GAME THOERY

Dr. Keerti Jain

NIIT University, Neemrana
 Game Theory

Game theory attempts to study “decision making in situations

where two or more intelligent and rational opponents are involved
under conditions of conflict and competition”

 In business and economics literature, the term ‘game’ refers to a

situation of conflict and competition in which two or more
competitor(or participant) are involved in the decision making
process in anticipation of certain outcome over a period of time.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 2

Game theory
Mathematical theory that deals, in an formal, abstract way, with the
general features of competitive situations :
 Like parlor games, military battles, political campaigns, advertising and
marketing campaigns, etc.
 Where final outcome depends primarily upon combination of strategies
selected by adversaries.

Emphasis on decision‐making processes of adversaries

We will focus on simplest case:

 two‐person, zero‐sum games

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 3

Game theory

Primary objective is development of rational criteria

for selecting a strategy.

Two key assumptions are made:

1.Both players are rational;
2.Both players choose their strategies solely to promote their
own welfare ( no compassion for opponent).
Contrasts with decision analysis, where assumption is that
decision
– maker is playing a game with passive opponent – nature
which chooses its strategies in some random fashion.
Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 4
HISTORY

• Game theory came into existence in 20th Century.

• In 1944 John Von Neumann and Oscar Morgenstern published a book
Theory of game and Economic Behaviour.
• In the book, they discussed how businesses of all types may use this
technique to determine the best strategies given a competitive business
environment.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 5

Basic Terminology

 Number of players : If a game involves only two players or competitors then it

is called two -person game .
 Payoff: outcome of a game when different alternatives are adopted by the
competing players are called payoffs. The game's payoff can be either (of
win or loss or draw).
 Value of the game : It is the expected payoff of play, when all the players of
the game follow their optimum strategies.
 The game is called fair if the value of the game is zero and unfair, if it is non-zero.
Saddle Point: If the maximin value equals the minimax value, the game is said
to have a saddle (equilibrium) point and the corresponding strategies are called
“Optimal Strategies”

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 6

Strategy

 A strategy for a player is defined as a set of rules or alternative courses of

action available to him in advance , by which players decide the course of
action that he should adopt. strategies are of two types:
 Pure strategy
 Mixed strategy
 optimum strategy
 Payoff matrix: The payoff in terms of gains or losses when players select
their particular strategies (course of actions) can be represented in a form of
matrix.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 7

Pure Strategy

 If the player select the same strategy each time, then it is referred to a pure
strategy.
 In this case each player know exactly what the other is going to do.
 There is a deterministic situation and the objective of the player is to
maximize gain or to minimize losses.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 8

Mixed Strategy

 When the players use a combination of strategies in each player always kept
guessing as to which course of action is to be selected by the other player at
a particular occasion then this is known as a mixed strategy
 There is probabilistic situation
 The objective of the player is to maximize expected gains or to minimize
losses.
 Makes this strategy is a selection among pure strategies with fixed
probabilities

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 9

Optimum Strategy

 A course of action or play which puts the player in most preferred position,
irrespective of the strategies of his competitor is called our optimum strategy
 Any deviation from this strategy results in a decreased payoff for the player

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 10

Assumptions of the Game Theory

 There are finite number of competitors.

 There is conflict of interests between them.
 Each player has available with him finite number of possible strategies(courses
of action).
 One player attempts to maximize gains and the other attempts to minimize losses.

 Players know all possible available choices but does not know which one is going to
be chosen.
 Players simultaneously select their respective courses of action.
 The payoff is fixed and determined in advance.
 Players have to make individual decisions without direct communication.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 11

Two‐person, zero‐sum game
Strategies of player 1;
Characterized by: Strategies of player 2;
Payoff table.

Strategy: predetermined rule that specifies

completely how one intends to respond to each
possible circumstance at each stage of game

Payoff table: shows gain (positive or negative)

for one player that would result from each
combination of strategies for the 2 players.
Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 12
The odds and evens game

Player
Player 1 takes evens, player 2 takes odds
Strategy 2
Each player simultaneously shows 1 or 2 fingers\
(odd)
Player 1 wins if total of fingers is even and loses if it
is odd;
1 2
vice‐versa for Player 2
Player 1 1 1 ‐1
2 ‐1 1
Each player has 2 strategies: which?
(even)
Payoff table:

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 13

Two Person Non-Zero-Sum Game

 Where profit of one player is not equal to loss of other player.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 14

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 15
 Nash equilibrium

• Nash equilibrium is a concept within game theory where the optimal outcome of
a game is where there is no incentive to deviate from their initial strategy.

