Which to use, a Constant or a Variable Bank

Bankroll management is of critical importance to successful gambling.  It is one thing to find an edge but it is another to maximise the rate of growth of your bankroll.  If you are betting with an edge, you need to decide on a bankroll management strategy that will squeeze every last bit of profit out of the approach while still not causing you any undue concern.

  Recently in the Renegade forum, the debate of constant versus variable betting was raised.  A constant bank is to keep the bank size the same each time you bet.  Regardless of winning or losing runs, you always stake as if you had the same amount of money.

  Opposite to a constant bank, a variable bank is when you adjust your bank size in relation to the size of the bank.  A punter may adjust the bank once every six months, once a month, once a day, or even after every bet.

  As you win, your bank is increasing resulting in larger amounts bet on each selection.  Conversely if you are losing, your bank size is decreased and thus your bets amounts would be less.

  There is much debate as to the correct method to use.  This article will provide you with an answer to the posed question.

  You need to utilize your bank (total sum of money available to bet), to the fullest extent while still betting at a safe level.  As you win, your profits rise and thus your bank rises.  It stands to reason that if your bank has grown and you haven't changed your bank size then you are betting too small for your bankroll.

  If you had $10,000 sitting in a bank account earning next to no interest, wouldn't it makes sense to put that money into play by investing in a managed fund with a long history of positive returns or something similar?

  Most constant bank advocates say that after a certain amount of time, they will adjust their bank. But this is essentially the same as betting with a variable bank. So why not let the bank move after each week, or day or even bet?

  Let us consider what will happen if we have a true variable bank and constant bank. Firstly if you have a punter that doesn't bet with an edge and shows a long-term loss, then you will find that a constant bank strategy will soon fall bankrupt well before the variable bank strategy. In fact, it's technically impossible to go bankrupt using a variable bank.  In reality however, there will come a point if you are consistently losing that you will be betting such a small amount that it would be impractical to bet.  If you are a gambler who doesn't win then a variable bank is the way to go. Actually, either don't bother betting at all or if you do, keep the stakes down!

  If you are a gambler that generally breaks even then with constant betting, you will simply break even.

  With a variable bank however, there is a very good chance that you will lose some money; nothing drastic, but enough to conclude that using a constant bank is more beneficial if you are the type of gambler that tends to lose over the long term. Of course, if you are one of these gamblers, it wouldn't be worth your effort to bet in the first place unless it was of course a pleasurable hobby.

  And finally, if you are a good gambler and make steady profits, you will find that using a variable bank is considerably better than using a constant bank. To prove this is the case, I've completed 1,000 simulations so the results can be analysed as shown below.

 The constant bank had a starting bank of $1,000 and $50 (or 5% of the bank) was wagered on each bet at odds of 1.95.  There are three different scenarios; one where the tipster on average predicts 49% correct, two, where the tipster predicts 51.3% correct and three, where the tipster predicts 54% correct. After 1,000 simulations we averaged the results as shown above. The gambler with a variable bank bet 5% of their current bank amount.

  As you can see, on average the initial $1,000 bank was quickly diminished when only correctly tipping 49% correct. The fact that the constant bank had lost more than the variable bank suggests that using a variable bank is better if you don't bet with an edge. Similarly, when a tipster is 'average' the constant bank method lost less than the variable bank although the difference is marginal.

In both situations, a negative result means it is not worth betting anyway.  The third scenario is the only one that has proved to provide a profit.  Amazingly, using a variable bank is on average, 5 times better than a constant bank. The difference is significant and remarkable.

  Out of interest, you will see the difference between a constant versus a variable bank betting the Renegade Method below.

  Between the 4^th of June and the 12^th of November, a $1,000 starting bank betting constant has increased to $3,304.00 while betting variable has increased to $5,105.00.  This is a massive difference.  The results above only include betting Saturdays which is only 22 days over the results period. 

  In the short-term, constant may well outperform variable as this is dictated by the short-run sequence of winners and losers.  That is the reason while constant outperformed variable in half of the subscriptions offered on the Punting Ace website in 2005.

  This is one of the instances that gambling perception can let you down.  Many punters will argue that constant is preferred over variable.  The incorrect reasoning is when you strike a few good winners in a row your bank will increase thus your bet size, and then you will strike a losing run and give back all the profits and more.

  In the short-run, this can and does happen and is the reason why betting a constant bank can yield greater results than betting with a variable bank.

  One method many punters use to combat this short-term fluctuation is to split their bank into two banks; a constant and a variable bank.

  If you are a successful gambler, like the tipster who tips 54% correct, then the more you bet the greater the profits. This is of course only true when you don't bet ridiculously greater than what you should. In the above example, we bet 5% of our banks, but the more we increase this, the quicker the constant Kelly method of gambler parallels the variable bank strategy. For example if you were foolish enough to bet 20% of your bank on each bet, you will soon 'tap out' betting with a variable bank.  It is possible that you may get lucky and make good profits using a constant bank.

  But of course, the proof is there. If you have proven a long-term edge betting with a variable bank is the way to go.  You will be betting the best of the bank compounding affect.

By Matt Elliott

This article is protected by international Copyright © Elk Publications Pty Ltd February 2005 Please contact if you wish to reproduce this article elsewhere.