What I'm saying is that some sites seem to take the numbers approach (X does better in a specific web browsing test) and then subjectively proclaim that device has the best battery life, even if it's only better in one specific scenario. You can't do that, just like I can't go and say because the Note 2 has better overall battery life than the iPhone 5, it is better in any kind of battery test.
The things I had setup identically on the Note 2 and iPhone 5:
Exchange and Gmail push (combined 50-60 emails a day)
Twitter push
What's App (100-200 messages a day)
Roughly 2 ~300MB Dropbox downloads each day (TV episodes, watch 2 a night)
Maps/Navigation usage, about an hour a day (mostly to check traffic on various routes in the morning and evening)
The Note 2 also had:
Google Talk
Google Now, set to show and alert weather on location, as well as traffic alerts from location in the mornings and evenings.
The Note 2 consistently lasts me two days with this type of usage. The iPhone? one day.
The iPhone 5 gets one more hour of playback in the tests, but who watches 8 hours of video in a day? Same with the Moto phones getting 20+ hours of talk time, no body spends that much time talking on a phone. What we need is a quantifiable mix of all these different uses that would more closely resemble real world usage.
Also the Maxx is the flagship Razr Phone. The flagship is your best. The One X was the flagship until they launched the One X+ or whatever it's called, which then became the flagship. It only makes sense not to include a phone like the Maxx if it had a battery so big the device was comically large and thick like those extended 3rd party batteries. It looks like a typical flagship phone though.
I do agree with you benchmarks don't tell the entire picture. I that's why there's also typical standby benchmarks that need to be done, but that's more difficult. However, I don't think I'm entirely wrong in saying Android in general tends to use more battery power in standby.
iOS uses push strictly, while Android tends to use background processes to sync. Now that Twitter finally has push on Android, in order to get push, you also have to periodically pull. I had some debate here with some folks, and while they told me that they like the pull function, it doesn't take rocket science to figure out that adding pull uses more battery.
If you look at idle numbers for the iPhone 4, 4S, etc, they all idle with less power than the HTC One X and SGS3 (Source: Anandtech). I don't know how they do it, but Apple squeezes more battery life out of their devices.
So maybe the GNote is done well. Who knows? My basis is the Nexus 4, which runs horrendously. But what do you do on an iPhone that kills it in one day? I'd have to play at least an hour of temple run and have all my syncing to bring it to a point where I have to charge at night. I already showed you what I sync on my iPhone. I also forget that I have some 8gb of photos on there. I have Google+ sync on my iPhone, so it's constantly uploading new photos.
I tend to not like to rely on anecdotal data which is why I even view my anecdotal evidence as meh. If the best objective data we have is from benchmarks, then they should be used for discussion. They may not represent true use of the device, but I'd say it's better than nothing.
As for the Maxx, it's the flagship yes, but the Razr HD is marketed as the main phone. The MAXX is just a HD with a bigger battery. That's how it's marketed. It's designed to be a battery monster, so yes I expect it to beat everything. I'm not saying you can't point that out, but having the MAXX beat any phone is a meaningless statement. Also with a MAXX beating an iPhone doesn't make it middle of the pack now.
Middle of the pack is probably what the HTC One X or SGS3 international are scoring in battery tests. The Nexus 4 would score low, and the iPhone 5 would score high. The MAXX would score even higher, but isn't that expected with a 3500mah battery? Also isn't the MAXX a Verizon only phone? It's practically unavailable everywhere else if that's the case. I don't see how that counts as a competitor.