Premiere League 2017
Alex Chitsazan
September 7, 2017
Casey’s Fantasy Grades
Hey guys, so since I’m a nerd and do biology statistical analytics for a living, I wrote a quick bit of code to give us a visual view of how our draft went. I agreed with Casey so I thought it would be fun to mark this down and be able to look back at it and see how much a draft really affects our season. Hopefully at the end of the season, I can do another write up to see how much the draft affects outcome in this league.
Grades
Ignore the code but I converted Casey’s grades to a table for each player in the league.
library(knitr)
setwd("~/Desktop/fant/")
grades <- read.table("fant.csv", header = TRUE, sep = ",", stringsAsFactors = FALSE)
kable(grades, align = "c")| Players | QB | RB.Core | WR.core | TE | Starters | Bench |
|---|---|---|---|---|---|---|
| Ryan | A | B | A | A | A- | B+ |
| Ari | B | B+ | B+ | B- | B | C |
| Tommy | B | B- | B- | C+ | B- | B- |
| Sam | A- | C+ | A- | B+ | B | C |
| Casey | B- | B | A | A+ | B | B+ |
| Turtle | A | B | B | B+ | A- | C |
| Alex | B | A | A- | B | B+ | B+ |
| Urbach | B+ | A+ | C | A- | A- | C |
| Connor | A | B | A- | B+ | A- | B+ |
| Sean | B+ | B | B | A- | B | B |
Grades
The next thing I did was convert each draft grade to a GPA total, the same system U of O had (A+ = 4.3, A = 4.0, etc).
| Players | QB | RB.Core | WR.core | TE | Starters | Bench |
|---|---|---|---|---|---|---|
| Ryan | 4.0 | 3.0 | 4.0 | 4.0 | 3.7 | 3.3 |
| Ari | 3.0 | 3.3 | 3.3 | 2.7 | 3.0 | 2.0 |
| Tommy | 3.0 | 2.7 | 2.7 | 2.3 | 2.7 | 2.7 |
| Sam | 3.7 | 2.3 | 3.7 | 3.3 | 3.0 | 2.0 |
| Casey | 2.7 | 3.0 | 4.0 | 4.3 | 3.0 | 3.3 |
| Turtle | 4.0 | 3.0 | 3.0 | 3.3 | 3.7 | 2.0 |
| Alex | 3.0 | 4.0 | 3.7 | 3.0 | 3.3 | 3.3 |
| Urbach | 3.3 | 4.3 | 2.0 | 3.7 | 3.7 | 2.0 |
| Connor | 4.0 | 3.0 | 3.7 | 3.3 | 3.7 | 3.3 |
| Sean | 3.3 | 3.0 | 3.0 | 3.7 | 3.0 | 3.0 |
Make a bar plot of outcomes
A function that will plot all of the categories.
Overall team power index
Lastly, this is the one that I will think will be most interesting. This is the average of all of the categories combinded, so (QB + RB + WR + TE + STARTERS + BENCH) / 6. I call it just the overall team power index. At the end of the season I’ll do a regression of each of our team’s placement vs the overall team power index to see how draft (or casey’s interpretation) affects overall outcome. Also, I’ll say it’s pretty cool all of our teams fall in the A-B range, we are all looking good.. May the best team win!
