OK, so how about some other examples. How about contrasting Gnome Rider instead of Dragonfly to reach the conclusion that +1 quanta production and +1 toughness are worth about the same as +1 attack, and therefore 1 card is worth 1 quanta production and one toughness?
Essentially, no matter what cards you're using, you're cherrypicking from a set large enough that examples that run contrary to your 'logic' are available, and it belies the authority thereof.
Obviously not all hp increases are the same. +1hp to Spark is significant. Damaging CC has peaks.
Dragonfly is 1|1 Gen
for 1
+ 1 card
Gnome Rider is 1|2 Gen
for 1
+ 1 card
Since neither is seen as OP or UP, I would conclude that there is very little difference in creature resilience between 1 and 2 hp.
The theory is based on starting with the cards with the fewest variables and building up from there.
It derives its authority from 3 factors: Fairly accurate predictions, explanatory ability and my efforts to get others to make a better one.
It does a good job at predicting if a card will be judged OP or UP by the community.
It does a good job of explaining why creatures seen as overpowered / underpowered creatures are overpowered or underpowerd. (Almost no counterexamples, even theoretical ones, exist that are not considered UP/OP)
I have continuously encouraged members to create better theories to replace the current one. You know how long I have been here doing that.
However we can use an accurate pillar based test (cost compared to cost) to see if it matches your intuitions about balance.
A Pillar that cost 1 elemental quanta to play but replaced itself (drew a card) when cast would be UP, Balanced or OP?
Also consider a Tower that replaced itself (drew a card) rather than generating a quanta when cast.