static null 0 with BGP

Before a prefix can be picked and announced via BGP, it must first be in the routing table.  I always null route with a distance:

ip route 192.168.1.0 255.255.255.0 null0 255 (cisco)

ip route 192.168.1.0/24 null0 distance 255 (brocade)

This is usually because you need to announce a supernet and only its subnets are in your routing table.

Thanks for the reply.
"Before a prefix can be picked and announced via BGP, it must first be in the routing table" That could be accomplish by any IGP. That does not necessarily why we need the null route. Correct?

I am still confused looking at many different answers including https://supportforums.cisco.com/discussion/10908346/null0-static-route.

I just want to know why we need a null static route with BGP and what if I don't use it.

As long as the prefix which you want to announce is in the routing table, then no, you don't need the null route.

If I want stable BGP, I might still null route it with a distance in case the internal route is dropped just to keep my prefixes from withdrawing and announcing causing a possible dampened state.

" in case the internal route is dropped just to keep my prefixes from withdrawing and announcing" Can you give me an example on this?
In order for BGP to advertise the network, the route for that network needs to be in the routing table. Correct?
So let say I in my BGP process network 10.10.10.0 and 20.20.20.0 and it is in the table with static routes. Then to keep the BGP stable, do I need 2 static null0 statements?

I found some answers that say the null0 is to avoid loop. I don't understand.