Percentage staking plan, thoughts?

[Hooloovoo]

Got back into sportsbetting just recently and i have been thinking about my staking plan and how to "balance" it. have decided to go for the strategy where every bet stake is a percentage of my current bank, and this is recalculated every day. So if my bank is 100pts, and each bet is 3% of my bank, then bet size is 3pts However, I am betting quite high volume (i.e. many bets) every week, this is a typical week: monday - thursday: 5 bets a day friday: 7 bets saturday - sunday: 10-15 bets a day This means that on a saturday with 15 bets, i am betting 15 x 3% = 45% of my bank, and it doesn't feel right to have so much of the bank "live". If i have a really bad run, then almost half of my bank could be wiped out! One solution could be to allow myself to bet a certain percentage of the bank each day, for example max 21%. Then i could have a maximum of 7 bets per day. An alternative is to estimate how many bets there will be on a certain day and to spread out the 20% over these bets. The thing is, when i place my first bet of the day, i have no idea how many more bets there will be. I like to place my bets instantly to take advantage of odds before they drop. Any thoughts or suggestions?

[aliando]

Re: Percentage staking plan, thoughts? if you're placing a hig number of bets that are waiting on a result simultaneously, you could use a smaller % to protect yourself against a losing run of 'already placed bets' another alternative will be to maximise the number of bets placed at one time - so rather than having a max of 7 a day, you could have a max of 7 at a time. I'm sure I read on here a while back that the best % of bank to use (in a maths context rather than gambling) was 4.5% but no idea where/when/why this was said :$

[Lardonio]

Re: Percentage staking plan, thoughts? Does it really matter if you have 15 bets placed on one day rather than one week or one month? No, it doesn't, so if you're uncomfortable with 15 bets in a day it's because your % per bet is to high. Go down to 2% or 1%, whatever makes you comfortable betting the style you like.

[Hooloovoo]

Re: Percentage staking plan, thoughts? Thanks for the comments guys.

another alternative will be to maximise the number of bets placed at one time - so rather than having a max of 7 a day' date=' you could have a max of 7 at a time.[/quote'] Good point :)

[Hooloovoo]

Re: Percentage staking plan, thoughts? In other words, i am prepared to take higher risks (more aggressive BRM) in order to get the bank up faster. On the other hand, i want to have reasonable protection against a really bad run. These goals are somewhat contradictory and i'm trying to find the right balance here. Again, thanks for your comments and keep adding please

[Lardonio]

Re: Percentage staking plan, thoughts? Well, I think you have to figure it out for yourself whether you want to take high risk or protect your bank.... Either way, you will take risk. If you sometimes place 15 bets each day, but you force yourself to limit the number of bets to 7 you could end up having bigger total losses on the day than you would if you had bet all 15. Personally, again :), I would choose to have faith in my betting and lower the percentage per bet to a level where I would be comfortable betting all the bets I wanted to bet.

[Hooloovoo]

Re: Percentage staking plan, thoughts?

Personally' date=' again :), I would choose to have faith in my betting and lower the percentage per bet to a level where I would be comfortable betting all the bets I wanted to bet.[/quote'] This sounds reasonable, it will have the added benefit of lower variance, which is what we want all the time :) I could live with betting 30% of my bank in one day, if those percentages are spread over 15 bets.

[Datapunter]

Re: Percentage staking plan, thoughts? The problem you have is not one of numbers, it's one of human perception. When a large portion of your bank is exposed it appears as a scary high risk. This is your perception only and it is likely what everyone else would perceive. However you are managing your risk by determining a % of your bank per bet. It does not matter if you place one bet at a time or a whole group. The risk does not change, only your perception. If you do hit a losing run you won't know beforehand. Wether all bet are placed at the same time or one by one it will be the same losing run. But now we do come to an aspect which appears to remain a question for you. If you placed bets one by one and you had X losers in a row would you stop betting ? If the answer is yes after 10 bets then that is the maximum number of bets you are willing to suffer before you stop betting. However i do not believe the number of bets should be part of a staking plan therefore you should be betting in a way that if you had 20 bets at the same time you should be able to place all bets. The selections should not be controlled by staking. The solution is to lower your % per bet so that you are comfortable with a 20 bet losing streak, or at least that's the maxium you're willing to suffer. And then if there is a day with 20 bets, as that is what your selection process throws up, place the lot.

