Given the uncertainties and complexities at play, i was looking for a probability weighted models considering current\future interest and marginal tax rate patterns. monte carlo models for retirement planning was a major breakthrough* 20+ years ago, but that technology should be (hopefully) surpassed by now with a tool for this specific issue of Roth conversion. - weatherman

------------------

Probability models? Still imprecise. I think you are making this more complicated than it needs to be.

The method I have been using only looks to the next tax year and then to to convert the max possible up to the limit of the 24% bracket (reverting to 26% when Trump tax cuts expire). That only requires predicting one year into the future and if you wait until December to make your conversion, you only have one month to predict. Just convert the max possible given the constraint each year and let the chips fall where they may. There is no penalty for draining your TIRA a year or two or three before RMD's would have started. Once you are within two years of Medicare then the secondary constraint of IRMAA enters the picture.