Investing - Theory, News & General • Roth Conversion strategy feedback?

Hi,

There are lots of Roth Conversion discussion here and online, and there are different sides in the spectrum, from "Don't do it" = "Convert Some" = "Convert all". We are a couple at age of 51, and plan our retirement around 55, with $5M in pretax in 2026, so I watched tons of YT videos, and lots of websites, and tried multiple different websites/models/tools. This is a complex topic, as we have some many variables and factors, and same for a lot of people, we are trying to predict a future, maybe many years or even decades later, so I understand the disagreements and unknown for this.

After lots of thoughts, I personally think "Don't convert" and "Convert all" are NOT the right ways for our case, they are 2 extreme sides in the spectrum. But "convert how much" is really a hard question, I wonder if the following is a better strategy, and didn't see it discussed a lot online, which is focusing on today, with all the known facts, instead of based on the future. So I will use the following example, to discuss how I am thinking about it, and seek feedback. I wonder for those, especially already in RMD stage, what do you think? Thanks.

Backgrounds:
1) We are a couple in the same age. Currently in 35 Fed tax bracket, so won't consider it unless one of us lost job.
2) Currently we plan for $250K yearly retirement expense (including everything, but before the tax). I believe this is an important factor in Roth conversion planning.
3) There is also enough balance in Taxable, for retirement expense plus tax for Roth conversion.
4) We know that based SS website, our SS is $100K if we claim at age 70, in today's dollar, and it is the ONLY guarantee income for us.
5) We are doing this mainly for our kids, but it is too early to say their tax rates now.
6) I plan to keep around 2M in pretax. So personally, converting all is never a good idea for us.

The thinking strategy:
1) Let's assume we retire now, in 2026, with $5M in pretax, and wonder how much should we convert.
2) First, I will assume the RMD kicks in now, so RMD for age 75 (both of us) is $5M / 24.6 = $203K. Considering $100K SS with 85% taxable, it puts us into lower end of Fed 24 tax bracket, in 2026.
3) Our desired retirement expense is also at 24, so if we were 75 as today, with all the known facts, the RMD is > what we need, but in the same bracket.
4) In this case, I will pick up top of 24 bracket as target. I know this part maybe debatable, but we should be happy to pay at this marginal rate, for 2026, considering the RMD rate only goes up, plus widow's penalty later.
5) In Oct/Nov of 2027, base on then pretax balance, evaluate again. We should know the "real" return for 2027 then, of the pretax balance. There are different possibilities:
a) Good market, like 6%+ real return, which means $300K "real" return. This will move the direction to favor higher target tax bracket, eventually.
b) So so market, like 2% real return. This will keep the direction to favor same/lower target tax bracket (As we already convert some in 2026)
c) Bad market, like -20% real return. This is an interesting year. While it maybe mean less balance to convert, but we know it is in fact an opportunity. The question is how to balance with enough bond allocation. But with longer duration of bad market, it will change the direction for lower converting target.
6) Repeat this every year going forward, until RMD starts.

My thoughts are to remove future/unknown as much as possible, instead of focusing on this year's data/facts as much as possible.

Let me know what you think, thanks.

Statistics: Posted by arcane — Thu Aug 27, 2026 2:32 pm