• More specifically, the Nash equilibrium is a concept of game theory where the

optimal outcome of a game is one where no player has an incentive to deviate
from his chosen strategy after considering an opponent's choice.

• An individual can receive no incremental benefit from changing actions,

assuming other players remain constant in their strategies.

• A game may have multiple Nash equilibria or none at all.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 16

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 17
The Prisoner's Dilemma

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 18

Prototype example two person zero sum game

Two polititians running against each other for senate

Campaign plans must be made for final 2 days
Both polititians want to campaign in 2 key cities
Spend either 1 full day in each city or 2 full days in one
Campaign manager in each city assesses impact of possible
combinations for polititian and his opponent
Polititian shall use information to choose his best strategy on
how to use the 2 days
Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 19
 Prototype example

Two polititians running against each other for senate

Campaign plans must be made for final 2 days
Both polititians want to campaign in 2 key cities
Spend either 1 full day in each city or 2 full days in one

Campaign manager in each city assesses impact of possible combinations for

polititian and his opponent
Polititian shall use information to choose his best strategy on how to use
the
2 days

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 20

Formulation
Identify the 2 players, the strategies of each player
and the payoff table
Each player has 3 strategies:
1.Spend 1 day in each city
2.Spend 2 days in Bigtown
3.Spend 2 days in Megalopolis
Appropriate entries for payoff table for politician 1 are total
net votes won from the opponent resulting from 2 days of
campaigning.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 21

Formulation of Prototype Example

 Payoff Matrix
Total Net Votes Won by Politician
1
(in Units of 1,000 Votes)
Politician 2
1 2 3
Strategy

Politician 1 1 2 4
1
2 1 0 5
Dr. Keerti Jain, NIIT University, Neemrana
3 0 1 ‐1
10/03/2020 22
Maximin-Minimax principle

 Maximin principle : According to this principal, the player adopts a

pessimistic attitude and play safe. That is, his strategy is always that which
results in the best out of worst outcomes.
 In other words, the player A decide to play that strategy which corresponds to the
maximum of the minimum gain for his different course of action.
 Minimax principle: According to this principal, the player B always like to
play safe and, in that case,, he selects the strategy which corresponds to the
minimum of the maximum losses for his different courses of action.
 In other words the maximizing player adopts the maximin principle while the
minimizing player adopt stuff mini Max principle for the best of his plays of
the game

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 23

Variation 1 of Prototype Example

Total Net Votes Won by Politician

1
(in Units of 1,000 Votes)
Politician 2
Given the payoff table, 1 2 3
which strategy should Strategy
each player select?

Politician 1 1*# 2* 4
1
2 1 0# 5*
Interpretation
# stands 3for minimum
0 value and1* stands for ‐1#
Maximum Value
• Since Maximin =Minimax =1 =Value of the game(v)
(Game is Unfair)
• Saddle point exist at (1,1) implies optimum pure
strategy for politician 1 is 1 and for politician 2 is 1.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 24

Variation 2 of Prototype Example

Saddle Point
 Given the payoff table, which strategy should each player select? (equilibrium solution)

Total Net Votes Won by Politician 1

(in Units of 1,000 votes)
Politician 2
Strategy
1 2 3 Minimum
Maximum
Value
0 0
Politician 1 1 -3 -2 6 -3
2 2 00 2 0

Dr. Keerti Jain, NIIT University, Neemrana

3 5 Minimum -2 -4
10/03/2020
-4
25

Maximum 5 Value 0 6 It is a Fair Game

Example1: Method of Pure Strategy games (With
Saddle Point) 2X2 two- person zero sum game
 Determine the optimal strategies for both Player A and Player B and the value of
the game using maximin-minimax principle.