[Hooloovoo]

Re: Percentage staking plan, thoughts? Very good comment Datapunter, i have read it a couple of times and it makes perfect sense. Made me think about it in partly different ways :ok

[skylark2009]

[Gingertipster]

Re: Percentage staking plan, thoughts?

The thing is, when i place my first bet of the day, i have no idea how many more bets there will be. I like to place my bets instantly to take advantage of odds before they drop. Any thoughts or suggestions?

By that I take it you are a "value" punter Hooloovoo? Betting as and when you see anything over-priced. If indeed you are a "value" punter, shouldn't your stake be influenced by what chance you believe the selection has of winning and how much value you believe is in the bet? i.e. Should you really be betting the same amount on a 10/1 shot as you do at Evens? And Should you be betting the same amount on a 5/1 shot that you believe is a true 4/1 shot; as you would a 5/1 shot you believe a true 2/1 shot?

[jimjohn]

Re: Percentage staking plan, thoughts? Another informative thread. I've not been here long but keep getting answers to questions I've not asked yet. A nice formula there skylark, that will come in handy for my staking, cheers everyone, jimjohn.

[Hooloovoo]

Re: Percentage staking plan, thoughts?

By that I take it you are a "value" punter Hooloovoo? Betting as and when you see anything over-priced. If indeed you are a "value" punter, shouldn't your stake be influenced by what chance you believe the selection has of winning and how much value you believe is in the bet? i.e. Should you really be betting the same amount on a 10/1 shot as you do at Evens? And Should you be betting the same amount on a 5/1 shot that you believe is a true 4/1 shot; as you would a 5/1 shot you believe a true 2/1 shot?

Thanks for your comment Ginger. Well i only, ever, place a bet if i believe there is value. You are right, its a twofold question. If there are two shots with odds of 1.80 and 8.30 respectively, but with the same percentual percieved "value", then i let the stakes differ with regards to the difference in odds. This is only to reduce variance, and i might lose some value by doing this. I have chosen to go for full stake for every shot with odds less than 3.00, and half stake for odds between 3.00 - 5.00. I have yet to see a selection with odds over 5.00 from my selection process. Concerning the other case, with two selections with different percentual value, well, so far i have been following the same staking as in the previous example. This will also means losing out on possible value, but as i see it, the art of deciding how big the edge is, it's very difficult and risky business. But deffo something i will look into further, a refinement of the staking can certainly be made there.

[gloryhunters]

Re: Percentage staking plan, thoughts? There are several questions you need to be able to answer before you can decide how much to stake. What is your win rate and what variance do you have. If you average to win 1 unit per 10 bets, it does make a huge difference if this comes from betting on 20/1 shots or 1/10 shots. You should also recalculate the percentage bet after each bet. So the premise than it doesn't matter is wrong. For example if you bet 5% a go and had 20 bets that's your whole roll and you'd have a greater chance of going broke than if you bet 5% of your roll on each race as it came. Of course higher broke percentage also means higher reward when doesn't happen:) Finally, the risk you should take should also relate to how much it matters if you go broke and create a new roll.

[Lardonio]

Re: Percentage staking plan, thoughts?

Should you really be betting the same amount on a 10/1 shot as you do at Evens? And Should you be betting the same amount on a 5/1 shot that you believe is a true 4/1 shot; as you would a 5/1 shot you believe a true 2/1 shot?

