[dfinlay]

In addition to what malloyd said, it also curbs spontaneity. Part of what helps soften the blow of dying is the coolness of getting to come up with a new idea you are keen on right now and introducing them into the story.It draws some attention onto you and adds freshness to the campaign. All of this is lost if they are old established NPCs (who may have even been established by the GM) and even worse if they were trained by your old character and thus have similar skillsets. The only reason to make backup characters in advance is if you are gaming in a style where death is so common that you don't want to pause the action after someone dies to work out a replacement. These sorts of games are rare and it definitely sounds like you aren't playing one. That said, if a character says "I want to take over that guy" after they die, that guy is an appropriate power level and it won't screw up your story to have that guy be a PC, let them. But don't make them commit in advance or even worse, lose capability from their old character in order to be able to do it.

That said, I disagree with what malloyd said about replacement characters being lower point value being neccessarily an issue. I've played in a campaign where everyone started at [150] no matter what and made my character when another PC was [300]. It was a lot of fun. If ther campaign focus is not too narrow and the characters aren't built to do exactly the same things, the [150] character can still find a lot of moments to shine and you wind up with a comparatively inexperienced apprentice feel, which can be a lot of fun. If you're doing this, though, it should be a deliberate choice because you want to create that feel and not an attempt to punish/disincentivise dying (players already don't want to die. You don't have to work for that effect) and the compounding halving is definitely a bad idea.