Econ2214 Games and Decisions
Reflective Essay 1
Zheng Jingru
UID: 3035490149
Sequential and Simultaneous Games (Pure strategies)
Sequential games are played in a specific order, which means players will make choices in
turn. The more important feature is that the players can observe what the others have
chosen before and make decisions based on other choices. That means players who take
action first will influence the game.
The common way to present the sequential games is usually the extensive form—game
trees, which is used to represent all the available choices from the players respectively.
Consider two players, A and B are playing the sequential games. A will move first and can
choose up or down while B will move after and can choose left or right. The game tree of
the game will be as follows:
(0, 0)
Left
Up b right
(1, 2)
A
(2, 1)
a Down left
B
c right
(0, 0)
A complete plan of action needs to be listed out clearly as it can tell what choice each player
will take at each decision node. From the game tree above, it is not hard to get all the
available strategies to A and B:
A: up and down
B: (high, high) (high, low) (low, high) (low, low)
In sequential games, players can estimate their rivals’ reactions to the decisions made by
the first player. In this case, A must consider how B will make the choice when A choose up
or choose down to think of which one will be the best strategy. Since all the available
strategies are listed out, the next step is to filter out the unlikely choices by backward
induction. This means that the decision-maker observe backward from the terminal nodes
and cross out the unlikely options. Using the backward induction can map out all the likely
moves that the players will take rationally from the initial nodes to the payoff. For example,
consider the game mentioned before, B won’t take leave when A take up and B won’t take
right when A takes down as B makes choices in its best interests. At node b, playing left give
B nothing, while playing right give a payoff of 2. While at node c, playing right give nothing
to B but playing left can give a payoff of 1. Thus, it will be easy to cross out the unlikely
choice for B. When it comes to A, playing up can give a payoff of 1 while playing down can
give a payoff of 2. The latter is better than the former. Finally, the equilibrium of the game is
A playing down and B playing right if A playing up and playing left if A playing down. This is
the subgame perfect equilibrium of the game.
(0, 0)
Left
Up b right
(1, 2)
A
(2, 1)
a Down left
B
c right
(0, 0)
Simultaneous games are games that players make the decisions at the same time or they
have no information about other players’ actions at all. Each player needs to make the
decision based on other players’ likely plan.
The common way to present the simultaneous games is the normal form—a game table,
which is used to represent the available strategies and the payoffs. Consider two players, A
and B, who are playing a simultaneous game. A can play up or down, while B can play left or
right. The game table is as follows:
B
Left Right
A Up (80, 100) (40, 80)
Down (110, 65) (55, 40)
As the table is shown above, the next step will be to find the best strategy to get the highest
payoff. Given that the players are rational, they are supposed to always follow their
dominant strategy. A dominant strategy means that it can provide the players with the
highest payoff within any other strategies but it is not necessarily higher than its rivals.
For example, if B plays left, A’s best response is to play down and get a higher payoff
(110>80). If B plays right, A’s best response is still to play down and get the higher payoff
(55>40). Thus, A should choose down regardless of what B chooses. That means playing
down is A’s dominant strategy. Similarly, B’s dominant strategy is to play left. Since neither
player would change its mind to other choices, A playing down and B playing left is the
equilibrium.
However, there is a concept that contrast with dominant strategy—dominated strategy.
When the players make decisions rationally, they never play a dominated strategy, which
means there must be other available strategies that can provide a higher payoff. For
example, playing down already provided the higher payoffs than playing up for A regardless
of what B’s action. Therefore, playing up is a dominant strategy for A and A won’t be better
off if playing up.
Equilibrium is differently achieved in sequential and simultaneous games. In sequential
games, subgame perfect equilibrium can be reached by making decisions rationally based
on other players’ choices in the past. While in simultaneous games, the equilibrium is called
the Nash equilibrium and will be more difficult to reach. Nash equilibrium means that
players don’t have any incentive to switch their strategy to another as they have chosen the
one that provides the optimal payoff for them. Notice that subgame perfect equilibria are
Nash equilibrium but not all Nash equilibrium are subgame perfect equilibria. For example,
in the case of the sequential game, A will always choose down, B will choose right when A
choose up, and choose left when A chooses down. Thus, A choose down and B choose left is
the subgame perfect equilibria. However, if A always plays up and B always plays right, Nash
equilibrium can still be achieved in this case. From the game tree above, no matter what
strategy A or B switch to, they won’t be better off. Thus, it is Nash equilibrium in this case
but it is not subgame perfect equilibria as it violates the backward induction that B won’t
play right in node c.
(995 words)
