I think that's a reasonable idea and I may try including that option in the future. Right now, the start point is quite arbitrary: it's the first date where we have data for all the symbols requested. And date ranges dramatically affect any sort of relative performance comparison. So in the current code I decided to just normalize to today's nominal-dollar value so that the end values are relatable.

What would the ideal UX be on this?

If you want to have ragged starting information displaying pricing information for the various symbols in the range they're defined, I'd assume it's much more common that there's different start dates, but they're all defined for today. So, you could have a single checkbox to normalize based upon equal value at today's value and I'd default to having that on. As your baguette argument about the absolute values not having huge significance still applies it seems like it would work with no other modifications (i.e. it doesn't matter that it could require fractional shares and the like).

Just index it all to the same starting value, which will have the same effect. I don't think this works for the drawdown chart but works for the total return chart.

Usually in economic and financial reporting you show multiple return series by saying, for example, 1989 = 100 and then showing the total return from there for multiple series in the same chart.

Growth of $10K (growth of a hypothetical $10,000 investment) is a chart that is commonly-used to illustrate this. Your chart would then have VFINX, VBMFX, and USDOLLAR start with $10,000 on 1987-08-03 and go from there.

Showing $10k way back in the day would be a bit misleading since every graph starts in a different day and the whole idea is that inflation makes $10k translate to different real world value (baguettes) across time.

It could instead show how much back at the start date one would need for it to be todays $10k? Not sure. Maybe absulute dolar values should just be ommited and work only off percentages.

Start every asset at 100k? I might be oversimplifying something, but it would really help to see how each asset performed over time.

Isn't the comparison of relative performance in the trend line? That's a year over year % change for every asset, so it doesn't matter what value you start with.

Yeah it's the trendline's slope, but it might be a bit hard to compare them since the angles are not that dissimilar.