The problem with comparing tanks rDPS is that it is essentially useless for them. Any benefits that a tank gets under a raid buff is given to another job. However, this causes issues as you cannot see how much a tank contributes in the 2 minute buff window. Get the tank on a striking dummy and it will do the same DPS as when you compare rDPS in a group. So really, you need to measure tanks on how well they can put their bursts in the 2 minute window, which is where aDPS comes into it.

This discrepancy can be seen on FFlogs where the ranking for rDPS is GNB > DRK > PLD > WAR, however aDPS is DRK > GNB > WAR > PLD. This differences comes about because different tanks can utilise the buff window differently. DRK can save up Edge of Shadow and increase their aDPS, whereas if you were to just look at rDPS, this factor is not considered for the DRK as that contribution is split between several other party members.

To give you an idea on this, based on the upper quartile of players, this is the % of damage the lowest job is doing compared to the highest. In the case of rDPS, GNB is top and WAR is doing 98.6% of the GNB damage. Compare this to aDPS where PLD is doing 97.0% the damage of DRK (this is based on the whole of Abyssos over the last 2 weeks).

rDPS should not be the only thing you look at when trying to balance jobs, because using the wrong data in the wrong context is going to lead you down the wrong path.