Since I started producing the following graphic for the ETF Internal Analytics product, I found the weekly return bin information compelling. I became curious about whether there was an opportunity to be exploited in the distribution patterns. I distilled all the questions I had into two: 

1. Does the percentage of ETF component stocks at various return levels provide actionable information?
2. Can a long-short market-neutral strategy be constructed by analyzing the relative return dispersion of each ETF's stock components?

To answer these questions I used a combination of tools/data sources including State Street's SPDR Holdings data, the Yahoo Finance API, and Python.