Here on the outskirts of Atlanta, Sprint and T-Mobile have more or less equal coverage, with one picking up where the other doesn't most of the time. The issue with T-Mobile is that it's more difficult to get a signal indoors, depending on the building material.

For example, at my job we're in a large warehouse with metal walls and framing, lots of pipes, and suspended ceilings. That makes the building difficult but not impossible for radio waves to penetrate. All four major carriers saturate the area; stepping outside you get full service on all of them. But of the four, only T-Mobile has zero coverage once you step inside the building. The rest are usable to varying degrees, with AT&T and Verizon the two best, and Sprint is decent inside.

And just so this isn't purely anecdotal, T-Mobile has acknowledged issues with using their service indoors:

http://arstechnica.com/information-technology/2014/10/t-mobi...

Coverage != speed. You can have excellent signal strength, sharing the bandwidth with thousands of other mobile devices.

Bandwidth itself is impossible to value. The infrastructure providing the bandwidth is not free, no, but the idea of data caps to provide some sort of artificial value it fairly laughable, not to mention irrational. If they want to make this argument, charge me WHAT IT IS ACTUALLY VALUED (i.e. charge me for what I use based on a variable rate derived from maintenance costs), not some made up cap that likely has to do with what they'll estimate people might be willing to put up with.

Lucky!

When on 4G, I'm lucky to get 1mbps; even just browsing the web is an exercise in, 'what is this, a 512kbps satellite link?'

When on 3G, I can pretty much forget about doing anything other than email.

Here's my anecdote: Sprint took too long rolling LTE to Phoenix - a major market - so I switched to Verizon.