-
Notifications
You must be signed in to change notification settings - Fork 0
/
Simulating ND vs Louisville.R
257 lines (187 loc) · 7.65 KB
/
Simulating ND vs Louisville.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
#Inspire a Better Tomorrow
set.seed(11)
#Packages
library(tidyverse) #Gravy
library(rvest) #Web Scraping
library(fitdistrplus) #Distribution Fitting
# Data for analysis is included in the github repository (https://github.com/VinnyC-Analytics/Simulating_NDvsLOU)
# Obfuscated code for web scraping included as comments for guide to using rvest
# To execute this code, read-in data from github repo as "Scores.BrianKelly" and "Scores.ScottSatt" as appropriate.
#------------------------------------GAME SCORES FOR DISTRIBUTION FITTING-------------------------------------------------
#
# GAME SCORES FOR DISTRIBUTION FITTING
#
#
Capacity <- 40000
# Functions:
# Web Scrape Game Score. URL has been removed. Use function as reference only.
# GetGameScores <- function(INPUT_YEAR , INPUT_TEAM){
# templist <- list()
#
# for(year in INPUT_YEAR){
#
# url <- paste0("SOME_WEBSITE_FOR_DATA",year,INPUT_TEAM,"/index.html")
#
# temp <- url %>%
# read_html() %>%
# html_nodes(xpath=' INSERT XPATH HERE ') %>%
# html_table() %>%
# as.data.frame()
#
# templist[[year]] <- temp
#
# }
#
# OUTPUT.GetGameScores <- bind_rows(templist)
# return(OUTPUT.GetGameScores)
#
# remove(templist)
# gc()
#
# }
CleanGameScores <- function(INPUT){
bind_rows(INPUT) %>%
dplyr::select(1:3,5) %>%
filter(!grepl("@" , Result)) %>%
mutate(GameMonth =
case_when(
substr(Date,1,2) == "08" ~ "August/September",
substr(Date,1,2) == "09" ~ "August/September",
substr(Date,1,2) == "10" ~ "October",
substr(Date,1,2) == "11" ~ "November",
substr(Date,1,2) == "12" ~ "December/January",
substr(Date,1,2) == "01" ~ "December/January",
TRUE ~ as.character(Date)
)
) %>%
dplyr::select(GameMonth,
Result,
Attendance) %>%
separate(Result,c("PtsFor","PtsAgainst"), sep = "-") %>%
mutate(PtsFor = gsub("W ","",PtsFor),
PtsFor = gsub("L ","",PtsFor)) %>%
mutate(Attendance = gsub(",","",Attendance)) %>%
mutate(Attendance = as.numeric(Attendance),
PtsFor = as.numeric(PtsFor),
PtsAgainst = as.numeric(PtsAgainst)) %>%
mutate(PtsForAdj = ifelse(Attendance >= Capacity , PtsFor , PtsFor * 0.70),
PtsAgainstAdj = ifelse(Attendance >= Capacity , PtsAgainst , PtsAgainst * 0.70)) %>%
mutate(PtsForAdj = ifelse(PtsForAdj == 0 , 0.00001 , PtsForAdj),
PtsAgainstAdj = ifelse(PtsAgainstAdj == 0 , 0.00001 , PtsAgainstAdj))
}
Years.BK <- as.character(c(2010:2018)) #Brian Kelly Coaching Years at ND
Years.SS <- as.character(c(2014:2018)) #Scott Satterfield Total Coaching Years (limited experience). Data starts at 2014
tempSCORES_BK <- GetGameScores(Years.BK , "Notre Dame")
tempSCORES_SS <- GetGameScores(Years.SS , "Appalachian State")
Scores.BrianKelly <- CleanGameScores(tempSCORES_BK)
Scores.ScottSatt <- CleanGameScores(tempSCORES_SS)
remove(Capacity , Years.BK , Years.SS)
remove(tempSCORES_BK , tempSCORES_SS)
gc()
#--------------------------------------------DISTRIBUTION FITTING: Brian Kelly--------------------------------------------------
#August / September
#
#
BK.AugSept <- Scores.BrianKelly %>%
filter(GameMonth == "August/September") %>%
dplyr::select(5:6)
#Step 1 - Eyeballs on the distribution
#Both have a positive skew
plotdist(BK.AugSept$PtsForAdj , histo = TRUE , demp = TRUE) #normal-ish
plotdist(BK.AugSept$PtsAgainstAdj , histo = TRUE , demp = TRUE) #positive skew
#Step2 - Distribution Fitting
#PtsFor
fitND.n <- fitdist(BK.AugSept$PtsForAdj , "norm") #Why not try it
