macrocephalopod

or re-centre (sticky delta)? In reality it will be somewhere in between. The new spot price is a datapoint with which you can update your "prior" probability distribution. So the degree of delta-stickiness is related to how important you think the new spot price is.

@volmagorov what do you think? I'm not sure this is yet a practical observation, but it seems to be a fairly interesting philosophical link.

I’m not an options guy but I thought this debate was resolved nearly 100% in favour of sticky delta? Most places model the smile as a function of moneyness so vol deltas on a price move are smaller.

From the options dudes i've been speaking to, there is near-unanimous agreement that sticky strike is wrong but not everyone agrees that pure sticky delta is right.

Isn't this largely regime and asset dependent? But yeah, what macro said. I think I've seen some mixture models floating around.

Yup, but I'm trying to reason about how exactly it depends on regime/asset. What drives the appropriateness of sticky strike/delta? e.g the shape of the smile depends on regime/asset, but we can reason about it in terms of flows (conscious that this may just be a narrative)

for instance, equities smirk because of demand for downside protection while commods skew positive because of demand for upside protection from hedgers. So I wonder whether there is a more intuitive way of thinking about sticky strike vs delta and bayesian updating makes sense

One thing I’d think would be important is degree of mean reversion in the underlier. Equities don’t have much (so sticky delta more likely to be appropriate) whereas rates are more likely to mean revert (which would make sticky strike more relevant).