Had a talk with João about it, and here're some plot I'll look into:
The average subgrain size of different regions along the compression direction. Plot the curve and compare with other samples from different cooling rate.
The average subgrian sizes/misorientation of each grians in the same EBSD map, and compare different cooling rate.
Calculate the true strain at the centre accroding to the diameter of middle of the sample, and see if it affect the misorientation/subgrian size variance.
Try to plot the difference in subgrain sizes of each orientations (might not workable for my uniaxial compression case).
Use the Kuwahara Filter to identifiy the subgrians.