They were trying to figure out the root cause of a sudden uptick of millions of requests being made for a given image with no user agent or referer, presumably with a view to notifying the app responsible or figuring out a workaround.
A few thousand requests from clearly identifiable as coming from browsers and with a referer header from news.ycombinator.com would not exactly interfere with this and in the grand scheme of things isn't a huge burden in terms of network traffic.
