In my intro to statistic class, my group and I look at some distinct groups in the Major Leagues. We collected data from all the position players who played 116 games or more in 2013 and stratified them into groups: 50 American League players, 50 National League Players, 33 24-27 year old players, 34 28-31 year old players, and 33 32-35 year old players.
Using statistical software and proper judgement we found that there is not much evidence that proves that one group is better than the other in home runs. This chart to the right shows the home runs for the 50 players in each league who played at least 116 games. The evidence shows that the American League is not better than the National League as many think, they are literally almost exactly the same. The same outcome came up when comparing the number of home runs by the different age groups. The American League is only superior in one respect: the pitcher is not required to bat, but I did not incorporate that, I strictly wanted to test the players.
Batting average also doesn't change much between age group as many think. It is a common misconception that the 28-31 age group is the best age group, but in batting average they all seem to be the same. In this graph 0 represents the other two age groups, and 1 represents the 28-31 age group. They are statistically similar. A player may be healthiest and most agile when in this age group, but in 2013 all of the players seem to have similar statistics.
Remember that each age group was composed of about 33 players randomly selected from the total number of players in that age group who played 116 games or more. One notable player that was not selected randomly was Miguel Caberea. Perhaps after another selection the statistics may be slightly more different, but not by much.
No comments:
Post a Comment