Both quantitative finance and more general data science draw heavily from physics PhDs (you can see this if you look at job listings for these jobs; physics is almost always listed next to math in the list of desired PhD degrees). Depending on area there can also be a significant number of industry pure-research positions available; quantum information is a good example.
As long as you are willing to go into industry, you're looking at a 90-100k starting salary[1]. That's pretty damn good considering that in the US nobody pays for a physics PhD; the average TA stipend is around $30k with tuition and medical insurance covered. You also have a decent chance of getting a research assistanship to provide the stipend once you start working seriously with an advisor.
The high number of physics PhDs getting postdocs off the bat is driven by people who want to go into academia/research, not by a lack of industry options. Admitidely if you're dead set on going into research (especially right off the bat) you better be ready to play the same low-paid tenure lottery as everybody else, but if you're just looking for a good career a physics PhD is a pretty good bet.
[1]: https://www.aip.org/sites/default/files/statistics/employmen...
