Illuminant estimation algorithms are usually evaluated by measuring the angular error between the RGB vectors of the estimated illuminant and the ground-truth illuminant (recovery angular error). However, the recovery angular error reports a wide range of errors for a given illuminant estimation algorithm and a given scene viewed under multiple lights (despite the fact that when the estimated lights are divided out the reproductions are similar). Following this observation, the reproduction angular error was proposed which instead measures the angle between the RGBs of a white surface being color corrected by dividing out the estimated illuminant RGB and R=1, G=1 and B=1 (i.e. white if the illuminant is correctly discounted). In this work we look at the correlation between the two metrics for each individual algorithm applied on a set of images. We observe that where the images are from the same scene under different illuminations the recovery and reproduction errors are often uncorrelated. Whereas when the same algorithm is applied on the images of diverse scenes the two errors are highly correlated.