Basics of money management

[Datapunter]

I've promised myself that each week I would do an article, (or part of) for my site. Here's the first. I welcome any comments, suggestions, alternative points of view you might have :ok

[Datapunter]

Re: Basics of money management

Basics of money management

Why do any money management? Well actually you don't need to do any money management at all. In sports-betting money management is not required. You can just bet as much as you fancy, or have available. Sometimes you win, sometimes you lose, and along the way you'll have loads of fun. Of course you will never actually know if you are losing or winning will you? In particular if you don't keep records. So if you don't care about ending up with a profit then indeed; money management is a complete waste of time. In fact I think you should stop reading right now. The rest of this text, probably the whole site, is going to be utterly boring to you. Thanks for visiting and good luck. . . . . . . . . Hello, still with me? Well, I guess you must care about losing or winning then. Because that is the reason why you want to manage your money; you want to end up in profit. Through money management you control the result you get from your betting. Even if you end up losing, through money management it will be you who is in control, not your circumstances. So let's have a look at the key parts of money management. Setting a betting bank First thing to do is set a bank for betting. Very simple, you look at your financial situation in life and determine an amount you are willing to use for sports-betting. And at this point you apply the golden rule of sports-betting. (or any other form of betting and gambling ) Only bet with money you can afford to lose. If you are currently in debt, or if you are in a situation where at the end of the month you have no money left, you should not be doing any sports-betting. You simply don't have any money you can afford to lose. You should concentrate on improving your overall financial situation first. But I'm not here to lecture... moving on. So if you have some money you don't need for your day to day living you can use that as your betting bank. How much is a very personal matter. I think you should pick an amount you are willing to lose. Not that you are actually going to lose it. (After all the intention is to make a profit) Why an amount you are willing to lose? Well i don't want to get all psychological here but that really is what it comes down to. Imagine you lose the entire bank, just for a moment, imagine that situation. How does that feel? As it is money you can afford to lose it won't, should not, affect your day to day life. Obviously you'll be devastated for a while and then pick up your life again. But can you live with it? Can you live the knowledge that you lost the entire amount you've set? If you can live with it then that bank is right for you. If you can't then that amount is too big, you should pick an amount smaller until you get to the point you can live with losing it all. Why is this so important? It is because it will affect everything you do as a sports-bettor. It creates a state of mind; it is the source of your success or failure. Key ingredients in being successful as a sports-bettor are discipline and objectivity. If your mind is at ease you can be objective. If your mind is dealing with worries then that will affect your objectivity and that will have a negative effect on your success. This is only one way of looking at it. There are many. This one works for me. I have been fortunate in life financially. I'm not rich but I have done well. My current betting bank is quite substantial. Let's say an average European annual income. I suppose I could live off my betting yet I still have a part-time job. It's a deliberate choice that fits entirely in this approach to setting a bank. If I was to lose my entire bank I would indeed be devastated for weeks. And then pick up life again. And I would be able to live with the knowledge of having lost all that money. I played the game and I lost. I wouldn't like it, but that's a different story. And it would not affect my day to day life. The part-time job allows me to pay the rent and put food on the table. What I get from all this is a state of mind that is peaceful. I've done my homework, made my choices; I can live with the worst case scenario, now let's do some betting. And be objective and disciplined about it. Keeping records Keeping records is part of good money management. Some people use databases, others spreadsheets and then some use something as simple as a notebook. Its all fine, whatever works for you. As long as the record is kept outside your own mind. The human mind is the worst of record keepers. We tend to remember only those things we want to remember. You should give your record quite some attention. It is your most valuable tool in being successful. Your record shows you where you've been successful and where you've failed. It allows you to learn from your past and adjust your future actions. In short: your betting record is your best friend. Establishing a stake size Now that we have a bank and some method of keeping a record we can look at placing a bet. I'm not going to talk about what to bet on. Bet selection is a different subject, this is about money management. So I'm simply going to assume you have something you want to bet on. But how much money should you put on an individual bet? At the extreme you could put your entire bank on. But then you risk losing all in one go. That's not what we want; we want the bank to last, preferably forever. So you need to put only a portion of the bank on an individual bet. How big should this portion be? This is mainly a matter of managing risk. How much risk are you up for? The more you are up for the bigger the portion of the bank you can stake on a bet. To determine the actual size of the stake we need to consider 2 things: the average odds you bet with or strike rate and the amount of risk you are up for. The odds imply a percentage of winning and losing bets. This ratio is called strike rate. Simply divide the number of winning bets with the total number of bets and express that as a percentage, which is your strike rate. For example if you have 100 bets of which 60 are winners you get: 60/100 = 0.6 expressed as percentage that's 60% strike rate. Or looking at it from knowing the average odds: if you bet with average odds of for example 1.85 you can calculate the implied strike rate. Simply divide 1 by the odds: 1 / 1.85 = 0.54 again expressed as percentage the strike rate is 54%. In other words if you bet with odds of 1.85 then out of 100 bets you may expect 54 to win and 46 to lose. The strike rate is important because it gives you an idea of the distribution of winners and losers. By this I mean what you can expect in terms of consecutive winners and losers. With low odds and high strike rate you are likely to have a series of winners and then a loser. You can use a fairly high portion of your bank because the risk of losing the entire bank on a series of losers is quite small. The opposite happens when you have high odds and a low strike rate. You can expect many losers and then a winner. That means you should only stake a small portion of the bank on each bet otherwise a series of losers might wipe out the bank quite quickly. This all comes together in the risk of going bankrupt. Again expressed in a percentage the risk of going bankrupt can be determined for each combination of strike rate (or average odds) and stake size. For a given strike rate the bigger the stake the more risk there is of going bankrupt. On the flip side is that the smaller the stake, the less risk of going bankrupt, but also the less profit you can make. It is a balancing act.

