If you assume Gesink’s value is correct, and calculate for Yates and ten Dam based on the time difference, you get values very close to the ones they actually published (5.85 and 5.56, respectively). If you then calculate Froome’s value based on Gesink’s and the time difference, you get 6.15 W/kg, very close to the value one gets using Sky’s uncorrected 414 W and 67 kg (6.18 W/kg).
With regard to the argument that Gesink rode more in the wind than Froome: As far as I can tell from the stage description, Gesink attacked with about 11 km to go, and Froome attacked with about 7 km to go. So Gesink was alone for about 4 km more than Froome, though this may be an overestimate, because at some point after attacking, I believe Gesink was with TVG. Riding unprotected at this speed adds about 8-9%, or 30-35 W, to the energy requirement. However, even a rider who is protected experiences some drag, about 60% of that unprotected. So the extra energy Gesink put out is about (4/15) x (30-35) x .6 = 5 – 5.5 watts. This is about 0. 0.07 - 0.08 W/kg. If we factor that into his power reading, and re-calculate Froome’s power based on time difference, we get 6.07 – 6.08 W/kg. This is about the value one would get using the 414 W, a weight for Froome of a little over 66 kg (see Flor's post above), and a 3% oval ring correction.
This is a very rough calculation, but I think it demonstrates that most of the difference between Froome and Gesink can’t be accounted for by greater drag. When you add in the close agreement of Gesink's values with those of Yates and ten Dam, this becomes even clearer. We really don't know the correction factor, if any, for oval rings, but depending on Froome's actual weight, it could be several %. As I said before, I don't think one can assume a 6% correction as a blanket statement, applying to all riders.