The answer to the first question, from a "value" punter's point of view, is a solid yes. Any person who answers no to that question does not understand value and should not pretend to be a "value" punter. The second question is more tricky. Mathematically, if all numbers were known and 100% correct, you should be betting more on the 5/1 shot which is a 2/1 shot. Personally, I find that trying to find "value" is hard enough without trying to grade the value as well. I know people who have fancy charts with bet sizes according to 'how much' value they have found. But really, all they are doing is taking a number they don't know and make calculations from it. The excel charts with formulas look good, but they are risking turning a value project into a non-value project.

[Datapunter]

Re: Percentage staking plan, thoughts?

all they are doing is taking a number they don't know and make calculations from it

Seconded. All this talk about staking the same on 10.00 as on 2.00 is a lot of theoretical hot air. If you actually do have a selection process that throws up that extreme range of odds then first i seriously doubt the validity of the method. The dynamics involved at those ends of the range are so different no single method can be applied. Ok, let's say for arguments sake you do end up with that kind of odds range. It should not be considered one and the same strategy but split into sub-strategies. Where you make the split depends on any number of aspects but i'd say the dynamics of the markets you bet on are a good start. In other words high favourites versus longshots should be considered 2 different strategies which can both be members of the same family and have aspects in common. You then devise a staking plan for each sub-strategy. As each sub-strategy naturally produces selections of the same nature they will also reasonable fall in the same odds range. As long as the distribution of bets is reasonable spread out over the range of odds you'll find it makes very little difference if you have a fixed stake and variable profit or the other way around a variable stake and fixed profit. The resulting yield and profit will be about the same. Stick your bets in a spreadsheet, randomise them a few times and compare the 2, easily done and checked. The practice of changing the % based on the odds range alone is basically what this comes down to, although it's a rather crude approach.

[Hooloovoo]

Re: Percentage staking plan, thoughts? Different opinions makes for interesting read :) I can sum up my staking easily. My selection process gives selections in the odds range of about 1.40 - 4.00 As the bank is new and kind of small, i prefer a rather aggressive approach, i.e. i am willing to take some risks to increase the bank more rapidly. Odds-band 1.40 - 2.99: Betsize 5% of bank Odds-band 3.00 - 4.00: Betsize 2.5% of bank Betsizes are recalculated each morning. And finally, i don't let my perception of "how much" value that might lie in a selection influence the betsize. As a remark, it would proably be wise to bet the same amount (percentage) no matter the odds, i will have to think further about that.

[Gingertipster]

Re: Percentage staking plan, thoughts?

The answer to the first question' date=' from a "value" punter's point of view, is a solid yes. [b']Any person who answers no to that question does not understand value and should not pretend to be a "value" punter. The second question is more tricky. Mathematically, if all numbers were known and 100% correct, you should be betting more on the 5/1 shot which is a 2/1 shot. Personally, I find that trying to find "value" is hard enough without trying to grade the value as well. I know people who have fancy charts with bet sizes according to 'how much' value they have found. But really, all they are doing is taking a number they don't know and make calculations from it. The excel charts with formulas look good, but they are risking turning a value project into a non-value project.

Lardonio, I've agreed with you on many threads, talking a lot of sense. But your first paragraph seems more than a touch arrogant. If your answer is "yes" then fair enough, that is your opinion. But there are more ways of working with value than your way. Your third paragraph dismisses what many professional gamblers do. Many successful punters including Dave Nevison and Eddie Freemantle work out a 100% book and bet from that. I too usually work out my own tissue and back anything over-priced. If you "find that trying to find value is hard enough without trying to grade the value as well", fair enough. But some punters are capable of doing so without "fancy charts". If bookmakers (or rather Peter Michael or odds compiler) can work out an accurate enough tissue to offer just 2% more. Something he believes a true 20% 4/1 shot may well be offered at 22.2% 7/2 (or if a difficult market to work out 23.1% 100/30) then why can't us punters work out accurate tissues? My biggest win was on Rooster Booster for the 2003 Champion Hurdle ante-post. Immediately after winning the 2002 Greatwood Hurdle I considered him a 20% 4/1 chance for the Champion. Stan James were best priced 12/1 (7.7%). So in my opinion the 12/1 was over 12% better than my price (20 - 7.7). When my Timeform Perspective form book came a few days afterwards, they seemed to agree, saying the performance rating was good enough to win 3 of the last 4 Champion Hurdles. So I went in again at the by now top price of 8/1. Rooster Booster won at Cheltenham at SP of 9/2. Yet in your opinion it should have been just a normal sized bet at 12/1? The same size bet as if he'd been half the available odds! I found out when reading Dave Nevison's book, Rooster Booster was also his biggest individual win, backed at the same time.

