Low Risk – High Reward Staking


You may have seen the table below in previous blogs, but I make no apologies for repeating it. 

The connection between strike rate and losing runs should constantly be borne in mind. This table is a graphic depiction of why.

Strike rate and losing runs

It is  important  to remember  that the larger the number of bets, the longer the probable losing run will be.

So it is important to consider how many bets your system/method/ tipster provide.

If it is less than a thousand a year, you will probably get away with betting treble the proportion of your bank shown above, eg, 50% strike rate, use 30 point bank (3 x 10 max losing run). 

Psychology and staking

Never forget the psychological angle to all this. Most people who see their bank anywhere near half will panic and give up.

If you are backing more than a thousand horses  a year you will need to follow the advice I give below.

As an example, with a 25% strike rate in 1000 bets you are likely to hit a maximum losing run of 24.

However, if you were to have 5,000 bets with a similar strike rate, you will likely  hit a much bigger losing run .

Basically, the more bets you have the more likely you are to hit an horrendous losing run (same goes for a whopping win though)!

In that case, it’s useful to have an idea of the number of bets your selection method is going to produce  in a given period.

If we have a method that produces 300 bets a year, calculating your longest  losing run over 1000 bets is as above.

If we have a method that produces 5,000 bets a year,we need a more cautious approach.

If we are looking at a method that gives 1000 bets a year with a strike rate of 50%, the table tells us we are likely to have a losing run of 10 in that period.

We could choose a 10% staking plan (upping the stake proportionately after a winner and reducing it in the same way  after a loss). If we hit the worst run, amongst some lesser losing runs, we might give up.

Bank Size

We need our bank to be significantly bigger than the size of our longest likely losing run.

Over such a large number of bets, it is quite probable that the worst losing run would be a fair bit longer than in the table.

It is also possible that two long losing runs could virtually follow each other!

To be safe, we set our bank at 4 times the longest probable losing run shown in the table.

With  50% strike rate, it would be one fortieth, or a 2.5% rolling bank (increasing and decreasing stakes according to results, as described above).


First of all let's set an amount for our bank. And let's use the figures from our latest system, the one we used last time.

Our historical strike rate has been 27% for the win only bet and 49% win and place.

Given that this blog is about each way betting (49% won or placed) that means, referring to the table, our longest probable losing run is 10. 

Multiply that by 4 for insurance and sanity!

If our start bank is £200 that means the first bet is 2.5% (one fortieth) of  £200 = £5.

Stakes, low risk – high return

Now we come to  the agonising decisions of each way betting. Do we half the total stake or double it?

Well let's look at what we would do if we were backing win only. Our bank is £200. With a 27% win strike rate we could expect a losing run of around 25 once in every thousand sequences. 

Divide the £200 bank by 100 (4 x 25) and the first bet is £2

If we split that stake, that would give us £1 each way as the first bet.  £2 is 1% or 1 in 100 of the bank.

When you consider our new system provides about 700 selections a year, and the longest probable losing run is 10 this seems safe.

We could be considerably more aggressive. But we don't need to be. Always play with money won off the bookies is my maxim.

We use HorseRaceBase for all our system research. Unfortunately, they don't do a breakdown to show what exactly would happen if  we backed all the selections from our new system each way system.

Percentage level stakes

They do though show what would happen backing just the win part of the bets, starting with a £100 bank. Your first stake  would be a pound then you would increase and decrease your stake according to  the results.

Eg, first £1 win bet wins at 2/1. New bank = £102. Next stake = £1.02 win.

First bet loses. New bank 99.00. Next bet = 99p.

Here is a screenshot of what would happen, at ISP, if you started with £100 on 1/1/16:




The latter screenshot shows the end of year bank balances.

Of course notwithstanding its credibility (though who am I to argue with the always spot -on HRB?) the final chart is impracticable. Though just year one would do me fine! But it illustrates the power of compounding.

It also shows the possibility and importance of taking some out of the betting bank as and when. For example, I would have been happy to start 2017 with a 4K bankat that rate of growth!


Compounding and a misconception about level stakes

There seems to be two understandings about what level stakes means.

One conception is that you stick to the same stake all the time. So, using our above bank and the same staking, we would start with our £100 bank and simply back £1 win for the rest of our lives! After all, one hundredth of one hundred is one. So that's our level stake.

The problem with that approach is that, if you have any success, you will be backing to a smaller and smaller percentage of your bank. For example, if you double your bank to £200 your stake will be just one two hundredth.

The other view, the one we take, is that it should be a level stakes percentage. So that if your bank doubles, so does your stake. If it halves, your stake should too.

We do it in the way show above, level stakes percentage betting, so that, incrementally, as the bank increases, so do our stakes. That way we don't under bet the winning runs.

Equally, the stake reduces bit by bit on the losing runs so as not to over bet them.

This is the professional approach to bank size and staking which is both low risk and offers potentially very high rewards.