This is not quite true, because poker is fundamentally adversarial, while startups are mostly not adversarial, at least not directly.
Startups are a beauty contest where each player focuses on maximizing the things about them that appeal most to a panel of judges (customers). Similar to the scramble competition that Benenson cites here, rather than an arm-wrestling contest.
Right, but my argument is that there are many niches in the market where there is no competition, and startups should try to find that and then creates moats with IP, data, etc. There are many situations where startups don't have direct competition, because they are inventing something radically new. Often true in life sciences, for example. You're right that in those cases they are in a race against time, since someone will eventually come along, but they can go for years without a direct adversary.