 It is a 2X2 two-person zero sum game

Player A Player B Row Minima
B1 B2

A1 0# 4* 0
A2 8* 4*# 4*
Column maxima 8 4#
# stands for minimum value and * stands for Maximum Value

Interpretation
• Since Maximin =Minimax =4 =Value of the game(v) (Game is Unfair)
• Saddle point exist at (2,2) implies optimum pure strategy for player A is A2 and
for Player B is B2.
Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 26
Example 2: Method of Pure Strategy games (With
Saddle Point) mxn two- person zero sum game
 Determine the optimal strategies for both firm A and firm B and the value of
the game using maximin-minimax principle.
 It is a 4X5 two-person zero sum game
Firm A Firm B Row Interpretation
B1 B2 B3 B4 B5 Minima
Since Maximin =Minimax =6
=Value of the game(v)
A1 3 -1# 4 6 7 -1 (Game is Unfair)
A2 -1# 8 2 4 12* -1
A3 16* 8 6#* 14* 12* 6* Saddle point exist at (3,3)
implies optimum pure
A4 1 11* -4# 2 1 -4
strategy for firm A is A3 and
Column 16 11 6# 14 12 for firm B is B3.
Maxima

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 27

# stands for minimum value and * stands for Maximum Value
Example 3: Method of Mixed Strategy games (Without
Saddle Point) 2X2 two- person zero sum game

 Determine the optimal strategies for both Player A and Player B and the value
of the game.
 It is a 2X2 two-person zero sum game
Player A Player B Row Minima
B1 B2

A1 25* 5# 5
A2 10# 15* 10*
Column maxima 25 15#
# stands for minimum value and * stands for Maximum Value
Here Maximin is not equal to minimax, therefore saddle point does not exist
We need to determine mixed strategies.
Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 28
 Solution
Step 1: Subtract the smaller payoff in each row from the larger one and the smaller
payoff in each column from the larger one
Player A Player B Difference
B1 B2

A1 25 5 20
A2 10 15 5
Difference 15 10
Step 2: Interchange each of these pairs of subtracted numbers found in step one as :

Player A Player B Interchanged

Difference
B1 B2

A1 25 5 5
A2 10 15 20
Interchanged
Dr. Keerti Jain, NIIT University, Neemrana 10 15 10/03/2020 29

Difference
 Step 3: Put each of the interchanged numbers over the sum of the pair of
numbers

Player A Player B Probability

B1 B2

A1 25 5 [5/(5+20)]=1/5
A2 10 15 [20/(5+20)]=4/5
Probability [10/(10+15)]= 2/5 [15/(10+15)]=3/5

Therefore the strategies for Player A are (A1, A2) =(1/5, 4/5)
for player B are (B1, B2)= (2/5, 3/5)

Value of game = (a11Xa22-a12Xa21)/(a11+a22)-(a12+a21)=325/25=13

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 30

Identifying Dominant Strategies
Sometimes in a matrix game, a player will have a strategy that, given all of the resulting
outcomes, would not be worth playing.

Such a strategy would not be worth playing if it is never better and sometimes worse
than some other strategy, regardless of the strategies of other players.

 For a given player, strategies that are never better and sometimes worse than other
strategies are
called dominated strategies. (We can think of this as equal or worse than all of the other
strategies.)

On the other hand, a dominant strategy is one that is sometimes better and never
worse than all other strategies, regardless of the strategies of the other players. (We can
think of this as equal or better than all of the other strategies.)
Dominance rule
 The principle of dominance states that if one strategy of a player dominates over
the other strategy in all conditions then the later strategy can be ignored. A
strategy dominates over the other only if it is preferable over other in all
conditions.
 The concept of dominance is especially useful for the evaluation of two-person
zero-sum games where a saddle point does not exist.
 Rule: 1. If all the elements of a column (say ith column) are greater than or
equal to the corresponding elements of any other column (say jth column), then
the ith column is dominated by the jth column and can be deleted from the
matrix.
 Rule: 2. If all the elements of a row (say ith row) are less than or equal to the
corresponding elements of any other row (say jth row), then the ith row is
dominated by the jth row and can be deleted from the matrix.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 32

Concept of dominated strategy is useful
for:

 Reducing size of payoff table to be considered.

 Identifying optimal solution of the game (special cases).