[Gingertipster]

Re: Percentage staking plan, thoughts? I used to bet in level stakes. But found it mattered less about whether my short priced selections won or lost. Everything depended on how many outsiders won. Obviously with level stakes a punter needs to win ten Even money chances to win the same as one 10/1 shot. It did not seem right so I changed to betting to win the same amount. But this time it mattered less about the outsiders winning or losing. As I was losing much more on the shorties. Again it did not seem right. I went in search of a staking plan that would enable me to put more on the shorties, as they by and large have a greater chance of winning. Yet win more on the outsiders, as - with less money risked - (imo) they deserve to win more. It occurred to me that anything I believe has double the chance of winning should have double the stake. i.e. Anything I rate as a fair Even money shot (50% chance), should have double the stake of what I believe has a fair 3/1 (25%) chance, which in turn should have double the stake of one I believe has a fair 7/1 (12.5%) chance, and so on. So anything I believe has a 50% chance of winning gets 50 points, 25% gets 25 and 12.5% rounded up to 13 points. However, despite decreasing stakes strangely enough by and large the less money staked the bigger the price, more money is won. 50 points @ 11/10 (47.6%, difference between the two (50% and 47.6%) being 2.4%) wins 55 points. 25 points @ 7/2 (25 - 22.2 = 2.8% difference) wins 87.5 points. 13 points @ 9/1 (2.5% difference) wins 117 points. Less money staked more money won. However, my stake should also take in to account the amount of value I believe is in the bet. i.e. Something I believe a 20% (fair 4/1) chance, who is available at best price 6/1 (fair 14.3%) should have more stake than if the best price was only 5/1 (fair 16.7%). So I came up with an idea to add a figure directly related to the amount of value. With my 20% chance available at 6/1 (14.3%). Difference being 5.7% (20 – 14.3 = 5.7). Then x 3. So 5.7 x 3 = 17.1. Rounded down (or up). So anything believed to be a 20% chance who is available at top price 6/1 gets 20 + 17 = 37 points @ 6/1 winning 222 points. Any 20% chance only available at top price 5/1 gets 30 points staked. 20 – 16.7 = 3.3 x 3 = 9.9 so 20 + 10 = 30 points @ 5/1 winning 150 points. Am not really in favour of “banks”. Obviously every punter should bet what he can afford and every now and again a punter should up or decrease stakes. So every month or few months he should consider how much each point should amount to.

[Lardonio]

Re: Percentage staking plan, thoughts? Fair play, Gingertipster, my "background" is betting on football, and I know very little about horses. I have read books about horse punters who identify a winning horse and then pile all the money they can find on it. That is a world I don't understand, and to me it seems there's a lot more going on in the horse markets than just horses with small people on top of them running vs each other. Regarding your comment about bookmakers calculating an accurate tissue, well they don't really, otherwise we wouldn't be here. But they're allowing them a big enough margin of error to profit, AND they constantly change their prices, block winning punters etc.... a luxury we as punters can't do. But, in theory, I agree 100% with you that for example if I was offered a 100/1 bet on there being rain tomorrow I would bet all I could spare. When it comes to betting on sports, my experience is that I won't trust myself to change the bet sizes because then I'll usually lose my cool and get excited. If I place huge bets on something then I keep it aside from my usual betting projects. I heartily congratulate everyone who does massive gambles based on value and profits more from it. As for my arrogant paragraph, 2%, 5% or 10% value gives the same return regardless of whether the odds are 1.1 or 1000 - mathematically. Understanding that is the core of understanding what people call 'value betting'. Then you can start adding your personal preferences and skills into it, and you might get a different strategy.

