Another downside that I don't think was mentioned is that your "ramen profitable" number goes up by about 50%.
Even with that, I think the 50% bump in (free and maniacal) manpower plus the extra perspective is totally worth it.
I always say that it's better to optimize your CHANCE of success rather than the magnitude of it, should it happen. Hard to argue that a good (free!) co-founder wouldn't do that.
There doesn't seem to be a 'correct' number of founders (though clearly having 4+ or 1 has been known to be more difficult). You should choose your number of founders based on what you think will give you the highest probability of success given your situation, personality types, etc.
The title of the article should be, 'How Having 3 Founders Worked Really Well For Us'.
It's probably just our engineering instincts trying to find the patterns in things.
Normative startup advice seems to be about as useful as normative life advice-- not very.
You frame a problem in a context, or a time frame, or a motivation and you can argue whatever point you want.
All predictions about the future are by definition opinion.
One of my previous bosses made me promise to never start a company with only two founders for exactly this reason. This was after a very painful startup experience that ended with lawyers talking to lawyers, one founder buying out the other, and a fair bit of acrimony. I don't know how common this is, but I think I'll stick to the promise.