averageplot <- function(table) {
rownames(table) <- table$Players
table <- table[,-1]
avg <- rowMeans(table)
assign("Average", data.frame(OPI=avg), envir = .GlobalEnv)
print(kable(data.frame(Team=names(avg), score=avg, row.names = 1:10)))
avg <- data.frame(Player = names(avg), Average = as.numeric(avg))
plotbar(avg$Average, color = "#CC79A7", title = "Overall Team Power Index")
}
averageplot(gpa)##
##
## Team score
## ------- ---------
## Ryan 3.666667
## Ari 2.883333
## Tommy 2.683333
## Sam 3.000000
## Casey 3.383333
## Turtle 3.166667
## Alex 3.383333
## Urbach 3.166667
## Connor 3.500000
## Sean 3.166667
Week 2 Update
Week 2 takeaways. Casey seems to have gotten the steal of the draft in Kareem Hunt. Secondly, Urby is screwed. Actually screwed is an understatement for urby, urby is beyond screwed. Actually if i could use a picture as a metaphor for Urby’s team it’d be: 
Good luck this week boys! Except for Casey. ! ## Stats
Games <- read.table("week2.csv", sep=",", header=T)
kable(Games, align='c')| TEAMOWNER | PF | PA | HOME | AWAY | DIV | STREAK |
|---|---|---|---|---|---|---|
| Casey | 258.5 | 208.0 | 1-0-0 | 1-0-0 | 2-0-0 | W2 |
| Connor | 235.5 | 244.5 | 0-1-0 | 1-0-0 | 1-1-0 | W1 |
| Sean | 231.0 | 190.0 | 0-1-0 | 1-0-0 | 1-1-0 | L1 |
| Ryan | 224.5 | 202.5 | 0-1-0 | 1-0-0 | 1-1-0 | W1 |
| Alex | 223.0 | 239.5 | 0-1-0 | 1-0-0 | 1-1-0 | L1 |
| Ari | 215.5 | 231.0 | 0-0-0 | 1-1-0 | 1-1-0 | L1 |
| Tommy | 212.5 | 203.0 | 0-1-0 | 1-0-0 | 1-1-0 | W1 |
| Turtle | 211.0 | 195.0 | 0-1-0 | 1-0-0 | 1-1-0 | W1 |
| Sam | 191.0 | 228.0 | 0-1-0 | 1-0-0 | 1-1-0 | L1 |
| Urbach | 169.0 | 230.0 | 0-2-0 | 0-0-0 | 0-2-0 | L2 |
Average$Players <- row.names(Average)
Complete <- merge(Average, Games, by.x = "Players",by.y = "TEAMOWNER")
rownames(Complete) <- sort(rownames(Average))
rownames(Complete) <- sort(rownames(Average))
PF <- lm(formula = Complete$OPI ~ Complete$PF)
scatter <- function(dataframe, x, y, title) {
dat <- dataframe
lin<- coef(lm(dat[[x]]~dat[[y]]))
ggplot(dat, aes_string(y, x, label = "rownames(Complete)"), environment = .GlobalEnv) +
geom_point() +
stat_smooth(method = "lm", se = FALSE, col = "red") +
geom_text(vjust = 0) +
ggtitle(title)
}
winPCT <- function(df) {
record <- read.table(text = as.character(df$DIV), sep="-", colClasses = "character")
colnames(record) <- c("Wins", "Losses", "Ties")
record$PCT <- as.numeric(record$Wins)/(as.numeric(record$Losses) + as.numeric(record$Wins))
record$PCT[which(record$PCT == Inf)] <- 1.00
return(record$PCT)
}
Complete$WinPCT <- winPCT(Complete)
plotbar2 <- function(category, color, title ) {
reorder_teams <- function(x) {
factor(Complete$Players, levels = Complete$Players[order(x, decreasing = TRUE)])
}
ggplot(gpa, aes(x = reorder_teams(category), y= category)) +
geom_bar(stat = "identity", fill=color) +
ggtitle(title) +
ylab(title) +
xlab("Jabroni")
}
plotbar2(Complete$WinPCT, color = "#009E73", title = "Win Percentage")
plotbar2(Complete$PF, color = "#E69F00", title = "Points For")
plotbar2(Complete$PA, color = "#0072B2", title = "Points Allowed")
Regressions
Intresting to see how the Overall Power Index (draft GPA) has affected the season.
