What about putting the average delta with a spread like 23C ± 0.6C so if a block have a larger variation, the spread would be larger. This would give a better idea if block X perform better than block Y and if mounting is harder to be consistent with a certain block.