We still haven't learned from the Villeneuve-Hill-Frentzen comparison, it seems.
Hill beat Villeneuve in '96; Villeneuve beat Frentzen in '97; Frentzen beat Hill in '99.
Also consider these:
Alonso > Fisichella > Massa = Räikkönen. But Räikkönen destroyed Fisichella at Ferrari. If you say Massa > Räikkönen then it's even worse.
Hamilton > Kovalainen = Fisichella > Button. Which basically means Hamilton should had had an even bigger advantage over Button than over Kovalainen. It didn't happen.
Heidfeld had the better of Kubica in 2007, Kubica had the better of Heidfeld in 2008, they were evenly matched in 2009.
Barrichello used to match and even beat Button on poor cars (2007 and 2008), but was blown away in 2008 and 2009.
Different drivers in different cars designed to different rules with different tyres and different engines against different drivers and at different points of their careers CAN'T be compared. But I'm stating the obvious.
I don't think bluestar means that every comparision is equal because they are not. Some involve mitigating factors that can make the comparison much weaker, because all things are not equal, and your examples are perfect examples of this.
1. Villeneve was a rookie against Hill. The regulations in 99 were enormously different from 1996, with grooved tyres compared to slicks, so this comparison is a useless measurement because of these factors.
2, Button was far less experienced than Fisichella in 2002.
3. Fisichella jumped into a new team and car mid season with no testing, and was a few months before retirement. That is not comparable to a prime fisichella established at Renault.
4. Rubens was not blown away by Button in 08/09, they have always been evenly matched in speed, except for 2006 when Rubens had braking issues.
Unlike these examples you listed, the Massa-Alonso= Kimi comparision is almost perfect. 2 drivers swapping seats in a successive season, with Massa remaining, and a minimal regulation change. As I explained its far more equal than the hill, villeneuve one. Of course if you are not capable of true analysis, and just blindly want to pluck any comparision out of the air, then of course it might end up being confusing, but all it takes is a little bit of analysis to sort out the meaningful data from the rest.