Be carefull when comparing 2D with 3D aerodynamic data. The angle of attack will be higher in 3D (over 20°) than on 2D, but the cLmax is smaller, because the slope is lower. And take the right Reynoldsnumber (e.g. 500000) and turbulence level (n=3)as it has a huge effect. Thicker airfoils have higher cLmax, but more drag and are more prone to cavitation, which is more important for a rudder than cLmax. Hence I modified a 16% thick Naca section for the T to get the cavitation right, of course we had higher drag... Can send this you as well, if you want