some maybe-biased data for a steel man in [1]:

"The Census Bureau measure overstates current income inequality between the highest and lowest 20% of earners by more than 300% and claims that income inequality has risen by 21% since 1967, when in fact it has fallen by 3% ... In 2017, among working-age households, the bottom 20% earned only $6,941 on average, and only 36% were employed. But after transfer payments and taxes, those households had an average income of $48,806. The average working-age household in the second quintile earned $31,811 and 85% of them were employed. But after transfers and taxes, they had income of $50,492, a mere 3.5% more than the bottom quintile."

[1] https://www.wsj.com/opinion/income-equality-not-inequality-i...

Data is also a really, really potent rhetorical tool, because it is definitionally never complete (a map that fully captures a territory is the territory), and by those omissions, the data can be made to say anything at all in a way that looks unbiased.

What's the question I should be asking, and what data answers it? I'm genuinely asking

Not you particularly, I meant the original: "it feels like every share of income is at its lowest except for the ultra wealthy."

It's ambiguous in several way, no time scale, "ultra wealthy" isn't defined, and "income" somewhat ambiguous.

Median household income is the stat usually used to measure income and it's still increasing.

If my income goes up by 1% and my expenses go up by 2%, has my financial situation improved?