Three hundred grams of chips per hectoliter give intense oak on a white, and moderate oak on a red. Same wood, same dose, same duration. This is not an error in the table, it is the wine that does not react the same way.
It is why the two columns of our conversion table are never read on the same line, and it is the point we are most often asked to explain.
What red wine does to oak compounds
A red wine is not a coloured white. It contains anthocyanins and grape tannins that bind part of the compounds extracted from the wood. What is bound is no longer perceived in the same way: the molecule is still there, but it no longer expresses itself on the nose with the same intensity.
A white has no such competing structure. What leaves the oak reaches the nose and the palate almost in full. At an equal dose a white therefore reads as more marked, and the gap is not marginal: published work on identical chips reports ratios of up to one to three between a red wine and a white wine.
pH plays its part too
Extraction itself depends on the medium. The pH of a white sits around 3.4, that of a red around 3.8, and that difference changes the solubility of several families of compounds. The effect is not linear and it does not go the same way for every compound, which rules out any simple ratio.
It is also why we do not publish a single coefficient between white and red. We publish two columns, each established by tasting on the wine concerned.
What it means in practice
On a white, the equivalent of a new barrel is reached around 350 g/hL. On a red, it takes about 500 g/hL. The gap holds across the intermediate levels: what gives light oak on a red already gives distinct oak on a white.
Rosés we treat closer to whites than to reds. The phenolic structure is low, colour alone does not make a red as far as extraction is concerned, and a rosé dosed like a red almost always comes out over-oaked.
The classic mistake
It consists in applying to a white a dose set on a red, usually because the red was worked first in the cellar. The result is a white crushed by wood, and there is not much to be done about it: you cannot take oak back out, you can only dilute it with unoaked wine, when you have some.
The reverse happens too, more quietly: a red treated with a white's dose simply reads as disappointing, and the conclusion is that the product does not work, when it was dosed at half of what was needed.
And the profile, in all this
None of the above concerns the character of the oak. Vanilla, caramel, spice or roasted: that comes from the toast, not from the dose. The colour of the wine decides the perceived intensity, the toast decides the profile. These are two independent settings, and confusing them is the second most common mistake after the one about units.
When in doubt, the drop by drop trial settles it within the hour on your wine, with your pH and your structure, which no average will ever do for you.
Read next: Oak conversion table: chips, staves, tannins · Le Nuancier des Possibles: the "Primary Colors"