OK, other than the fact that it is rather crap, the biggest problem I have with Golden Balls is the Share or Steal at the end. Whilst I never studied Game theory, this is pretty much a Prisoner’s dilemma. However the payoff matrix is weighted in an odd format.
Assume that you have a 6th sense, you are sat in the chair and the amount is £50,000.
If you knew that your opponent who you will never see again is going to share what would you do?
If you share it’s 50/50, and you walk away with £25,000, but if you steal thats £50,000 in your pocket!
However if you knew that they were to steal, your f*cked. You’re going home with nothing, so what do you choose? Well if it’s share, you’ve just given your opponent fifty grand, but if it’s steal you could shaft them and ensure that they go home with nothing too.
So if you look back, on both scenarios steal was the best option, which means that you should always steal.
The show Friend or Foe had this exact same thing, and for me it was a major turn-off.
I would only watch shows like this if they added a 3rd option that ‘Busted’ someone for choosing Steal but only got a tiny payout if the other person Splits. Fans of the format say “it’s game theory etc”, but I say its an unbalanced game, almost like a version of Rock-Paper-Scissors where Rock is banned.
Here’s how I work a endgame like Golden Balls or Friend or Foe?:
Split x Split: Money divided 50-50
Split x Steal: Stealer gets 90%, splitter gets nothing.
Steal x Steal: Both get nothing
Steal x Bust: Bust gets 60%, Steal gets nothing or 10%, with Busting player choosing if Stealer gets a mercy prize or not.
Split x Bust: Split gets 60%, Bust gets 10%.
Bust x Bust: 1 Question face-off is held with winner getting 30% and loser getting none.
My way combines a trust test with elements of Rock-Paper-Scissors. In all my results, choosing Steal or Bust automatically hands money back to the show. There’s motive to Steal, but there’s also RISK. The current format doesn’t have risk with Stealing, so I agree that it is faulted by design.