scatter(Complete, x = "PF", y = "OPI", title = "PF vs OPI")
#scatter(Complete, x = "WinPCT", y = "OPI", title = "Win% vs OPI")
scatter(Complete, x = "PF", y = "PA", title = "PF vs PA")
Week 3 Update
Week 3 takeaways. Everyone is still alive and no one is undefeated! Urby still has a tiny fraction of a life left (..not really…), Kareem Hunt is still good, and Stefon Diggs is lookind reallll niceeeeee (Shameful plug) Sean had a one of the worst weeks in years this week. Unfotunatly it’s not looking like a bounce back season for him.. Actually I believe this photo was taken of Sean this past sunday… Sorry Seannnnnn…
Stats
winPCT <- function(df) {
record <- read.table(text = as.character(df$DIV), sep="-", colClasses = "character")
colnames(record) <- c("Wins", "Losses", "Ties")
record$PCT <- as.numeric(record$Wins)/(as.numeric(record$Losses) + as.numeric(record$Wins))
record$PCT[which(record$PCT == Inf)] <- 1.00
return(record$PCT)
}
add_columns <- function(filepath) {
Games <- read.table(filepath, sep=",", header=T)
Complete <- merge(Average, Games, by.x = "Players",by.y = "TEAMOWNER")
rownames(Complete) <- sort(rownames(Average))
rownames(Complete) <- sort(rownames(Average))
Complete$WinPCT <- winPCT(Complete)
Complete$PlusMinus <- Complete$PF - Complete$PA
return(Complete)
}
Complete <- add_columns("~/Desktop/fant/week3.csv")
kable(Complete, align='c')| Players | OPI | PF | PA | HOME | AWAY | DIV | STREAK | WinPCT | PlusMinus | |
|---|---|---|---|---|---|---|---|---|---|---|
| Alex | Alex | 3.383333 | 361.5 | 360.0 | 1-1-0 | 1-0-0 | 2-1-0 | W1 | 0.6666667 | 1.5 |
| Ari | Ari | 2.883333 | 334.0 | 395.5 | 0-1-0 | 1-1-0 | 1-2-0 | L2 | 0.3333333 | -61.5 |
| Casey | Casey | 3.383333 | 379.0 | 346.5 | 1-0-0 | 1-1-0 | 2-1-0 | L1 | 0.6666667 | 32.5 |
| Connor | Connor | 3.500000 | 351.5 | 381.5 | 0-2-0 | 1-0-0 | 1-2-0 | L1 | 0.3333333 | -30.0 |
| Ryan | Ryan | 3.666667 | 318.0 | 252.5 | 1-1-0 | 1-0-0 | 2-1-0 | W2 | 0.6666667 | 65.5 |
| Sam | Sam | 3.000000 | 355.5 | 346.5 | 0-1-0 | 2-0-0 | 2-1-0 | W1 | 0.6666667 | 9.0 |
| Sean | Sean | 3.166667 | 281.0 | 283.5 | 0-1-0 | 1-1-0 | 1-2-0 | L2 | 0.3333333 | -2.5 |
| Tommy | Tommy | 2.683333 | 349.5 | 319.0 | 0-1-0 | 2-0-0 | 2-1-0 | W2 | 0.6666667 | 30.5 |
| Turtle | Turtle | 3.166667 | 303.5 | 289.0 | 0-2-0 | 1-0-0 | 1-2-0 | L1 | 0.3333333 | 14.5 |
| Urbach | Urbach | 3.166667 | 263.0 | 322.5 | 0-2-0 | 1-0-0 | 1-2-0 | W1 | 0.3333333 | -59.5 |
plotbar2(Complete$WinPCT, color = "#009E73", title = "Win Percentage")
plotbar2(Complete$PF, color = "#E69F00", title = "Points For")
plotbar2(Complete$PA, color = "#0072B2", title = "Points Allowed")
plotbar2(Complete$PlusMinus, color = "#CC79A7", title = "Overall Plus Minus on Season")
Complete$AverageWin <- Complete$PlusMinus / 3
plotbar2(Complete$AverageWin, color = "#009E73", title = "Average Margin of Vitory")
Regressions
Too early to really see how draft affects correlation but not looking like draft has a big impact.
scatter(Complete, x = "PF", y = "OPI", title = "PF vs OPI")
scatter(Complete, x = "WinPCT", y = "OPI", title = "Win% vs OPI")
scatter(Complete, x = "PF", y = "PA", title = "PF vs PA")
Week 5 Update
Week 5 takeaways. We seem to be seperating out finally.
Tyler and Turtle are gonna need a huge miracle to make it to the playoffs unfortunately, as they sit at 1-4.
On the oppisite end, Ryan and Sam gotta feel real good right now about their playoff chances as they are currently in the lead at 4-1. Sam’s team is undeniably doing the best this year while stats are showing that Singer’s team is fairing well in the luck column. He ranks 5th in points to his credit, but he’s dead last (by far) in points allowed. Don’t choke Singer!
My team had a pretty good week but after losing OBJ and Dalvin Cook, my future is not seeming bright so well see what happens.
Connor’s team is decent but he is in bye week hell this week so this week might end up in an L.
Tommy Ari and Casey are coming into a big week as they sit 2-3 and this week might be do or die. 3-3 sounds a hell of a lot better than 2-4 so win this week!
On a seperate note, It’s important to note that Casey seems to be in a freefall. He’s averaging just 103 points in the last two games and has lost three straight. On top of that he just spent HALF ($75!!) on latavious murray this week, putting his remaining budget at just $49. He’s got to be worried.