[Gingertipster]

Re: Percentage staking plan, thoughts?

2%' date=' 5% or 10% value gives the same return regardless of whether the odds are 1.1 or 1000 - mathematically. Understanding that is the core of understanding what people call 'value betting'. Then you can start adding your personal preferences and skills into it, and you might get a different strategy.[/quote'] Not sure I know what you mean by that Lardonio. Don't think we can be talking about the same thing. Say I believe a selection has a 25% 3/1 (4.00) chance of winning. If it were top price of 100/30 (4.33) 23.1%, (25 - 23 = 2%) it is just 2% value (1.9% to be exact). If it were top price 4/1 (5.00) 20%, it is 5% better (25 - 20 = 5%). 5% value. If it were 11/2 (6.5) 15.4%, (25 - 15 = 10%) it is 10% better. 10% value (9.6% to be exact). Every point @ 100/30 returns 4.33 points. Every point @ 4/1 returns 5 points. Every point @ 11/2 wins 6.5 points. So it does make a big difference to returns.

[gloryhunters]

Re: Percentage staking plan, thoughts? By following that system and assuming an edge of 1% on all bets, then you’ll basically go broke about 5% of the time over a series of 500 bets.

At the end of 200 bets you’ll have decreased you betting bank over 50% of the time. This is also true after 500 bets.

If your edge increased to 3% then you’ll only go broke (eg have less than 10% of original betting bank) about 0.1% of the time. You’ll increase your bank after 200 bets over 70% of the time and around 80% of the time after 500 bets.

[Gingertipster]

Re: Percentage staking plan, thoughts?

Fair play, Gingertipster, my "background" is betting on football, and I know very little about horses. I have read books about horse punters who identify a winning horse and then pile all the money they can find on it. That is a world I don't understand, and to me it seems there's a lot more going on in the horse markets than just horses with small people on top of them running vs each other. Regarding your comment about bookmakers calculating an accurate tissue, well they don't really, otherwise we wouldn't be here. But they're allowing them a big enough margin of error to profit, AND they constantly change their prices, block winning punters etc.... a luxury we as punters can't do. But, in theory, I agree 100% with you that for example if I was offered a 100/1 bet on there being rain tomorrow I would bet all I could spare. When it comes to betting on sports, my experience is that I won't trust myself to change the bet sizes because then I'll usually lose my cool and get excited. If I place huge bets on something then I keep it aside from my usual betting projects. I heartily congratulate everyone who does massive gambles based on value and profits more from it.

Would've thought every sport is the same, with football it is a three horse race. Team A, Team B and the Draw. Study the form, work out what you believe the percentage chance of all three results are. Convert to odds and back the value selection. Bookmakers (odds compilers) work out a book "accurate" enough to only enable a fraction of punters to make a profit. Punters (in any sport) can use their "skill" to work out a market "accurate" enough to be one of that "fraction". The ability to "trust" oneself to grade value comes with experience and profit, once a punter is confident enough to find that value. :ok

[Lardonio]

Re: Percentage staking plan, thoughts?

Not sure I know what you mean by that Lardonio. Don't think we can be talking about the same thing. Say I believe a selection has a 25% 3/1 (4.00) chance of winning. If it were top price of 100/30 (4.33) 23.1%, (25 - 23 = 2%) it is just 2% value (1.9% to be exact). If it were top price 4/1 (5.00) 20%, it is 5% better (25 - 20 = 5%). 5% value. If it were 11/2 (6.5) 15.4%, (25 - 15 = 10%) it is 10% better. 10% value (9.6% to be exact). Every point @ 100/30 returns 4.33 points. Every point @ 4/1 returns 5 points. Every point @ 11/2 wins 6.5 points. So it does make a big difference to returns.