At this point I could get all mathematical but as this is called "The basics of" I'll leave you with a simple table. This table was established using a computer program that simulates betting. (Monte Carlo simulation) It shows the risk of going bankrupt for a given strike rate / average odds and stake size combination. It's the best starting point I can give you. Determine your strike rate or average odds, then determine how much risk of bankruptcy you are up for and the table shows you the size of your stake for each individual bet. TABLE GOES HERE

I'm doing the program with the simulation next week. And obviously the table also. So the above may change a bit Notice here that I have said nothing about profit. That is because the approach I take is a carefull one. I manage risk and then I'll take whatever profit I can get. There is another way to determine stake size. That is if you approach things with a target. You set yourself a target of profit that you want to make. And then you take, within limits, whatever risk is necessary to achieve your target. It is not that one approach is better than the other, both are valid. In the end it comes down to personal choice. I do believe however that before you get into betting towards a target you should have a good grasp of what it means to manage risk. And that means getting some experience with the approach outlined so far. So I'll say no more on the subject, you'll come across it again in the advanced sections of the site.

[kevsul]

Re: Basics of money management Very Very informative. Mr D.P. never used spread sheets or databases. just used to write down my starting amount at the begin of the month and see what i ended up with. I think i might take this approach in to the New Year with me.:ok

[slapdash]

Re: Basics of money management Well-written and full of good advice. One aspect I slightly disagree with is at the end where you talk about the risk of going bankrupt with a given stake. If I had a betting bank of 50,000 units and typically bet 1000 units, then if my betting bank dropped to 20,000 units after a bad run I wouldn't still bet 1000 units, I'd bet about 400 units. The question I first ask myself is not "How many units should I bet?" but "What proportion of my bank should I bet?". So a simulation showing how likely I am to go bankrupt if I continue to bet 1000 units is a little irrelevant, although I realize it's easier to explain than the mathematics of proportional staking. I may be the only gambler in the world who thinks not in terms of the expected gain in his betting bank, but in terms of the expected gain in the logarithm of his betting bank.

[PAULM03]

Re: Basics of money management Very well written DP

I may be the only gambler in the world who thinks not in terms of the expected gain in his betting bank, but in terms of the expected gain in the logarithm of his betting bank.

That's an interesting approach, and one that in my very limited experience have never come accross. What advantage does it give and how do you apply it?

