[KaVir]

Actually, yes, it does.

Many muds and clients already support existing protocols - which means at worst a mud owner has plenty of examples to follow, and at best their mud already has them built in. Inventing your own protocol can be a lot of work, and it's unlikely other mud clients would add support for it, meaning you'd most likely end up either adding support for two protocols that did the same thing, or alienating the players who preferred their existing mud clients.

Take a look at MCCP for example. Back in 2001, Lasher gave some statistics for the amount of bandwidth it saved Aardwolf, which showed that with a little under 250 MCCP users the bandwidth usage dropped to around 18-19% of its original amount. He went on to say "if we ever have to move to a host that charges for bandwidth it could very well mean the difference between being able to afford to stay online or not". You've said you're targeting large muds, so I think this is definitely the sort of thing that's worth looking at.

Now perhaps Lasher could have created his own compression protocol, but his goal was to encourage as many people as possible to use it, and that means it's in his best interests to have a widely supported protocol. Not everyone will use the same client, so using a standardised protocol supported by multiple clients makes a lot of sense.