From the animated images you can see that the probability distributions have mostly stabilized by ~30 turns. That’s about 5 times around the board, which in my experience is enough for people to start building up monopolies but isn’t quite the end game.

The exact probabilities for turn `t` can be calculated in matlab from the transition matrix as `A^t`

Monopoly games are not infinite. They end out of either fatigue, mutual agreement, or utter defeat at some point. Additionally, there is only so much money in the bank, etc. Considering there is an end point (let’s say, for example, 50 turns is about the extent of it) and that all players (2-6) start on the same square, does this skew the results? For instance, Illinois Ave. may eventually be the most likely square, but it is the most likely square in the first 5, 10, 25, or 50 moves? I highly doubt anyone will want to “go for the reds” if, indeed, the oranges or purples (or light blues) of the world are far more likely to be landed on while the game is in session (rather than 4,300 turns later). Is there a way to calculate probability in certain time increments (every 3 or 5 turns?) rather than an overall metric?

]]>