Bert Willing wrote:

Your example is not working.

If there is no wind, a perfect thermal would rise vertically and you fly
constant circles to stay in.
If you have a constant wind, the whole airmass - including the thermal -
drifts with the wind. If you stay with your constant circles, you drift at
the same speed as the thermal so you stay perfectly centered.
Just basic vector addition.

Corrections are made because there is no ideal thermal, but corrections are
made into the core, regardless of the direction of wind.
Corrections into the wind are made if the thermal is of orographic, i.e.
rotors.


Nevertheless Helmut Reichman in his famous book says the same thing
as Mark James Boyd. Except he mentions also a case where you have to
do the opposite. You are right that the thermal drifts with the wind
but the glider sinks in the thermal. You may either figure the thermal
as an oblique column of rising or a sequence of bubbles rising while
drifting downwind and so connected by an oblique line. In both cases
sinking in the thermal will bring you below it, and in order to get back
into the column or the next bubble, you have to move upwind. The case
above mentionned where you have to do the opposite is the case of a
continuous column with a strong maximum, below which the lift is weaker
and converging, as well a weaker and diverging above it. In this case,
despite your sink in the thermal, it will bring the glider in the upper
part of weaker lift, and at this height the stronger lift is downwind.

A thing that Mark James Boyd made me discover is that the needed upwind
move is higher with lower climb speed, up to completely cancelling the
drift when the climb speed is zero.