Baseball Team Season Save Leaders with the Lowest Totals since 2000

Introduction

The Lahman database is a tremendously useful dataset that has been maintained by many individuals over the years. This has been compiled into an R package that makes using this database easy.

Question: Which pitchers led their teams in saves each year? Of these, who had the lowest total in baseball that year?

Methods

Here I loaded the relevant packages and the Pitching table.

library(Lahman)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(ggplot2)
data(Pitching)
data(Master)
Pitching <- tbl_df(Pitching)
Master <- tbl_df(Master)

Classical R code would do a split, apply, and combine strategy. I wanted to see if I could do this in dplyr. The examples are pretty good for finding things like HR leaders although there are no examples of doing this by team. I tried group_by but it didn’t work like I thought it might, so I resorted to using a helper function.

This function takes a year as the input. It filters the pitching data for that year, groups players by team, and finds the team saves leader. It then takes those team saves leaders and finds out which one (or more in the case of ties) had the fewest saves that season.

team_SV_leaders <- function(year) {
  max_saves <- Pitching %>% filter(yearID == year) %>% 
    select(playerID, yearID, stint, teamID, SV) %>% 
    group_by(teamID) %>%
    top_n(1, SV) %>%        # Determine team saves leader
    arrange(SV)
  max_saves[which(max_saves$SV == min(max_saves$SV)), ]
}

For example, in 2012 Glen Perkins led the Minnesota Twins with 16 saves. His was the lowest save total for any team’s closer that year.

team_SV_leaders(2012)

## Source: local data frame [1 x 5]
## Groups: teamID [1]
## 
##    playerID yearID stint teamID    SV
##       <chr>  <int> <int> <fctr> <int>
## 1 perkigl01   2012     1    MIN    16

Results

I bound together the results from calling this function on every year since 2000. This resulted in the following table.

SV_table <- data_frame()
for(i in 2000:2015) {
  SV_table <- bind_rows(SV_table, team_SV_leaders(i))
}
SV_table <- SV_table %>% inner_join(Master) %>% 
  select(yearID, nameFirst, nameLast, stint, teamID, SV)

## Joining, by = "playerID"

as.data.frame(SV_table)

##    yearID nameFirst  nameLast stint teamID SV
## 1    2000      Ryan Kohlmeier     1    BAL 13
## 2    2001     Buddy     Groom     1    BAL 11
## 3    2002     Keith    Foulke     1    CHA 11
## 4    2002   Antonio     Osuna     1    CHA 11
## 5    2003  Franklyn    German     1    DET  5
## 6    2003     Chris     Mears     1    DET  5
## 7    2004    Jeremy   Affeldt     1    KCA 13
## 8    2004       Bob   Wickman     1    CLE 13
## 9    2005  Fernando    Rodney     1    DET  9
## 10   2005    Ugueth    Urbina     1    DET  9
## 11   2006     Tyler    Walker     2    TBA 10
## 12   2007      Eric     Gagne     1    TEX 16
## 13   2007     Chris       Ray     1    BAL 16
## 14   2008    Jensen     Lewis     1    CLE 13
## 15   2009     Jason    Frasor     1    TOR 11
## 16   2010     J. C. Gutierrez     1    ARI 15
## 17   2011      Matt     Capps     1    MIN 15
## 18   2012      Glen   Perkins     1    MIN 16
## 19   2013     Heath      Bell     1    ARI 15
## 20   2014      Jake  Petricka     1    CHA 14
## 21   2015  Fernando    Rodney     1    SEA 16

ggplot(SV_table, aes(x = yearID, y = SV)) +
  geom_point(shape=1) +
  ggtitle("Lowest Team Saves Leader by Year")

Discussion

In 2003 the Detroit Tigers had a record of 49-113 and not surprisingly had only 27 saves. Of these, Franklyn German and Chris Mears both had 5 to lead the team. This figure seems to be the principal reason for why the save leaders had such low totals.

The Tigers were improved in 2005 but still only went 71-91. Fernando Rodney and Ugueth Urbina both led the team with 9 saves. Urbina was traded in June to the Phillies, limiting his total saves for the Tigers. That was his last season in the MLB because he was arrested in the winter for attempted murder. He was ultimately convicted and spent 7 years in prison (source) .

Interestingly, Troy Percival, who as of this writing, is #10 on the career saves list, was also on the team and only had 8 saves. His season ended early after an injury in July, and after another year on the DL, he retired from playing baseball.

Conclusion

It was not common for a team to have a saves leader who saved fewer than 10 games.