Myth: Reduce HTTP requests to go faster

Myth: Reduce HTTP requests to go faster
Merged 40 requests into 12: LCP 2.4s -> 2.5s.
The request-count rule is from HTTP/1.1's 6-connection limit. On HTTP/2 the bottleneck moved to bytes and priority, not count.
— Inlining a 90KB sprite to cut requests bloated the critical HTML and delayed first paint.
— What actually moved LCP: dropping a single 300KB render-blocking CSS file via critical-CSS extraction (2.4s -> 1.6s).
Count the bytes on the critical path, not the requests.
Takeaway: -28 requests = +0; -300KB blocking CSS = -0.8s.