One of the quickest ways to disaster is to be ignorant about losing sequences. They can occur without warning at any time, and they can quickly devastate a betting bank if it is of insufficient size to absorb them.

First of all, how do you know what the length of these losing sequences is likely to be? Well actually, that is fairly easy to predict, provided you have an accurate idea of your strike rate.

Sequences are directly related to strike rate, and you can mathematically determine the likelihood of specific sequences appearing, provided you know what percentage of the bets in a series are going to win. You can then use this criteria to set up your betting bank.

My view is that the size of your opening bank should equal the longest losing sequence you could encounter during an extended series of bets, multiplied by your maximum stake.

To illustrate, I have produced the table below, which you can use to determine your staking strategy. Let's say, for example, you estimate that your strike rate over a series of 600 bets will be around 40%.

Please note at this stage that there are no 'good' or 'bad' strike rates - they are all purely relative to the price range you decide to target - please refer to my article How Many Winners? in conjunction with this procedure (opens in a new window).

As can be seen from the table below, with a 40% strike rate you can expect to have a losing sequence of 10 somewhere in the series (there is a 75% chance of that happening). However, there is a 50% chance of your hitting a losing sequence of 12 and a 25% chance of a run of 14 losers. That sequence of 14 is less than likely to occur, but it will still happen, on average, once in every four series.

On the other side of the coin, you have a 75% chance of hitting at least one winning run of 6 during the series. A sequence of 7 is a 50:50 possibility and a successful run of 8 has a 25% chance of occurring. Those figures can be seen in the 60% row (take your strike rate away from 100 and look in the row for the difference in order to anticipate winning sequences).

Finally, with a less than 1% chance, you are likely to confront a losing sequence of 20 no more often than once in every hundred series. However, that is the figure you should use in constructing your betting bank. If you are to operate professionally that is what you must do.

Using the formula I mentioned earlier, you would multiply the longest losing sequence you could encounter during the series (in the above example that would be 20) by your maximum stake. If that is, say, 5 points, then your starting bank should be 100 points.

Please don't make the common mistake of thinking these sequences won't happen to you. They are mathematically determined by your strike rate and will happen as often as they are entitled to, whether you like it or not. Nobody can avoid them, but the difference between a professional and an amateur is that the professional is fully prepared for them. When they do occur it certainly doesn't mean that you are 'out of form' or doing anything wrong.

It is also recommended that you set aside a reserve bank which can be used to temporarily top up the main bank in extreme circumstances. For example, a downturn which is over and above any which had originally been catered for (perhaps the original strike rate calculation was based on insufficient data and needs amending).

Once you have accurately anticipated your strike rate, you will be able to use the table below to optimise your bank structure.

Provided the average price of your winners is sufficient for your strike rate, as explained in How Many Winners?, and your bank is set up in such a way as to absorb sequences, then you will never have a problem. You will be well on your way to securing a successful future in a professional manner.

Please also see the Banking And Staking page.

Steve Jones

Professional Betting Advice and Strategy

Please note: The table below is based on a series of between 600 and 650 bets. A shorter series would be less likely to contain sequences as long as these and a longer series would be more likely.