I agree with the article for the most part.
Usually when someone says they play by “feel” in fighting games it means they don’t frame count or know exact numbers or hitbox placement and I usually interpret that as they’ll be more likely make a risky decision starting out until they’re sure it can be punished (or they’ll be stubborn and keep doing it until they’re dead lol).
Still they have to learn and grasp stuff like timing, spacing and risk/reward along with using whatever favorable mechanics in that particular game’s engine to their advantage if they want to get better. So even if they don’t go by hard numbers to figure this out, they definitely can learn the types of decisions that are more likely to put them in a good position vs the ones that are likely to get them killed/put in a very bad position and then make adjustments to avoid that.
For fighting games this is pretty much always based on what the opponent is doing about what you’re doing. 'Would be kind of hard to determine the true risk/reward of something if your competition doesn’t know how to capitalize on your mistakes.
At the same time, this could also be an edge when you don’t utilize stuff like hitbox/frame data but you can figure out quick enough that your opponent personally can’t do anything about your moves (“feeling” your opponent out) or you can open someone up more often with something that ends up completely unexpected because typically it’s “too risky”. I usually feel that way at times when I don’t know the frames vs. when I do. If I know my decision is mid to high-risk, I might not continue to use it after it worked once or twice in fear that my opponent might be waiting to kill me for it the next time (even if in actuality he doesn’t know that he can). The more I know, the harder things get sometimes when the knowledge becomes a limiter like that. It sucks, but that’s how it is… it ain’t stopping from using it to my advantage when I can tho.