You might like to try this progression scale.

With small div results a compounding progression works best.

The idea is to inrease in stages of 2 lots each time.
e.g.
1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10 etc

If you strike a winner that does not end up showing a total profit , repeat that bet.
If striking a looser , follow the progression

If you can strike winners that average out at half the odds required to break even , you will either be in front or have broken even.

e.g. In the run shown above , this totals 110 units outlayed for a run of 20 .

At level stakes , you need a 19/1 winner to break even or winners adding up to that ,e.g. say you had 5 wins in a row adding up to 19/1 , but with the progression you would only need approx half those odds to break even , in this case 10/1.

Stop once you have broken even or in front ,before starting at the beginning.

Its a good idea to start at the beginning once you have broken even ,then start at the beginning ,because its amazing how the wheels fall off , just when you think you are on a winning sreak.

Its best to have 3 banks of 110 =330. if the 1st bank goes belly up , your 2nd bank will have double the outlay on each bet e.g. 2,2,4,4,6,6, etc. stopping once more, once that series has broken even or shown a profit.

You will find all staking plans rely one factor , the ability to pick winners in the sequence you are in , thats the tricky bit .

If you stop betting once you have 6 outs in a row , then commence again once a winner gets up , it will give yo more confidence to follow a progession plan like this & hopefully avoid the 20 loosers in a row , or in this case 3 lots of 6 loosers in a row.

__________________
Cheers.