[Datapunter]

Re: Basics of money management I get what you are saying slapdash, my point in the article is to give people a starting point, one that is sound and safe, it is after all "The basics" Your approach will pop-up when i get to 'Staking plans' Another point to consider is this: suppose your system goes tits up and you start losing. Suppose after a series of bets you see your bank slowly but steadily getting smaller. Will you actually continue until it is all gone ? I wouldn't. At some point i would say to myself: 'this ain't working, back to the drawingboard' So what's the point of setting aside a pot of money if you don't have the intention of using all of it in the first place ? The point is 'state of mind' . When i started with the Hot Favourites System this season i set aside a big bank. I applied all i have written in the article and came up with: 100points. But i did not put the entire 100 points in my Betfair account, there was no need, i put about half in and left the other half in a savings account in the bank. Suppose the system had gone tits-up would i actually have deposited the other 50% and continued ? Not very likely, maybe part of it, but not all. But when i put my first bet on, with a stake much bigger than what i had ever done before, i was completely at ease. I had done my homework, made my choices, come what may. That stake was more than my monthly income, which is a scary amount to risk on a game of football, and i was completely relaxed about it. Thats what you get from this, simple, starting point. p.s. Very nice to hear you think it's well written, enjoyed writing that :ok

[Datapunter]

Re: Basics of money management I see now something is missing, I need to mention the starting point is 'level stakes' , and there are other staking plans possible. :ok

[slapdash]

Re: Basics of money management

Very well written DP That's an interesting approach, and one that in my very limited experience have never come accross. What advantage does it give and how do you apply it?

Well, I was only being 75% serious. I don't think on a day to day basis about logarithms. But ... [i know I'm going to regret this ... some mathematics follows ...] Suppose that every day you have a bet available at odds of 2.20 where the actual chance of winning is 50%, so on average you will make a profit of 10% of your stake. How much should you bet on it? If you use a fixed stake, then there is a mathematical theorem (the "Law of Large Numbers") that says (roughly) that after a large number of bets your profit, as a percentage of the total amount you have staked, will be about 10% ... if you have not gone bankrupt following an initial unlucky spell. You might ask how much you should bet each time to make your expected profit as large as possible. That's easy, you should bet your whole current bank every time. Then after 100 bets you will have made an average profit of nearly 13780 times your original bank!!! The problem is that the expected profit is made up from making a profit of about 17500000000000000000000000000000000 times your original bank about one time in 1268000000000000000000000000000 and losing everything the rest of the time (and unless you're betting with Pinnacle you'd run into problems with bookie's limits if you did win!!). The problem is that because you are raising your stakes when you win, the Law of Large Numbers doesn't apply, and it's not true that in the long run your profit is likely to be close to the average profit: in the long run you'll almost certainly lose your entire bank. Keeping to level stakes has the problem that if you start with a small bank you will gradually build up your bank and eventually the amount you are betting will be very small compared with your bank and you could obviously make more money, safely, by raising your stake. One approach is to periodically reevaluate your stake taking into account, as DP hints at, the likelihood of going bankrupt given your new stake and your new bank. A mathematically more elegant approach is to constantly bet a fixed proportion of your current bank, so your stake continuously varies as your bank changes. Since your stake is changing, the Law of Large Numbers doesn't apply to your actual profit, and it won't be true that after a large number of bets is sure to be close to 10% of the total amount you have staked. However, each time you bet, if you lose your bank will be reduced by a fixed proportion, and if you win it will increase by a fixed proportion. This means that the logarithm of your bank will decrease/increase by a fixed amount when you lose/win, which means that the Law of Large Numbers will apply to the logarithm of your bank. Now you can do a calculation to decide what proportion of your bank you should bet each time to maximize the expected increase in the logarithm of your bank. This is an A-level exercise in calculus and will give you the Kelly formula, which in the example above says that you should bet 10%/1.2, or about 8.33%, of your current bank each time. If you bet more or less than this, then your bank will grow less quickly, and if you bet too much more than this then your bank will almost certainly decrease in the long run. [WARNING: This is very sensitive to overestimating the actual probability of your bet winning, so it is prudent to bet a smaller proportion of your bank than the Kelly formula suggests.] So far this doesn't require you to think about logarithms. You could just take the Kelly formula and not worry about how it is derived. But as you may know, I've done a lot of each-way betting on horses, and even placing each-way doubles on horses, which makes things more complicated, and I actually have done calculations with logarithms to get some idea of how much I should be betting.

