Here's the situation:
It's a fantasy soccer game where players score points for my team based on their actual performances on the pitch.
I have a team of 11 players and their is no limit to the number of teams that can select a player. All the teams exist in the same league in the game.
I have 10 picked and need to pick between two similar candidates for the final spot, both play for the same team, both have fairly similar expectations in terms of points scored over the season.
Player A has been selected by almost 60% of the players (3,000,000+)
Player B has been selected by about 15% of the players.
My questions concerns the situation where my pick scores fewer points over, say, 10 weeks.
Suppose that one of these players scores 100 points and the other scores 90 points over this period. If I have picked the 'loser' then my team has 10 points less than it could have. This is true whichever player that happens to be.
It seems intuitive to me, however, that my team will be relatively 'better off' if I pick Player A (and he is the loser) than if I pick player B (and he is the loser).
I have arrived at the conclusion that if I pick the less popular player, the rest of my team (i.e the other 10 players) would have to contribute more points to my overall total in order for my team to maintain the same relative rank.
Am I correct? I am trying to explain why I think this to someone and getting nowhere - I can't even really convince myself! They insist it makes no difference. Maybe they are correct.
Hopefully someone can help a novice, Im not even sure what tags to give this...
Thanks
Who are you competing with and what is your objective? It seems there are $5,000,000$ people who have picked teams. Are you ranked against all of them, or only some local group? If against the whole $5,000,000$ you are unlikely to win outright. Do you want to come in the top so many percent?
Basically you can think of the two players as having the same expectation, but B has a higher variance in your ranking against other teams. If you are in a local competition and want the best chance to win outright, variance is good. If B beats A you are ahead of a lot of people. If your happiness is linear in finishing place, you are more concerned with expectation and it doesn't matter who you take.