macrocephalopod

This is exactly the same as saying that by diversifying you can reduce the volatility drag of a portfolio of assets, right? ie “diversification is the only free lunch in investing”.

No it's not. The most diversified portfolio it not the right portfolio to rebalance back to capture the full effect. Typically the most diversified portfolio should be held and not rebalanced freqently to increase returns.

Yeah I agree. The "equivalent" thing would be the premium you tend to get from rebalancing assets with certain cross-sectional time-series properties - not the diversification (correlation) effect itself, I think.

Can you point me to somewhere that explains it in more detail? I feel like I have a pretty good mental model for these things, and (in a theoretical world with two assets with equal risk) it is obvious to me that frequent rebalancing to equal weight is optimal.

Hence if you tell me any *other* strategy is optimal, I think you’re making a claim about the real-world properties of the assets.

I believe you know what you’re doing and your calculations look right — I just don’t know what point you’re making?

This is already a practical way to benefit from a zero geomean asset, assuming it is such. So we shouldn't label these things as purely theoretical.