So you got me curious about those percentages and I did some counting:
Of all the games given away in Feudalife Challenges since the beginning,
almost exactly 15% of titles had duped keys. In this case I tried to count all the games, including those that I never won.
If I begin the timeline in October 2020 (to try and make it more current, since some games like Nosferatu, QUBIC or Prismatica seem to have stopped dropping at that point and other games started to drop), then
around 12% of titles had dupes. That's only counting my own wins, because I don't think it's possible to make a really time-accurate list from user reports, so I won't even try
Out of all keys I won 5.6%(all time) and 7.7% (since October) were dupes, but those numbers are lower than they should be, because I have a habit of hoarding well known problematic keys (like Commandos), so I have a bunch of replacements for when I end up giving them away to raise chances of success.
I think the estimation that "
10% of the games are dupes and the other 5% it's from random games (for example there are a lot of games that always activate but in very rare cases they're dupes)" is wrong though - out of all games only a dozen have no known working keys (like Chaos Engine or Rhiannon) which gives us roughly 2%. Even if I add games, with majority of dupes and only a handful of working keys found (like Time Ramesside or Sacred Gold), the category grows to about 5-6%. Though I guess if you count the games with 50/50 chances of dupes in that category as well, then it fits.
Did I read this right? 0, 4, 8, 12, 16??? 20???
How on earth is any challenge possible with 20 let alone 16.
See my signature for videos, archers are doable in 44 seconds if you're lucky with their spawns and you can do hoeing in 44 seconds fairly consistently if you roll a decent map. I think hoeing is possible to do in 40 seconds and I used to do it back before the devs tightened up the difficulty, but these days I almost never feel in good enough shape, so I don't try.