Hmmm... Well, you have some data, but how about a theory to back up your point?
Let's try this one on for size. We'd expect that the number of likes is proportional to the number of people who have seen that link, and as for the reason you brough it up the number of people that actually like CiV. Let's define some variables and see if we can get to the bottom of this:
P: Number of people
P_4: People who like Civ IV
P_5: People who like Civ V
To a good fit the number of people may defined as a function P(t) where t is defined as the number of days since the facebook began. Let's just assume that the probability of someone finding this group is only dependant upon them enjoying the game and not just spotting someone else liking it like most people of facebook. From this it is safe to assume that the number of people who 'like' each can be defined as:
Likes_IV=Int[P(t)*P_4,{t,r_IV,t_0}]
Likes_V=Int[P(t)*P_5,{t,r_V,t_0}]
Where here r_IV is the release of Civ IV and r_V is the same for Civ V. Here from data on wikipedia the function P(t)=(2*10^6)*Exp(0.0024t)=a*Exp[b*t] for simplicity. Here r_IV=663 and r_V=2455 as well as t_0=2622. As such we may solve the above equations for the relative number people liking each game.
Likes_IV=Int[P(t)*P_4,{t,r_IV,t_0}]
Likes_IV=(a(Exp[b*t_0]-Exp[b*r_IV])P_4)/b
Which means that:
P_4=b*Likes_IV/(a(Exp[b*t_0]-Exp[b*r_IV]))
And of course:
P_5=b*Likes_V/(a(Exp[b*t_0]-Exp[b*r_V]))
Calculating this gives:
P_4=5.27569*10^-8
P_5=5.4846*10^-8
Which taking this as a ratio gives:
P_5/P_4=1.04
In other words according to this the numbers are about the same, which considering the fan community is probably about right. Whilst many people loved Civ IV and quite a few don't like Civ V as it stands, (or stood with this new patch) there are undoubtably some new fans out there that would make up the lost numbers. That and it's shiny and new... Also note that the probability isn't that a few in a 100 mil people like Civ, but rather that the number that was calculated was actually the number of people that will like Civ IV or Civ V per day per person.
...come to think of it that was kind of pointless... Oh well. You can find statistics to backup any point though.