There ain't no ideal thickness-
Start with a rolling pin
(a cylinder shaped wing)
The lift is almost nil but the limit to angle of angle attack is ALSO also nil
smash the rolling pin to be 1/2 as high as it is wide
now we get lift at some angle of attack but still the angle is not critical
Also the lift possible is pretty low
Obviously - there is a relationship between chord thickness and critical angle of attack.
So- from cut n try (experience) most modelers have found that a airfoil shape of 10% to 20% is a workable parameter
It also (very important) allows for a light structure
Best of both worlds for a strong ,light, slow speed aerobatic airfoil is a fat one.
Once we get wingloading waaaaay down we find that the thickness can really be reduced -much further than seems practical
I have used airfoils which are only 2-4% thick! (flat Depron foam) and the shape does not stall out unpredictably. Itis simply because the craft is so light and strength is adequate..
These shapes also "fly" predictably when fully stalled - (tho stalled they are maneuverable )
Sounds impossible but in the area of super low wing loadings the rules can be badly bent.