Counters are playing to a 1 1/2% advantage.
Non-counters are playing to a 1/2% disadvantage.

Both groups are expecting their game to corralate to the stated odds in the long term.
In the short term, if there is any difference between the actual results and the odds, the
difference is calculated as variance.

Counters play and wait for their advantage to materialize.
Non-counters can't wait, they must overcome more than the 1/2% disadvantage in order to
make profit.

One method is to keep all bets low until the odds are satisfied.
How do you know when the odds have been accomodated, maybe the amount you're down
is just negative variance? Your record keeping will give you a good indication. You know,
ahead of time, how many units you should be down after 100, 200, 300 or 1000 hands etc..
It's all one continuous session with breaks of 10 minutes, 10 hours, 10 days or 10 years.

If you allow for a 2% disadvantage, when the actual disadvantage is only 1/2%, you can play to
to an expectation of 1 1/2% which is actually variance.

A good visual analogy would be a pendulum swinging back and forth, with your mean or centre point being set at -1/2%. On average, 1/3 of the time you'll be left of centre, 1/3 of the time
around centre and 1/3 time at right of centre.

I suggest you use a flat bet progression, as long as the pendulum is left of centre (negative variance) until you've recovered all the negative variance or until you show a satisfactory profit.
While you're letting the negative variance build