sandy4321 / baseballdbr Goto Github PK

---
output:
  md_document:
    variant: markdown_github
---

<!-- README.md is generated from README.Rmd. Please edit that file -->

```{r, echo = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "README-",
  warning = FALSE,
  message=FALSE
)
library(baseballDBR)
```
# BaseballDBR
[![Build Status](https://travis-ci.org/keberwein/baseballDBR.png?branch=master)](https://travis-ci.org/keberwein/baseballDBR)
[![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/baseballDBR)](http://www.r-pkg.org/badges/version/baseballDBR)
[![Project Status: Active - The project has reached a stable, usable state and is being actively developed.](http://www.repostatus.org/badges/latest/active.svg)](http://www.repostatus.org/#active)

# Install

* The latest development version from GitHub:

```{r eval=FALSE}
devtools::install_github("keberwein/baseballDBR")
```

# Gathering Data

The `baseballDBR` package requires data that is formatted similar to the [Baseball Databank](https://github.com/chadwickbureau/baseballdatabank) or Sean Lahman's [Baseball Database](http://www.seanlahman.com/baseball-archive/statistics/). The `Lahman` package is great source for these data. The package also contains the `get_bbdb()` function, which allows us to download the most up-to-date tables directly from the Chadwick Bureau's GitHub repository. For example, we can easily load the "Batting" table into our R environment.

```{r}
library(baseballDBR)

get_bbdb(table = "Batting")
head(Batting)
```

# Adding Basic Metrics

Simple batting metrics can be easily added to any batting data frame. For example, we can add slugging percentage, on-base percentage and on-base plus slugging. Note that OPS and OBP appears as "NA" for the years before IBB was tracked. 

```{r}
library(baseballDBR)

Batting$SLG <- SLG(Batting)

Batting$OBP <- OBP(Batting)

head(Batting, 3)
```

# Advanced Metrics

The package includes a suite of advanced metrics such as wOBA, RAA, and FIP among others. Many of the advanced metrics require multiple tables. For example, the wOBA metric requires the Batting, Pitching, and Fielding tables in order to establish a player's regular defensive position.

```{r}
library(baseballDBR)

get_bbdb(table = c("Batting", "Pitching", "Fielding"))

Batting$wOBA <- wOBA(Batting, Pitching, Fielding, Fangraphs = T)
head(Batting, 3)
```

The code above uses [Fangraphs](http://www.fangraphs.com/guts.aspx?type=cn) wOBA values. The default behavior is to uses [Tom Tango's formula](http://www.insidethebook.com/ee/index.php/site/article/woba_year_by_year_calculations/). Other options include `Sep.Leagues`, which may act as a buffer to any bias created by the designated hitter.

```{r}
library(baseballDBR)

get_bbdb(table = c("Batting", "Pitching", "Fielding"))

Batting$wOBA <- wOBA(Batting, Pitching, Fielding, Fangraphs = F, Sep.Leagues = T)
head(Batting, 3)
```

We can also produce a data frame that only shows the wOBA multipliers. Notice the Fangraphs wOBA multipliers slightly differ from the Tango multipliers.

```{r}
library(baseballDBR)

get_bbdb(table = c("Batting", "Pitching", "Fielding"))

fangraphs_woba <- wOBA_values(Batting, Pitching, Fielding, Fangraphs=T)
head(fangraphs_woba, 3)

tango_woba <- wOBA_values(Batting, Pitching, Fielding, Fangraphs=F)
head(tango_woba, 3)


```

# Create Local Database

A relational database is not needed to work with these data. However, we may want to store the data to be called more quickly at a later time. We can download all of the tables at once with the `get_bbdb()` function and then write them to an empty schema in our favorite database. The example uses a newly created PostgreSQL instance, but other database tools can be used assuming an appropriate R package exists.

```{r, eval=F}
library(baseballDBR)
library(RPostgreSQL)

# Load all tables into the Global Environment.
get_bbdb(AllTables = TRUE)

# Make a list of all data frames.
dbTables <- names(Filter(isTRUE, eapply(.GlobalEnv, is.data.frame)))

# Load data base drivers and load all data frames in a loop.
drv <- dbDriver("PostgreSQL")
con <- dbConnect(drv, host= "localhost", dbname= "lahman", user= "kriseberwein", password = "pissoff")

for (i in 1:length(dbTables)) { 
    if (dbExistsTable(con, dbTables[i])) {
        dbRemoveTable(dbTables[i])
    }
    dbWriteTable(con, name =  dbTables[i], value = get0(dbTables[i])) 
}

# Disconnect from database.
dbDisconnect(con)
rm(con, drv)
```