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 33

 Player B
Minimum of row
B1 B2 B3
A1 -5 10 20 -5 Maximin

Player A A2
5 -10 -10 -10  Given game have no saddle point
5 -20 -20 because Maximin is not equal to
A3 -20 Minimax. Therefore, further we
Must try to reduce the size of the
Maximum 5 20 given payoff matrix as further as
of 10 possible .
column Minimax
Note that : every element of column B3 is more than or equal to every corresponding element
of Row B2 Evidently , the choice of strategy B3, will always result in more losses as compared to
that of selecting the strategy B2.Thus , strategy B3 is inferior to B2. Hence, delete The B3
strategy from the payoff matrix. Reduced payoff matrix is …
Player B
B1 B2
A1
-5 10
Player A

A2 5 -10
5 -20
• Now, it may be noted that strategy A2 of player A is dominated by his A3 strategy (every
element of column A2 is more than or equal to every corresponding element of Row A3 ),
since the profit due to strategy A2 is greater than or equal to the profit due to strategy A3.
Hence, strategy A3(row 3) can be deleted from further consideration.
Thus the reduced payoff matrix becomes:

Player B
Minimum of row

B1 B2
Player A

A1 -5 10 -5 Maximin

A2 5 -10 -10

Maximum 5 10
of column
Minimax

Note : the maximin value is not equal to minimax value, hence there is no saddle point.
For this type of game situation, it is possible to obtain a solution by applying the concept
of mixed strategies.
 Stable and Unstable solutions

Using only pure strategies, games not having a

saddle point turned out to be unstable because 
Players wanted to change their strategies to improve their
positions
For games with mixed strategies, it is necessary that  for
optimal solution to be stable
This condition always holds for such games according to the
minimax theorem of game theory

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 36

Variation 3 Prototype Example

 Given the payoff table, which strategy should each player select?

Total Net Votes Won by Politician 1

(in Units of 1,000 votes)
Politician 2
Strategy
1 2 3 Minimum Maximum
-2 Value

Politician 1 1 0 -2 2 -2
2 5 42 -3 -3
3 2 3 -4 Saddle
-4 Point does not exist
Dr. Keerti Jain, NIIT University, Neemrana
Minimum 10/03/2020 37

Maximum 5 4 2
Value
Example 3 (Cont…)

 Originally suggested solution is an unstable solution (no

saddle point).
 Whenever one player’s strategy is predictable, the opponent
can take advantage of this information to improve his
position.
 An essential feature of a rational plan for playing a game
such is this one is that neither player should be able to
deduce which strategy the other will use.
 It is necessary to choose among alternative acceptable
strategies on some kind of random basis.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 38

Games with mixed strategies
 Whenever a game does not posses a saddle point, game theory
advises each player to assign a probability distribution over their
set of strategies.
 Let xi=probability that player 1 will use strategy I (i=1,2,…,m)
yi= Probability that player 2 will use strategy j (j=1,2,…n)
 Probabilities need to be non negative and add to 1.
 These plans (x1, x2, …, xm) and (y1, y2,…, yn) are usually referred
to as mixed strategies, and the original strategies are called pure
strategies.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 39

When the game is actually played…
 It is necessary for each player to use one of the pure
strategies.
 Pure strategy would be chosen by using some random
device to obtain a random observation from the
probability distribution specified by the mixed
strategy.
 This observation would indicate which particular
pure strategy to use.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 40

Expected Payoff

  Supposepoliticians 1 and 2 select the mixed strategies (x1, x2,
x3)=(½, ½, 0) and (y1, y2, y3)=(0, ½, ½)
 Each player could then flip a coin to determine which of his two
acceptable pure strategies he will actually use.
 Useful measure of performance is expected payoff:
Expected payoff for player 1=
 is payoff if player 1 uses pure strategy I and player 2 uses pure
strategy j.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 41

Expected payoff (contd…)

 4 possible payoffs (-2, 2, 4, -3), each with probability ¼.

 Expected payoff is ¼ (-2+2+4-3)=¼
 This measure of performance does not disclose anything about the risks
involved in playing the game.
 It indicates what the average payoff will tend to be if the game is played
many times.
 Game theory extends the concept of the minimax criterion to games that lack
a saddle point and thus need mixed strategies.

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 42

Minimax Criterion for mixed strategies

   given player should select the mixed strategy that maximizes the minimum
A
expected payoff to the player.
 Optimal mixed strategy for player 1 is the one that provides the guarantee
(minimum expected payoff) that is best (maximal).
 Value of best guarantee is the maximum value v
 Optimal strategy for player 2 provides the best (minimal) guarantee
(maximum expected loss)
 Value of best guarantee is the minimax value

Dr. Keerti Jain, NIIT University, Neemrana 10/03/2020 43