First, I  am reporting and writing a feature for Las Vegas Weekly on Anthony Curtis, the owner of the gamblers' consumer newsletter Las Vegas Advisor as well as the publisher of Huntington Press. I do not know much about poker, but a couple weeks ago, Curtis was talking confidently to me about how one of his authors was about to win this television game show. She was doing this, he said, largely by applying game theory to the play. Curtis was quite convinced this strategy would allow Annie Duke to win "Celebrity Apprentice," and he wanted to have a book by her ready to go right after the show finishes and she wins. I was surprised he harbored so little doubt of her victory considering the huge roles that variables like television cameras and personalities play on these things. But Curtis had little doubt his author would easily adapt to any situation. His arguments to me that day a couple of weeks ago as to why Duke was going to win the show are interestingly almost identical to the ones made by Melissa Rivers in the above clip of bonus material from the show.

The other reason I know how Duke is approaching "Celebrity Apprentice" is that Duke is telling anyone who cares to read about her choices on her blog. But interestingly,  Duke blogs that her decision to go against Rivers was actually an example of her failure to apply game theory to the situation:

"So it came down to Brande and Melissa and who would be fired. When I was asked about this I had an interesting dilemma. At this point in the game, Brande and I had become very good friends so my loyalties certainly were with her. However, I also knew that going into the next task I would a) have to be project manager if Brande went along with me since she had already stepped up twice, and b) that if we were to lose I would have a much weaker argument against Brande than I would against Melissa. Being PM for sure on the next task would be a tough spot since the PM is usually the most under fire. And Brande was the stronger player to boot. She had raised much more money than Melissa, second only to me in fact in the game. She was an incredibly hard worker. And this is really important, she was really well liked. Much more well liked than me, obviously. Melissa, on the other hand, was a weak fundraiser, was not well liked at all and would have to be project manager on the next task since she had lost on her first try so would need to try for a win in order to win the whole game. So taking Melissa long with me would be a better play.

"But I really went against game play here and decided to argue for Melissa getting fired. Why? Because I really just felt Brande deserved to be in the game more, and I just plain really like and respect her as a person. So when push came to shove, I recommended Melissa get fired even though it felt like it was against my short-term best interests. I figured the life equity issue was much more important to me at that point … walking away from the game knowing that I had been loyal to my friend, to Brande."

And so Curtis should note there is a weakness in Duke's armor. Strictly speaking friendship should never have weighed into this decision. And once you are willing to be exposed that way, anything is possible in a game. Being too well liked to attack, I am sure Duke would admit could be a strategy applied by Brande, the model, who knowing she cannot beat Duke strategically has managed to use her charm to outplay her. 

I will start watching this show to see how things turn out.

Do any of you who have been watching think Duke is lock to win?