Across lots of different industries and repeated time and time again you can find cases of people (or groups of people) making obvious mistakes that in retrospect are pretty obvious. Things like New Coke or releasing an album with a 'baby butcher' cover must have seemed like a good idea at the time, but it didn't take long for the mistake to be uncovered when exposed to the public.
WAR's beta should be a text book case of how to do things right yet end up with a very wrong conclusion. At this point, it is obvious what happened, but it needs to be pointed out so that it doesn't happen again. I wasn't in beta but had heard some things, so I asked how Mythic had run its beta process; the responses I got pointed to a common theme.
This theme is that by testing WAR a lot in separate pieces but rarely the whole as it would launch, Mythic missed the basic problems of WAR's size versus player population distribution. Apart from the start, when beta players complained that things were too repetitive, Mythic had beta testers test sections of content in concentrated bursts. Players would be levelled up as required to do the content in the test/ This worked very well and the beta players did get involved were able to go through the zone(s) and test the required areas easily. When that was done, there were plenty of players of about the right level to do some realm vs. realm combat (RvR) or public quests (PQs) or whatever else took peoples' fancy. No-one had to worry about levelling because the next period would see the characters wiped and new characters level bumped as required.
An RvR area that would have been full in beta testing lies empty post-launch.
This worked great as a beta test - lots of testing focus, lots of players in a limited area to test how things worked and could possibly be broken. However, little time was left to examine how the whole game fit together when players had to level from 1 to 40 under their own steam. The sheer size of WAR's world, together with a large number of world servers, means that players have been distributed far and wide - something that wasn't adequately tested for in beta. It seems likely that Mythic looked at the feedback taking from the tiered beta testing and thought it would apply equally to the title post-launch. Sadly, this is obviously not the case.
The lesson of this tale (and yes, it seems obvious) is that you need to test things under 'real world' conditions as much as possible to know how they will behave in the real world. A tiered beta test is a good idea for examining each section in isolation, but the game as a whole needs to be tested too in order to find out how it hangs together. WAR doesn't hang together particularly well since it appears balanced for a much larger number of players in each chapter - something that isn't happening in the mid-levels of this game. Scenarios dominate RvR - the supposed core of WAR - because they are easier to take part in and give better rewards i.e. allow people to level up more quickly.
Some people might want to blame EA for such things, but I find that one hard to accept given that Mythic appears to have run the beta themselves and just not quite ever got to the point of letting players test from level 1 to 40 in a natural state. Also, given that Mythic already has one successful MMO title under its belt, you would have thought they would have known better.
Unfortunately the case is often that obvious mistakes are only obvious after you have made them.