We're not talking about the same thing. I'm talking about betting on varying odds with constant expected value, as this is how I understood the original question: "Should you really be betting the same amount on a 10/1 shot as you do at Evens?" If you place 100 bets at odds 100 and you win 1.1% (meaning 10% value) of them you profit 10. If you place 100 bets at odds 10 and you win 11% (meaning 10% value) of them you profit 10. If you place 100 bets at odds 2 and you win 55% of them (meaning 10% value) you profit 10.

[Hooloovoo]

Re: Percentage staking plan, thoughts? Interesting discussion guys. Are we talking edge = roi here ? The selections i use have an 107% roi, in that case my "edge" is 7%, right? Given that, would you recommend to: * bet the same percentage of current bank no matter the odds or * bet different percentages of current bank depending on odds bands? As i stated before, i am willing to take a higher risk, initially, when the bank is still "small".

[Lardonio]

Re: Percentage staking plan, thoughts?

Would've thought every sport is the same, with football it is a three horse race. Team A, Team B and the Draw. Study the form, work out what you believe the percentage chance of all three results are. Convert to odds and back the value selection.

My experience is that every sport is very far from the same. Of course, the mathematical principles of every game of chance are the same, but that does not mean that every sport is the same.

Bookmakers (odds compilers) work out a book "accurate" enough to only enable a fraction of punters to make a profit. Punters (in any sport) can use their "skill" to work out a market "accurate" enough to be one of that "fraction".

Of course ;)

The ability to "trust" oneself to grade value comes with experience and profit, once a punter is confident enough to find that value. :ok

Or, experience and profit leads to having enough confidence and cool to treat all bets the same to ensure maximum profits. :ok

[Lardonio]

Re: Percentage staking plan, thoughts?

Interesting discussion guys. Are we talking edge = roi here ? The selections i use have an 107% roi, in that case my "edge" is 7%, right? Given that, would you recommend to: * bet the same percentage of current bank no matter the odds or * bet different percentages of current bank depending on odds bands? As i stated before, i am willing to take a higher risk, initially, when the bank is still "small".

If your edge is 7% regardless of odds then you should bet the same percentage. If you analyze your bets and find anything interesting, for example you might be profiting on bets in certain odds 'bands' and losing on others, then you might do something different, dependant on how far you are willing to trust your 'data-mining'. Just remember any edge in gambling is a historical number and does not necessarily say anything interesting about the future.

[Hooloovoo]

Re: Percentage staking plan, thoughts? Thanks Lardonio. I find this subject to be complicated :\

[Gingertipster]

Re: Percentage staking plan, thoughts? Putting the same stake on an outsider as you would an odds on shot is for me not the right way to go. As it is (but not to the same degree) with your varied prices Hooloovoo, 4.00 and a 1.4. But with stakes it is often what a punter is comfortable with that matters. :ok

[slapdash]

Re: Percentage staking plan, thoughts?

Putting the same stake on an outsider as you would an odds on shot is for me not the right way to go.

I agree. I don't see at all why the fact the ROI is the same means that you should stake the same amount. In fact, in extreme cases it is clearly wrong. If I could identify bets at odds of 2.00 for which I was confident that the expected ROI was around +10% then I would be very comfortable to stake 2% or more of my bank. If I identified bets at odds of 101.00 with ROI of +10% then I would eventually go broke if I staked 2% of my bank on each bet. Deciding on stakes is about balancing expected profit and risk (or variance). Higher odds bets with the same expected profit for a given stake have higher variance, so to redress the balance between profit and risk it makes sense to reduce the stake.