Jay Ghosh

Street Dice

Street Dice Simulation

 I worked on simulating different street dice strategies

You can review the rules for street dice here.

I calculated the odds of the game:

Instant Win Odds: 11.11% [7,11] 
Instant Loss Odds: 8.33% [2,3,12]  
Point Value Odds: 
2: 2.78% 
3: 2.78% 
4: 5.56% 
5: 5.56% 
6: 8.33% 
7: 8.33% [Loss] 
8: 8.33% 
9: 5.56% 
10: 5.56% 
11: 2.78% 
12: 2.78%

The following three Monte Carlo simulations show 1,000 players each in a circle of 10 dice players, playing 1,000 rounds, with each player starting with $10,000 and betting $100 each round. The only difference between the three being the strategy examined. The strategies being whether the player always passes (bets for the shooter), doesn’t pass (bets against the shooter), or determines to pass or not randomly.

Pass Only:

This shows the strategy of betting for the shooter each round

Each line represents a player, the x axis represents the rounds, the y axis represents the players’ cash

This is a distribution curve of the players’ money at the end of the simulation. The x axis represents the dollar value

Here are the summary statistics:

count   1000.000000
mean   68531.006349
std     6192.250700
min    48226.984127
25%    64371.726190
50%    68664.880952
75%    72437.103175
max    91189.285714

Pass Randomly:

This shows the strategy of randomly deciding to bet for or against the shooter.

Each line represents a player, the x axis represents the rounds, the y axis represents the players’ cash

This is a distribution curve of the players’ money at the end of the simulation. The x axis represents the dollar value

Here are the summary statistics:

count    1000.000000
mean    93511.897222
std      7574.498375
min     70944.444444
25%     88405.952381
50%     93422.420635
75%     98502.777778
max    120413.095238

No Pass Only

This shows the strategy of betting against the shooter each round

Each line represents a player, the x axis represents the rounds, the y axis represents the players’ cash

This is a distribution curve of the players’ money at the end of the simulation. The x axis represents the dollar value

Here are the summary statistics:

count    1000.000000
mean   118535.532937
std      8816.742292
min     94180.555556
25%    112335.218254
50%    118262.500000
75%    124353.273810
max    150313.492063

A Comparison:

In conclusion:

By looking at the mean and median of each of our simulations I can rank the strategies from best to worst: Don’t Pass Only, Pass Randomly, and Pass Only. So in summary, when you’re playing street dice, betting against the shooter will yield better long term results.

 

You can take a look at my code here.