For instance, imagine a game listed as free:
- a freemium model let you play 15 levels and asks for 4.99 to unleash 500 more levels, maybe let a 1.99 option for 100 levels; this should convert 3-5% with a million testing players; v.s.
- free-to-play with locking levels every 10 levels that are so hard you have to use bonus, that would convert 1% to pay for either a Golden Orb for 1.99, or a Golden Blast for 5.99, etc. each who may or may not make a difference, and appease your frustration, and work only once. You get a couple of whales (desperate, clueless players) who pay 50$ a month, but even with a larger distribution (two to three millions players) you don't break even.
The idea in the original comment is not that a price twice higher will double sales, but more that having a price tag (which free-to-play games deny having) makes sense for good games.