This is a pretty clever insight. And if by “consistent” you mean low polarization then the data backs up your theory. Decks that are high in Discover opportunities have historically been low polarization decks. Even Rainbow Mage, a deck that includes a Sif fueled combo game ender, has low-mid polarization overall because it has so much Discover.
Decks high in Discover also tend to reward piloting skill better than normal.
—
Regarding the main topic, I think most people think they understand randomness because they think that people can understand randomness. Of course, this is mostly incorrect. The first thing to understand about randomness is to understand that the absence of ANY pattern is its defining feature and that the human brain is defined by its pattern recognition. You are not equipped on a biological level to understand randomness in a directly perceptual way. Randomness is like the mathematical concept of infinity in that it can be understood in an abstract way but it can’t be understood at all in a visual way, and the same types of thinking that work with the concept of the infinite also are necessary for the fullest possible conception of randomness.
On these forums I see CONSTANTLY people pointing to one time coincidences, or to streaks, and saying that “this can’t be random.” This is what I mean about thinking about randomness wrong. If you generate a random outcome an infinite number of times, not only will literally everything happen at least once, but literally everything will happen an infinite number of times. As the number of randomly generated outcomes approaches infinity, your longest streak of getting exactly the same outcome in a row approaches infinity; this is obvious if you think about randomness in an abstract way but deeply counterintuitive if you’re thinking perceptually.
There is no outcome whatsoever that can’t happen under random generation, providing that it is possible, so outcomes are NOT the criteria for whether something is or isn’t random. PREDICTABILITY of outcomes is the standard for randomness, and that’s a completely separate concept from the outcomes themselves.