fitND.w <- fitdist(BK.AugSept$PtsForAdj , "weibull")
fitND.ln <- fitdist(BK.AugSept$PtsForAdj , "lnorm")
fitND.g <- fitdist(BK.AugSept$PtsForAdj , "gamma")
#PtsAgainst
fitND.n2 <- fitdist(BK.AugSept$PtsAgainstAdj , "norm") linke
fitND.w2 <- fitdist(BK.AugSept$PtsAgainstAdj , "weibull")
fitND.ln2 <- fitdist(BK.AugSept$PtsAgainstAdj , "lnorm")
fitND.g2 <- fitdist(BK.AugSept$PtsAgainstAdj , "gamma")
TheLegend <- c("Weibull", "Lognormal","Gamma")
denscomp(list(fitND.w, fitND.ln, fitND.g), legendtext = TheLegend)
cdfcomp(list(fitND.w, fitND.ln,fitND.g), legendtext = TheLegend)
fitND.g$aic
fitND.g$estimate
summary(fitND.g)
rand.ND_PF <- rgamma(n = 1000000,
shape = fitND.g$estimate[[1]],
rate = fitND.g$estimate[[2]])
denscomp(list(fitND.n2, fitND.w2, fitND.g2))
cdfcomp(list(fitND.n2, fitND.w2, fitND.g2))
fitND.n2$aic
fitND.w2$aic
fitND.g2$aic
fitND.n2$estimate
rand.ND_PA <- rnorm(n = 1000000,
mean = fitND.n2$estimate[[1]],
sd = fitND.n2$estimate[[2]])
sim.ND_Scores <- cbind(rand.ND_PF,rand.ND_PA) %>% as.data.frame()
remove(rand.ND_PF , rand.ND_PA)
remove(fitND.g , fitND.ln , fitND.n , fitND.w)
remove(fitND.g2 , fitND.ln2 , fitND.n2 , fitND.w2)
gc()
#-----------------------------------------DISTRIBUTION FITTING: Scott Satterfield--------------------------------------------------
#Step 1 - Eyeballs on the distribution
#Points For Adjusted
temp.SCOTTPF <-filter(Scores.ScottSatt , GameMonth == "August/September")[,5]
plotdist(temp.SCOTTPF , histo = TRUE , demp = TRUE)
#Step2 - Distribution Fitting
#PtsFor - August/September
fit.n <- fitdist(filter(Scores.ScottSatt , GameMonth == "August/September")[,5] , "norm")
fit.w <- fitdist(filter(Scores.ScottSatt , GameMonth == "August/September")[,5] , "weibull")
fit.ln <- fitdist(filter(Scores.ScottSatt , GameMonth == "August/September")[,5] , "lnorm")
fit.g <- fitdist(filter(Scores.ScottSatt , GameMonth == "August/September")[,5] , "gamma")
TheLegend <- c("Weibull", "Lognormal","Gamma")
denscomp(list(fit.n,fit.w, fit.ln, fit.g))
cdfcomp(list(fit.n,fit.w, fit.ln, fit.g))
fit.n$aic
fit.w$aic
fit.ln$aic
fit.g$aic
fit.w$estimate
rand.L_PF <- rweibull(n = 1000000,
shape = fit.w$estimate[[1]],
scale = fit.w$estimate[[2]])
#Points Against Adjusted
temp.SCOTTPA <- filter(Scores.ScottSatt , GameMonth == "August/September")[,6]
hist(temp.SCOTTPA)
plotdist(temp.SCOTTPA , histo = TRUE , demp = TRUE) #Negative Skew
#PtsAgainst - August/September
fit.n <- fitdist(temp.SCOTTPA , "norm")
fit.w <- fitdist(temp.SCOTTPA , "weibull")
fit.ln <- fitdist(temp.SCOTTPA , "lnorm")
fit.g <- fitdist(temp.SCOTTPA , "gamma")
denscomp(list(fit.w, fit.ln, fit.g))
cdfcomp(list(fit.w, fit.ln, fit.g))
fit.n$aic
fit.w$aic
fit.ln$aic
fit.g$aic
#gamma has the lowest aic
summary(fit.g)
fit.g$estimate
rand.L_PA <- rgamma(n = 1000000,
shape = fit.g$estimate[[1]],
rate = fit.g$estimate[[2]])
hist(rand.L_PA)
sim.L_Scores <- cbind(rand.L_PF,rand.L_PA) %>% as.data.frame()
remove(temp.SCOTTPA , temp.SCOTTPF)
remove(fit.g , fit.ln , fit.n , fit.w)
remove(rand.L_PA , rand.L_PF)
gc()
#------------------------------------SIMULATED GAME SCORES-------------------------------------------------
#
# GAME SCORES FOR PREDICTION
#
#
sim.tempbind <- cbind(sim.ND_Scores , sim.L_Scores) %>%
apply(2 , function(x) ifelse(x < 0 , 0 , x)) %>%
as.data.frame() %>%
mutate(ND_Score = (rand.ND_PF * 0.8) + (rand.L_PA * 0.2),
Lou_Score = (rand.L_PF * 0.8) + (rand.ND_PA * 0.2),
ND_Victory = ifelse(ND_Score > Lou_Score , 1 , 0))
#WIN OUTCOME: ND vs Louisville
prop.table(table(sim.tempbind$ND_Victory))