[DeepBlueDevilFish]

Re: Basics of money management Good article DP. Re Slapdash's point on proportional staking. I've found that using this form of staking will quite often result in more profit at the end of the day but that you may well take longer to get there (because it takes longer to rebuild profits after a losing run). This time aspect to staking can sometimes be over looked. For example with DP's Soccer Hot favourites system I think the chance that you won't be in profit after about 200-250 bets using proportional staking is almost double that of a level staking plan.

[slapdash]

Re: Basics of money management

Re Slapdash's point on proportional staking. I've found that using this form of staking will quite often result in more profit at the end of the day but that you may well take longer to get there

Get where? I'm not being flippant, it's a serious question.

For example with DP's Soccer Hot favourites system I think the chance that you won't be in profit after about 200-250 bets using proportional staking is almost double that of a level staking plan.

That depends on what proportion and what level stakes you use. If your aim is really just to maximize the chance that you're in profit (not worrying how big the profit is) after a given number of bets, then you should start with a tiny stake and use some kind of Martingale strategy, reducing your stake to zero as soon as you're in profit. If you use a strict fixed stake, then there is a non-zero probability that you will go bankrupt. Most people who espouse "fixed stake" betting actually use some kind of "plateau" staking, increasing their "fixed" stake when their bank increases enough (and presumably decreasing it if their bank decreases to a frightening level). Proportional staking is just a smoother way of doing this.

[DeepBlueDevilFish]

Re: Basics of money management

Get where? I'm not being flippant, it's a serious question.

Get into some sort of target profit situation. With a proportional staking strategy it's quite possible that on a very bad losing run your stakes could go so low that's it's hardly worth trying to get back to your starting point again (i.e zero loss / zero profit).

That depends on what proportion and what level stakes you use.

I was assuming DP's 10% level stakes there which he states he uses at or near the beginning of his thread.

If your aim is really just to maximize the chance that you're in profit (not worrying how big the profit is) after a given number of bets, then you should start with a tiny stake and use some kind of Martingale strategy, reducing your stake to zero as soon as you're in profit. If you use a strict fixed stake, then there is a non-zero probability that you will go bankrupt. Most people who espouse "fixed stake" betting actually use some kind of "plateau" staking, increasing their "fixed" stake when their bank increases enough (and presumably decreasing it if their bank decreases to a frightening level). Proportional staking is just a smoother way of doing this.

Yeh that's all true and works fine when things are going OK and your in profit or not making much of a loss. But would you really continue with a proportional staking strategy if you've lost 50% of your bank ? Is it worth your while with a much reduced and reducing chance (if your bank drops further) of getting back to a profitable situation ? What would you actually do in this situation ?

[slapdash]

Re: Basics of money management I made my last post just before going to bed and may have been more than usually cryptic. Apologies to DP for hijacking his thread: this always seems to happen when I get involved in a thread about money management or staking plans.

That depends on what proportion and what level stakes you use.

Actually, it doesn't depend on the level stakes you use, though the chance of not being in profit after a given number of bets using proportional staking does depend on the proportion you use (the smaller the proportion, the closer it is to level stakes). It's true that using proportional staking you are more likely not to be in profit after a given number of bets (although typically, depending on the precise parameters, you will also be more likely to have made a 100% profit and less likely to have lost half your betting bank). But the question of whether you are in profit seems to me to be concentrating on what is really just a psychological barrier: if your betting bank really is a betting bank, and not money you need for another purpose, then the 0% profit level is really no more significant than the 10% profit level or the -10% profit level. My point about Martingale was not entirely flippant, either. Martingale staking is the extreme version of chasing your losses by increasing your stakes: it will probably get your money back, but if it doesn't then you're in real trouble. If you use strict level staking and choose your stake so that initially you have, say, only a 1% chance of going bankrupt, but then have an unlucky spell and lose half your bank, then if you continue with the same level stakes you now have a much higher chance of going bankrupt, so you are increasing your risk by continuing. That's the only reason you (usually) rebuild your profits quicker after a losing run using level stakes than using proportional staking: because you're using a riskier strategy.

