The average punter is fighting a tough battle vs the bookmaker largely as a result of poor money management. Bookmakers can from time to time present very good prices, especially after the bookmaker has had a few good weeks (often weeks with surprise results), and punters' accounts are running empty. Bookmakers know that even if they present a few good odds (I mean really good odds), average punters will most probably deposit more money, and most certainly win this time, thereby increasing their bookmaker account.
This is fine by the bookie, because he knows that the customer most likely will have an empty account in a few weeks time due to his poor money management anyway. The bookmaker will see this as a short term investment, because the average punter will feel aroused by his winnings, he's getting eager, and start to bet more often and often with higher stakes. Unevitably, the bookie's getting his money back, with a solid interest on top.....
Professional punters know that bookmakers operate this way, (presenting good prices every now and then, get money circulation, let the customer feel he's on top, thus generating more cash flow, and eventually more money in the bookmaker coffers), and therefore they wait patiently for the next round with extra good prices offered.
The professional punters do not get carried away after a few wins. They know what to expect, and act thereafter. The professional punter is not necessarily better to predict an events outcome than other punters , but he's definitivly much better with money management
There are 3 systems for money management which are widely used by punters: Martingale, "Row of numbers" (a.k.a "Die Abstreichmetode"),and the Kelly criteria
Ofcourse, these systems are known under other names, but the function should be the same no matter the name.

Martingale
Martingale is probably the best known system for money management. Originally designed for use in casinos, this system has nothing to do with picking objects. Martingale is a system which helps you control your stakes. With Martingale, you can be a lousy bettor, but still be successful. Please note that Martingale is a system with very high risk of going broke.
The principle of the Martingale system is as follows:
If you lose, you double your stakes, and if you win, you start all over on your starting stake. This means that you eventually will be ensured a profit. Follow this example: You've found a game with odds 2.0 for home victory. You bet 100$, but you lose, as the game ends in a draw. Next time you bet 200$ on a game with odds 2.0. If you lose again, you must bet 400$ on a game with odds=2.0. If you win this time, you've placed a total stake of 100+200+400 =700$, and you've 100$ to the good for your efforts. The 100$ payoff is equal to what you would've gained on your first bet.
Good system, or what ?
Certainly, if you had unlimited funds available. Or if the bookie would drop his maximum bet limit. Or if you have the sufficient cool when you hit a long bad run. Remember, with odds=2.0, you will have to wager 32 times your original stake if you lose 5 times in a row. And losing 5 times in a row is not too difficult with odds 2.0..... And even if you have lost 5 times in a row, you have no greater chance of winning the sixth time. Too many people think they'll have to win soon when they've lost many in a row. The fact is that the chances for a win is just as big (or small if you like) as when you started the betting sequence. Martingale is in other words a very dangerous system, and should be used by punters with large funds and no nerves. In my opinion, Martingale is not a suitable money management system for the average punter due to it's heavy level of progression

Row of numbers
The German word for this system is "Die Abstreichmetode". This is also a system well known from casinos, and has a lot in common with Martingale (recovering from losses by increasing stakes). But Row of numbers is not as tough as Martingale though. While Martingale can quickly give you your stakes back including a profit, Row of numbers does not repay your losses that quick, but it does not lead to big stakes either (which Martingale does).
Row of numbers gives you more flexibility compared to Martingale because you can adjust your stakes in a more suitable way. The disadvantage is that this system requires more effort from you, as you have to make a row of numbers on a piece of paper and add numbers when you lose, and remove numbers when you win. You need to have better control with your betting, so to say.
Let's have a look at the system:
First decide how much you want to win. E.g 1000 $. Then you have to figure out how long it's going to take to reach your goal. This is the tricky part, because when you win you must remove the first and last number in the row, but when you lose you must add a number to the row.
As a result of this, you must try to decide an average probability for a win, given a fixed odds. If you want to play on games with an average odds of 2.0, set your probability to 40% (or thereabouts). It's better to underestimate rather than overestimate your abilities. If you overestimate your ability, you will have to make up for it by increasing your stakes, and increasing the stakes can easily become rather unpleasant if you hit a losing streak.
If we imagine the 1000$ split into 20 50$ wins, we can calculate how long you will have to bet in order to get a 1000$ total win, given the 40% chance of winning with odds=2.0 (a 50% chance means you're betting even with the bookmaker).
With a 40% chance of predicting the 2.0 object correctly, you will lose 60% of the picks. Thus you lose 50% more often than you win (60/40). In the case of winning, you get to remove 2 numbers on the list (the first and last number). When you lose you must add a number at the end of the list. If we draw this row of numbers, it would look like this:
50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50
Determining your first stake:
As explained above, the first and last number will be removed if you win. 50+50=100. In order to win 100$, you must use a stake which will give you 100$ in net profit. If the odds given is 2.0, you'll have to bet 100$. If the odds is 1.50 you must bet 200$. If the odds is 1.75 you must bet 133,33$. The method of determining the stake is:
Net winning

odds  1
The stakes for your forthcoming bets can be calculated in the same way. If the first bet is 125$ at odds 1.8, and you lose this bet, you must add 125 at the end of the list. The next bet will then be the first and last number in the row, which now is 50+125, divided by (odds1).
50 + 125

odds  1
You must ofcourse substitute "odds" with the actual odds offered on your object.
If you win, you must remove the first and last number in the row, and then start fresh with 50+50/(odds1) again.
Even though this system does not produce such a high level of progresion as Martingale, it's still a rather dangerous system. If you hit a losing streak, things can easily get out of hand. Therefore, your prebetting calculations is of highest importance. Before starting with this kind of system, be sure to run a test period with fake stakes for a month or so. You will learn the system, and see the importance of modesty when determining your own predicting ability.

