Data Overview
Row
Baseball Data
player homeruns years atbat hits runs rbi walks league86 team86 league87 team87
1 Reggie Jackson 548 20 9528 2510 1509 1659 1342 A Cal A Oak
2 Dave Kingman 442 16 6677 1575 901 1210 608 A Oak A Oak
3 Graig Nettles 384 20 8716 2172 1172 1267 1057 N SD N Atl
4 Tony Perez 379 23 9778 2732 1272 1652 925 N Cin N Cin
5 Jim Rice 351 13 7127 2163 1104 1289 564 A Bos A Bos
6 George Foster 348 18 7023 1925 986 1239 666 N NY N NYThis data set originally had 322 observations and 25 variables. After pre-processing 12 variables remain. “homeruns” indicates the player’s number of career homeruns. “years” indicates the player’s total number of years in the major leagues. “atbat” indicates a count of the player’s career at-bat’s. “hits” indicates the number of career hits for each player. Similarly, “runs”, “rbi”, and “walks” indicate the counts of the player’s career runs scored, runs batted in, and walks (respectively). Lastly, “league86” and “team86” describe the league/conference or team for which the athlete played in during the 1986 season while “league87” and “team87” describe the league/conference or team for which the athlete played in during the 1987 season.
Row
Ratio of American League in 1986
Approximately 54% of all players in the MLB in 1986 were members of the American League.
# A tibble: 1 x 1
n
<int>
1 175Ratio of National League in 1986
Approximately 46% of all players in the MLB in 1986 were members of the National League.
# A tibble: 1 x 1
n
<int>
1 147Ratio of American League in 1987
Approximately 55% of all players in the MLB in 1987 were members of the American League.
# A tibble: 1 x 1
n
<int>
1 176Ratio of National League in 1987
Approximately 45% of all players in the MLB in 1987 were members of the National League.
# A tibble: 1 x 1
n
<int>
1 146Team Consideration
Column
1986
1987
Row
Switching Teams between ’86 and ’87 Seasons
275 of the 322 players remained at their teams for both the ’86 and ’87 seasons while 47 players did not. Therefore, approximately 15% of players switched teams that year. This is representative of typical MLB patterns for players changing teams since players can be added to or removed from the 25 man roster depending on injury, performance level, contracts, etc. The data presents that these two seasons do not need to be considered abnormal in regards to roster changing.
Relationship between Team and Runs
This plot compares each team in the MLB on the basis of the particular players career run totals. Each dot represents a player. During the 1986 season, the two players with the highest record for career runs played for the Cincinatti Reds and California Angels. Yet, during the 1987 season, the California Angels no longer had one of the highest career run scoring players. Instead, the Oakland Athletics rose to the top. The Cincinatti Reds remained at the top as well.
Data Table
Model
In order to better predict a player’s career homerun totals, I build a linear regression mode with atbat, rbi, runs, hits and walks as predictors. All indicators were found to be significant with an adjusted r squared value of 0.97.
Call:
lm(formula = homeruns ~ atbat + rbi + runs + hits + walks, data = edit_baseball,
family = binomial)
Residuals:
Min 1Q Median 3Q Max
-62.524 -5.802 0.966 6.802 45.229
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.738351 1.241151 0.595 0.552
atbat 0.015221 0.003832 3.972 8.83e-05 ***
rbi 0.431491 0.007574 56.973 < 2e-16 ***
runs 0.284017 0.017403 16.320 < 2e-16 ***
hits -0.282260 0.014799 -19.073 < 2e-16 ***
walks -0.051369 0.009158 -5.609 4.44e-08 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Residual standard error: 13.9 on 316 degrees of freedom
Multiple R-squared: 0.9744, Adjusted R-squared: 0.974
F-statistic: 2409 on 5 and 316 DF, p-value: < 2.2e-16
