Support percentage y-axis scale

[ghostwords]

To make it easier to see the magnitude of changes. Should there also be an option for zeroing the axes?

[ghostwords]

Also, would it be useful and possible to (optionally) weigh the change in ownership by the growth of Robinhood's userbase? (Is something actually becoming more popular or is it just Robinhood's user growth?)

[dalexj]

@Ameobea percentage change sounds doable. does the popularity means direct ownership or is that a separate field we could pull in?

[Ameobea]

Hey @ghostwords. For the percentage axes, do you picture the relative change being calculated in terms of the minimum popularity, the lowest popularity, or something different? I'm not sure that either of those two representations would be valuable; they seem very arbitrary and rather difficult to make useful.

One other issue with that is that if a popularity goes down by 1% one day and then up 1% the next, the absolute popularity doesn't go back to where it was before since 1% of 99% is less than 1% of 100%. This is why the major stock indices are often quoted in points rather than percentages.

Something that I'd agree with more would be to display a small table on the stats page for individual stocks showing how much popularity has increased over different periods of time (1 day, 1 week, 1 month, etc.). Those differences could be shown as both absolute changes as well as percentages.

For your second idea of weighing changes by the total number of active Robinhood users, yes that would be a great feature to have. The difficulty comes from trying to calculate that active users figure. From what I know, there's no way to get that figure accurately, so it would have to be extrapolated from the individual ownership of different stocks. We could use a very common symbol like SPY as a benchmark, but we're then exposed to people reacting to moves in the market. There's probably a decent way to calculate this by averaging popularities across many different symbols in a way that avoids market bias, but finding that would be very tricky since there's no real way to see if it's correct or not.

One last thing - what did you mean by "zeroing the axes?" I don't really understand what you mean there.

[ghostwords]

Percentage y-axis: Relative to the starting value. Updating the range will update the percentages. For example, see Y! Finance charts. Click on the gear icon on the right to set y-axis scale. Try changing the date range to see the final right-hand side percent value get updated.

Zeroing the axes: Setting one or both y-axis min values to zero. The y-axis range now is always fixed to the range of shown values, so something can swing wildly between let's say 100,000 and 101,000 users but the actual magnitude of these swings is tiny, which would be obvious if you could set the user scale to start from zero.

These are just some ideas meant to help make better sense of the charts, I dunno.

[Ameobea]

Ahh - having the percentages change as you change the view of the window makes a lot more sense, and your explanation of zeroing the axes is perfect too. Thank you very much for explaining them.

I think both of those things would be really nice features to have; I'll see how tough they'd be to implement with the charting library I'm using. Thanks again so much for the ideas and info!