Stats
Complete <- add_columns("~/Desktop/fant/week_5.csv")
kable(Complete, align='c')| Players | OPI | PF | PA | HOME | AWAY | DIV | STREAK | WinPCT | PlusMinus | |
|---|---|---|---|---|---|---|---|---|---|---|
| Alex | Alex | 3.383333 | 585.0 | 573.5 | 2-1-0 | 1-1-0 | 3-2-0 | W1 | 0.6 | 11.5 |
| Ari | Ari | 2.883333 | 553.5 | 621.5 | 0-2-0 | 2-1-0 | 2-3-0 | L1 | 0.4 | -68.0 |
| Casey | Casey | 3.383333 | 583.5 | 595.5 | 1-1-0 | 1-2-0 | 2-3-0 | L3 | 0.4 | -12.0 |
| Connor | Connor | 3.500000 | 613.5 | 596.5 | 1-2-0 | 2-0-0 | 3-2-0 | W2 | 0.6 | 17.0 |
| Ryan | Ryan | 3.666667 | 555.5 | 408.0 | 2-1-0 | 2-0-0 | 4-1-0 | W4 | 0.8 | 147.5 |
| Sam | Sam | 3.000000 | 634.5 | 563.0 | 1-1-0 | 3-0-0 | 4-1-0 | W3 | 0.8 | 71.5 |
| Sean | Sean | 3.166667 | 496.0 | 477.5 | 2-1-0 | 1-1-0 | 3-2-0 | W2 | 0.6 | 18.5 |
| Tommy | Tommy | 2.683333 | 545.5 | 570.0 | 0-2-0 | 2-1-0 | 2-3-0 | L2 | 0.4 | -24.5 |
| Turtle | Turtle | 3.166667 | 490.5 | 529.0 | 0-2-0 | 1-2-0 | 1-4-0 | L3 | 0.2 | -38.5 |
| Urbach | Urbach | 3.166667 | 463.0 | 586.0 | 0-3-0 | 1-1-0 | 1-4-0 | L2 | 0.2 | -123.0 |
plotbar2(Complete$WinPCT, color = "#009E73", title = "Win Percentage")
plotbar2(Complete$PF, color = "#E69F00", title = "Points For")
plotbar2(Complete$PA, color = "#0072B2", title = "Points Allowed")
plotbar2(Complete$PlusMinus, color = "#CC79A7", title = "Overall Plus Minus on Season")
Complete$AverageWin <- Complete$PlusMinus / 5
plotbar2(Complete$AverageWin, color = "#009E73", title = "Average Margin of Vitory")
Regressions
Too early to really see how draft affects correlation but not looking like draft has a big impact.
scatter(Complete, x = "PF", y = "OPI", title = "PF vs OPI")
scatter(Complete, x = "WinPCT", y = "OPI", title = "Win% vs OPI")
scatter(Complete, x = "PF", y = "PA", title = "PF vs PA")
Week 7
Stats
Quick Takeaways:
- Singer is a lucky dude. By far least points for
- Lots of teams at 4-3
Complete <- add_columns("~/Desktop/fant/week7.csv")
kable(Complete, align='c')| Players | OPI | PF | PA | HOME | AWAY | DIV | STREAK | WinPCT | PlusMinus | |
|---|---|---|---|---|---|---|---|---|---|---|
| Alex | Alex | 3.383333 | 842.5 | 801.5 | 2-2-0 | 2-1-0 | 4-3-0 | L1 | 0.5714286 | 41.0 |
| Ari | Ari | 2.883333 | 803.0 | 856.5 | 0-2-0 | 3-2-0 | 3-4-0 | W1 | 0.4285714 | -53.5 |
| Casey | Casey | 3.383333 | 782.0 | 773.0 | 2-1-0 | 1-3-0 | 3-4-0 | L1 | 0.4285714 | 9.0 |
| Connor | Connor | 3.500000 | 841.5 | 839.5 | 1-3-0 | 3-0-0 | 4-3-0 | W1 | 0.5714286 | 2.0 |
| Ryan | Ryan | 3.666667 | 786.5 | 622.5 | 3-1-0 | 3-0-0 | 6-1-0 | W6 | 0.8571429 | 164.0 |
| Sam | Sam | 3.000000 | 855.5 | 769.0 | 1-2-0 | 4-0-0 | 5-2-0 | L1 | 0.7142857 | 86.5 |
| Sean | Sean | 3.166667 | 729.5 | 749.0 | 3-1-0 | 1-2-0 | 4-3-0 | W1 | 0.5714286 | -19.5 |
| Tommy | Tommy | 2.683333 | 783.5 | 697.0 | 2-2-0 | 2-1-0 | 4-3-0 | W2 | 0.5714286 | 86.5 |
| Turtle | Turtle | 3.166667 | 713.5 | 845.0 | 0-3-0 | 1-3-0 | 1-6-0 | L5 | 0.1428571 | -131.5 |
| Urbach | Urbach | 3.166667 | 666.0 | 850.5 | 0-4-0 | 1-2-0 | 1-6-0 | L4 | 0.1428571 | -184.5 |
plotbar2(Complete$WinPCT, color = "#009E73", title = "Win Percentage")
plotbar2(Complete$PF, color = "#E69F00", title = "Points For")
plotbar2(Complete$PA, color = "#0072B2", title = "Points Allowed")
plotbar2(Complete$PlusMinus, color = "#CC79A7", title = "Overall Plus Minus on Season")
Complete$AverageWin <- Complete$PlusMinus / 7
plotbar2(Complete$AverageWin, color = "#009E73", title = "Average Margin of Vitory")
Regressions
The first graph is interesting, youre vertical distance from regression (line) is how much better or worse (above or below) you are doing compared to where you should be by points for and/or how lucky you are.
