a bet at odds of 3.00, then if you win you multiply your bank by 1.5, so you add log(1.5) to the logarithm of your bank. If you lose, you multiply your bank by 0.75, so you add log(0.75) to the logarithm of your bank (since log(A) is negative if A Link to comment Share on other sites More sharing options...
Datapunter Posted January 5, 2006 Author Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) thank you very much :clap I've just read that about 5 times and i think i'm actualy beginning to understand :loon Digesting further .... Link to comment Share on other sites More sharing options...
merlin Posted January 5, 2006 Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) .... looks like far too many bets to contend with for this weekend DP.I have seen Joe's list too, and although the selections look sound enough, the prices are absolutely ridiculously low.Iv'e gone with three namely, Man C, Sunderland and Milan where I got what I consider to be reasonable prices, and left the rest due to skinny prices and some bank roll for another day! Did have a go with Peterborough U @ 1.75 with a saver on the draw.Always an interesting thread and I always look forward to turning this thread up.Don't like the Kelly formula too much, maybe your plateau system of bets holds more merits for me.Interesting point you put forward about Betfair and showing the odds you are asking, may try this in the near future, might just make the difference.Good luck with the selections.;) Link to comment Share on other sites More sharing options...
Datapunter Posted January 5, 2006 Author Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) thanks merlin, yes the prices aren't brilliant but still value enough to fit the system, and i can understand if you don't like the "Kelly" subject, staking is just as much a matter of personal preference as it is about maths. If we were all robots then only the maths would matter. Unfortunatly we're human and our sentiments and feelings greatly influence our success in whatever we do. If it doesn't "feel" right, think twice. Meanwhile, Chelsea got matched at 1.08 , still trading at 1.07 Link to comment Share on other sites More sharing options...
merlin Posted January 5, 2006 Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) ... exactly my sentiments on the success rate and how it is influenced by ones thoughts and feelings.Most of the prices on offer certainly did not feel 'right', after all it's supposed to be the F A Cup where shocks often happen.May be different if I were slapping 100 stake on every match highlighted, but 10's is an entirely different kettle of fish! Go to make something to make it all feel worthwhile.:tongue2 Link to comment Share on other sites More sharing options...
Datapunter Posted January 5, 2006 Author Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Right, lets see if i got this... Suppose you bet 25% of your current bank on each bet. When you win, you win 50% of your current bank, so multiply your bank by 1.5. When you lose, you lose 25% of your current bank, so multiply your bank by 0.75. So one win and one loss, which is what you expect from two bets since the probability of winning is 50%, will multiply the bank by 1.5 x 0.75 = 1.125. This is the plain english i was after. And now i understand that there actually is a percentage that will make the bank grow most efficiently. There's a particular percentage of your bank to use that will make your bank grow fastest if you have average luck In the simple example it does not matter in what order you get a winner/loser. In an actual betting system you do get winning/losing runs but if the distribution winners/losers is normal, i.e. normal for a given strike rate then in the end it levels out. You could start off with a losing run and that would mean it takes a lot longer to get to the expected profit, and that would simply be bad luck. it turns out to be given by a simple ("Kelly") formula x% divided by N-1 in terms of the expected yield and odds. (If you want, I could explain how this formula is derived: the quickest way I know uses logarithms and some fairly basic calculus.) Ok, for myself i will pass on the detailled explanaition and take your word for it :ok But maybe others are interested, merits a thread of its own maybe. How does this apply when your bets have varying odds and yields, and what does it say about a single bet? Ok, the bottom part i had to read quite a few times. I did learn about logarithms in school but i haven't actually done anything with them in over 20 years. So the details of the maths escape me but let me say in my own words what i got and see if it fits. Applying the stake given by the "Kelly" formula gives most certainty that the expected yield/profit will actually happen. (all else being equal) There is an optimal size stake that will make the bank grow fastest. Stake less and it grows slower, stake more and it also grows slower or even deminishes. My expected yield is 4% no matter what staking i apply, the result may be different but the expectation is still 4%. So all else being equal if i stake the amount given by the formula i will end up with a 4% yield. If i stake anything else i may end up with 6% but i may also end up with 2% . Personally i prefer more certainty to end up with 4% than the possibility of 6% . Therefore the staking should be as optimal as possible. This is how i understand the part of determining the stake size. In the overall picture of the system there are other things to consider but i got this is a key factor. Link to comment Share on other sites More sharing options...
Datapunter Posted January 5, 2006 Author Share Posted January 5, 2006 Re: Soccer Hot Favourites System ( at +7.11% ) Just a quick calculation here, imagine i expect a yield of 5% at odds of 1.25 the most efficient stake is 20% of the bank. ( 5 / 0.25 ) Using my own table to determine the risk of bancruptcy, uncompleted but still, a level stake of 20 points with 500 bets at a strike rate of 80% gives a risk of going bancrupt of something between 60% and 75%. Sure the positive yield will bring this down as well as the effect of percentage staking, but still it will be quite high. Unacceptably high. A losing run of 3 bets, matches that happen to occur at the same time, will bring the bank down by 60% , wow, i do not like the thought of that. And that HAS happened. So an early conclusion is that i probably won't be able to use the optimal stake size because it means an unacceptable level of risk. The trick will be to balance optimal stake size with risk. Another practicality to consider is that with 20% of the bank on each bet i can only have 5 bets at the same moment in time. That would be difficult this weekend as 8 games are at the same time. Food for thought eh.... :lol I think i'll take a little break and let it all sink in a bit. ( btw i made a mistake with the plateau staking, i recalculated every time the bank grew with 25 points, that should have been 25 percent :$ big difference, sorry if i've put anyone on the wrong foot ) Next week i'll start all over :lol - determine what level of risk is acceptable, - calculate the optimal stake as now discussed, - find the right balance between the 2 , optimal stake and risk - compare percentile staking with other staking plans. - pick the one thats right for me. Many thanks slapdash :ok you've certainly shifted my views. Link to comment Share on other sites More sharing options...
Recommended Posts