Metcalf's law has proven to be false repeatedly. It's just a quack theory to overprice tech stocks.
Any evidence? Because all the mobile phone industry growth proves you wrong, so you are going to need something more than that.
I think you are confusing number of connections can be created "between users" with number of "Actual users".
If there are 4 users (A,B,C and D) then number of connections that can be made 6 (assuming connection is bi directional)
(A<=>B, A<=>C,A<=>D, B<=>C,B<=>D,C<=>D )
If user E join this network ,now only 1 new user added but number of connection that can be made now will be 10 (increase of 4 connection).
(A<=>B, A<=>C,A<=>D ,
A<=>E, B<=>C,B<=>D,
B<=>E, C<=>D,
C<=>E, ,
D<=>E )
So in simple words I will say if "X" new user adopt bitcoin (with "Y" existing user base) then bitcoin adoption is increased only by "X" people only
but number of connection (i.e transaction that can happen between user wallets,(assuming each user has 1 wallet) is increased by [(X+Y)*(X+Y-1)/2] -[X*(X-1)/2] = XY+Y
2/2-Y/2.
If you are assuming mobile industry growth with number of connections available then you are right but if you are equate with revenue/profit generation then you are in wrong side.