scatter(Complete, y = "PF", x = "WinPCT", title = "PF vs WinPCT")
scatter(Complete, x = "PF", y = "OPI", title = "PF vs OPI")
scatter(Complete, x = "WinPCT", y = "OPI", title = "Win% vs OPI")
scatter(Complete, x = "PF", y = "PA", title = "PF vs PA")
Week 8
Stats
Quick Takeaways:
Big matchups this week! League is tight!
Complete <- add_columns("~/Desktop/fant/week8.csv")
kable(Complete, align='c')| Players | OPI | PF | PA | HOME | AWAY | DIV | STREAK | WinPCT | PlusMinus | |
|---|---|---|---|---|---|---|---|---|---|---|
| Alex | Alex | 3.383333 | 1001.5 | 909.0 | 3-2-0 | 2-1-0 | 5-3-0 | W1 | 0.625 | 92.5 |
| Ari | Ari | 2.883333 | 890.5 | 985.5 | 0-3-0 | 3-2-0 | 3-5-0 | L1 | 0.375 | -95.0 |
| Casey | Casey | 3.383333 | 899.0 | 875.0 | 3-1-0 | 1-3-0 | 4-4-0 | W1 | 0.500 | 24.0 |
| Connor | Connor | 3.500000 | 943.5 | 956.5 | 1-3-0 | 3-1-0 | 4-4-0 | L1 | 0.500 | -13.0 |
| Ryan | Ryan | 3.666667 | 906.5 | 758.0 | 3-2-0 | 3-0-0 | 6-2-0 | L1 | 0.750 | 148.5 |
| Sam | Sam | 3.000000 | 963.0 | 928.0 | 1-2-0 | 4-1-0 | 5-3-0 | L2 | 0.625 | 35.0 |
| Sean | Sean | 3.166667 | 858.5 | 836.5 | 3-1-0 | 2-2-0 | 5-3-0 | W2 | 0.625 | 22.0 |
| Tommy | Tommy | 2.683333 | 885.0 | 782.0 | 2-2-0 | 3-1-0 | 5-3-0 | W3 | 0.625 | 103.0 |
| Turtle | Turtle | 3.166667 | 798.5 | 946.5 | 0-4-0 | 1-3-0 | 1-7-0 | L6 | 0.125 | -148.0 |
| Urbach | Urbach | 3.166667 | 801.5 | 970.5 | 0-4-0 | 2-2-0 | 2-6-0 | W1 | 0.250 | -169.0 |
plotbar2(Complete$WinPCT, color = "#009E73", title = "Win Percentage")
plotbar2(Complete$PF, color = "#E69F00", title = "Points For")
plotbar2(Complete$PA, color = "#0072B2", title = "Points Allowed")
plotbar2(Complete$PlusMinus, color = "#CC79A7", title = "Overall Plus Minus on Season")
Complete$AverageWin <- Complete$PlusMinus / 7
plotbar2(Complete$AverageWin, color = "#009E73", title = "Average Margin of Vitory")
Regressions
scatter(Complete, y = "PF", x = "WinPCT", title = "PF vs WinPCT")
scatter(Complete, x = "PF", y = "PA", title = "PF vs PA")
scatter(Complete, x = "PF", y = "OPI", title = "PF vs OPI")
scatter(Complete, x = "WinPCT", y = "OPI", title = "Win% vs OPI")