Apple, perhaps intentionally, has turned the once clear problem of the iPhone 4 displaying dramatically worse reception when the bottom left band is covered to a muddle of “facts” about bars and antennae.

Here’s what we know:

Many, perhaps all, phones have worsening performance when they are held in a certain way. Specifically, covering up the internal antenna with your hand is generally thought of as poor form. I remember an old Nokia I had specifically instructing me how the phone should be held and that I should be vigilant lest my thumb drift over the Nokia logo, which served double duty as an antenna.

Many people gauge “reception” by the bar meter displayed by basically all phones, but there is in fact no standard as to what these bars mean. They’re basically only useful for comparing relative signal strength between the same make/model of phone, and even a strong showing on the meter doesn’t guarantee good performance, though there is some correlation.

According to Apple, they had calibrated their meter such that only when the signal was relatively weak did the bars start dropping. They are attributing this to human error, but the more cynical could argue that it served their purposes in the past (more bars in more places), but not now (dramatic fall of in “bars” when held the wrong way).

The general consensus is that covering the black band on the left side of the phone is a bad idea because it “shorts” the two antennae together. This is a hardware issue because of the exposed antennae, though that same hardware feature probably improves the signal a bit when you’re not covering up the black band.

Why doesn’t this affect WiFi reception? Why doesn’t covering the black band on the top of the phone (another junction between the two antennae) have the same effect? I haven’t gotten good answers to this yet, but there probably is some explanation, since using a case (so you can’t touch the antenna metal) seems to alleviate the problem.

Apple’s fix is to recalibrate the reception meter to what it says is AT&T’s recommendation. But the iPhone is sold in more countries than the US, on other carriers. Does Apple calibrate the signal strength separately for each carrier? Probably not…my guess is that AT&T’s formula will just lessen the severity of the perceived signal drop and highlight areas with poor reception, which will be seen as the carrier’s fault (which it probably is, to a large degree).

Apple has promised a fix for the meter in “a few weeks”. If they’re just changing the formula, why wait a few weeks? This is my guess: Apple got caught off guard by this issue, probably because they test their phone in disguise, protecting the metal band. They want some time to figure out if there’s anything else they can do via software to remedy this problem, but they needed to put out some statement now because otherwise it makes them look clueless and distracts from all the features of the new iPhone.

The backup plan is likely not to offer free bumpers, since this will hurt case manufacturers, but to give an Apple Store credit for the value of a bumper. They should probably just do that and be done with it.