Kelly criteria
Systems like Martingale and Row of numbers use high level of progresion to make up for the punters lack of margins. In these systems, stakes are successivly increased when losing, thus the punter are running a high risk of bankruptcy.
In the Kelly Criteria, progresion increases when you are winning, and decreases when you are losing. The stakes are decided by a percentage of the size of your funds. In the Kelly Criteria, the risk of bankruptcy is virtually eliminated.
The Kelly Criteria requires that the punter have the probabilities on his side. When using Kelly, it's expected that the punter can bet even with, or better than the bookmaker. If a home team has got odds = 2.0, you will only bet on the home team if you think it has a 50% chance or more.
The Kelly Criteria has got it's name from John L. Kelly, the American who invented this theory and formula.
In short, Kelly's theory says that if you can determine a somewhat correct probability for an events outcome, then the formula will determine the exact amount of your funds which you should bet on that event.
If you overestimate your ability to predict an outcome (i.e you predict a 60% chance, when the correct prediction should be 52%), you will pay for it by losing your bets. If you underestimate your ability to predict the outcome (i.e you predict a 55% chance, when the chance is 60%), you will win your bets, but not as much as if you were betting with flat stakes. This is due to the fact that Kelly's formula optimizes your stakes if you are able to predict with a high degree of accurracy.
With Kelly you stand a better chance of winning your bets by having only a small advantage on every game you pick. If you instead have a small disadvantage for evey game you pick (i.e you overestimate your predictions), you will have a lesser chance of coming out on top compared to flat stake betting
Before starting with Kelly, decide upon the following:
Size of your fund
A fund sized 1015 times the size of your normal singlebet is enough. Ofcourse these funds must be funds you can afford to lose. Remember that with the Kelly Criteria, you will not lose all your money straight away, because the stakes are decided as a percentage of the actual size of your fund.
How often, and how well you are able to pick valueobjects
Experience is vey important in the art of sportsbetting.
Picking value objects correctly, and often enough, is ofcourse the trickiest part. Only by gaining experience you can become a clever punter. The bookmakers do not offer a lot of value objects, so you have to shop around with several bookmakers, and look for objects with a high spread. High spread means that bookmakers have different views to what is the correct odds on a match. Use services on the Internet to compare odds from several bookmakers.
Bookmakers rarely offer many value objects during a week. Do not expect to find value in more than 25 events during a week. When you find an object which you think has got value, check this event carefully. If the odds presented is 2.0, the team MUST have more than a 50% chance for victory, because the size of the stake is directly related to the winning probability (which is your subjective opinion).
Lenght of the project
For how long will you continue to play with this system?
If you have set yourself a goal with your betting, finish the project when your goal is reached, and start all over again. This way you get to realise the money you've earned, and thereby strenghtening your moral and dicipline. If you've not set yourself a goal, reset your fund when it approaches 100% payoff. Remember as long as your money is in the bookmaker account, it's the bookmaker who possesses the money, not you. And what's the point of earning when you cannot use the results of your hard work as you please ? Therefore, decide when to reset your fund when you have reach a predefined amount, and start it all over again. Make your earnings visible.
Lets examine the system closer:
The biggest advantage with the Kelly Criteria is that you will lose less money when your fund is low. This is due to the fact that your next stake is a percentage of the fund's actual size. When the size of the fund is low, your next stake will also be low. If your average stake is 10% of your total funds, then you can lose 6 times in a row, and still have nearly 48% of your funds left (take 0.9 and multiply with itself 6 times).
I won't bother you with advanced statistics (infact I'm not so sure I understand it myself either...), but if you've got the probabilities on your side (you bet with a 10% advantage on the bookie every time), the chance of losing 10 times in a row (at odds=2.0) is 1/3000 !!!
Kelly is not a system which provides rapid changes to your account. With Kelly you make money by having the required little margin, the extra 5% on each game. If you have the margins on your side, Kelly makes your stakes increase. If you lose, stakes will decrease. As a result of this, you will not experience great changes to your account. Kelly is a system for punters who do not bet jut for the sake of money, but also for the sake of their own satisfaction (when the prediction turns out to be a good one).
Conclusion
If you are a punter with a high risk profile (you like to gamble), and you know that you rarely predict better than the bookmaker, then Martingale, or Row of numbers, could be your money management system. But be ware of the dangers represented with these systems. Their progression rate can be devastating, and can take the mickey out of your projet long before it's even started.
If you have a low risk profile (you consider betting to be an investment with slightly more risk than a stockfund), and you know you are able to predict slightly better than the bookmaker, then Kelly is your system. Kelly is the system of the profesional punter, a system for punters who thrive for perfection. A system who optimizes the handling of your funds, if, and only if, you can predict better than the "enemy", i.e the bookmaker.
If you're looking for excitement during a televised match, then there is nothing more exciting than having a bet on spread. Read about "Spread betting" in the next article.
Parts of this article are inspired by Geir Høistads book "Oddstipping som levebr?d" (in Norwegian only). His excellent book can be purchased by mailing oddsgeir@hotmail.com