[DeepBlueDevilFish]

Re: Basics of money management It's probably not feasible to just dismiss the psychological aspect of gambling here. Everyone is going to be happier in a profit situation than a loss situation whether or not you need your betting bank. If you believe you've got a system that is going to be profitable in the long term you're probably going to be happier knowing that you'll recover quickly from the inevitable losing run rather than flounder around in a negative profit state for a longer time. Yes you could go bankrupt using fixed stakes but in practical terms this won't happen - you'd probably abandon the system when you hit your own personal "loss" threshold so the threat of bankruptcy isn't really there. Incidentally fixed profits staking may actually be the best strategy to use with DP's system (as you reduce the risk on the higher odds) Oh and Martingale is OK with an infinite bank. :loon And apologies from me too DP. :$

[Sgt.Sunshine]

Re: Basics of money management Interesting discussion guys. For me it all boils down to the psychology of dealing with it on a week to week basis. Like you, I guess, I've done all sorts of simulations testing out different staking plans etc. and the proportional staking (Kelly or same amount on each bet) always looks great on paper - big numbers, big profit etc. But I looked at the possible fluctuations in bank and therefore stake, after a winning or losing streak and always asked myself if could stick to the plan in the long run. Paper trading is easy, doing it for real ain't. I gave it a go and it took me about a month to lose my nerve completely :lol. I feel much more comfy with a level stake as a % of how my betting bank stands at the beginning of each season. Enjoy the winning streaks, and don't worry too much about the losing ones (which will inevitably follow at some point!). One problem I think is that most staking simulations assume a steady edge, and so while useful for discussing theory are in most cases removed from reality. Interesting stuff atbout the logs slapdash btw, hadn't thought of it like that :ok.

[slapdash]

Re: Basics of money management

But would you really continue with a proportional staking strategy if you've lost 50% of your bank ? Is it worth your while with a much reduced and reducing chance (if your bank drops further) of getting back to a profitable situation ? What would you actually do in this situation ?

It's happened. And I continued and recovered. These days I'm more conservative with the proportion of my bank I use, but it's still not unusual for me to lose something like 20% of my bank in the course of a month (luckily it's also not unusual for me to gain 25%). If I have a much reduced and reducing chance of getting back into profit, then that means the bets I'm making are not good ones. If the bets I make are good then I know that I will get back in profit in the long term.

[Datapunter]

Re: Basics of money management Apologies ? I'm not complaining, good stuff, keep it coming :ok I think Sgt.Sunshine hit's the nail on the head, in the end, after having done your homework, you need to pick a staking strategy that's right for you. And different people will pick different strategies.

[Datapunter]

[slapdash]

Re: Basics of money management Is this assuming you are getting fair odds? So for a strike rate of 80% you are getting odds of 1.25, for example?

[Datapunter]

Re: Basics of money management Yes, when you just start out in sports-betting, or you start a new project, you don't know if it will be successfull. So the calculation is done with fair odds, or in other words a yield of 0%, not winning, not losing. Thats your starting point. ( need to add that to the text somehow ) Obviously a negative or positive yield will shift the table quite a bit, but thats another article :lol I may do one with odds -10% representing the bookies overround, see what the effect is.

[Datapunter]

Re: Basics of money management Sorry about the formatting, to get it right i'd need to install a higher version of Excel :$ So, couple of points you need to know: - Odds used are based on Strike rate, so yield = 0% - When the bank is smaller than a single stake, it is considered bancrupt. - Risk of going bancrupt increases with the number of bets in a run, the longer the run the more chance of running into a losing streak. - Each run of bets was done 100000 times, accuracy is up to the 3rd digit, so if you see for example 32,5 then thats anything between 30 and 31 - When a run is bancrupt it is discontinued, it does not get a chance to recover. Another reason why long betting runs show a higher risk.