This is the code from my blog post for A.I. Sports.
1,000,001 simulations takes some time to run. If you lower the num_sims it will be quicker to play around with on your own.
set.seed(1234)
while (i <= num_sims) {
##### Wild Card Weekend
# AFC
wc.1.afc <- simulate.game(afc.3, afc.6)
wc.2.afc <- simulate.game(afc.4, afc.5)
# NFC
wc.1.nfc <- simulate.game(nfc.3, nfc.6)
wc.2.nfc <- simulate.game(nfc.4, nfc.5)
##### Divisional Weekend
# AFC
dw.1.afc <- simulate.game(afc.1, wc.2.afc)
dw.2.afc <- simulate.game(afc.2, wc.1.afc)
# NFC
dw.1.nfc <- simulate.game(nfc.1, wc.2.nfc)
dw.2.nfc <- simulate.game(nfc.2, wc.1.nfc)
##### Championship Weekend
# AFC
cw.1.afc <- simulate.game(dw.1.afc, dw.2.afc)
# NFC
cw.1.nfc <- simulate.game(dw.1.nfc, dw.2.nfc)
##### Superbowl
champ <- simulate.game(cw.1.afc, cw.1.nfc)
#print(paste0("This is the winner ",champ))
results.all <- c(
i,
wc.1.afc,
wc.2.afc,
wc.1.nfc,
wc.2.nfc,
dw.1.afc,
dw.2.afc,
dw.1.nfc,
dw.2.nfc,
cw.1.afc,
cw.1.nfc,
champ
)
simulation.results <- c(simulation.results, results.all)
i <- i + 1
}# View results
kable(all.chances.df)| team | Divisional | Conference | Superbowl | Champion |
|---|---|---|---|---|
| KC | 0% | 66.36% | 44.88% | 27.13% |
| NO | 0% | 58.41% | 32.96% | 17.74% |
| LAR | 0% | 59.6% | 31.67% | 15.43% |
| NE | 0% | 64.13% | 27.34% | 14.29% |
| CHI | 65.25% | 28.13% | 13.89% | 5.45% |
| SEA | 57.82% | 25.36% | 11.94% | 4.71% |
| LAC | 47.66% | 18.16% | 9.34% | 4.5% |
| BAL | 52.34% | 15.48% | 6.89% | 3.16% |
| HOU | 56.79% | 20.73% | 6.7% | 3% |
| IND | 43.21% | 15.14% | 4.85% | 2.12% |
| DAL | 42.18% | 16.22% | 5.14% | 1.29% |
| PHI | 34.75% | 12.26% | 4.4% | 1.16% |
Remember probabilities are just that... probabilities... It doesn't mean that they WILL win. But they have the best chance to win. I love the playoffs! Anything can happen! Let's see how this plays out!
Feel free to play around with the percentages in the "playoff_games.csv" file to see how it affects the simulation.