Monday, November 13, 2006
In the Post-Expansion Era, Does Good Hitting Beat Good Pitching?
TABLE BELOW
Well, I promised way back when that I was working on a study, comparing win/loss record versus game duration in Major League Baseball. I first had the idea a couple of years ago when the Indians went from being a perennial contender to a bottom feeder. Suddenly, 7 p.m. games were ending before 10 p.m., when just a few years previous, Indians games wouldn't conclude till 10:30 or later. With the help of my brother-in-law, we embarked on a mission to see what we might discover exploring game duration and any possible correlation to win/loss record. We decided to go as far back as 1996 because 1995 was a shortened season and 1994 was the lockout year. Using the last ten years gave us a pretty unique glance on how pitching and hitting have influenced win/loss records on post-expansion era baseball.
The process was rather painful. My brother-in-law downloaded the boxscore of every game, putting the information into a spreadsheet. I compiled a win/loss record for each team, noting overall record and yearly record. Having gathered data for over 16,000 games, I went back through, searching to fill in the blanks for games with either no information or skewed data. At times, this process was frustrating to say the least. (In the table below, we decided to normalize the games to 54 outs so that we could eliminate skewed results that variables like rain-shortened contests would have caused)
It was still necessary to find out the circumstances around these games so that we could have the option of including them in research down the road. Just to give you an idea of how tedious the process was...one day, I spent thirty-five minutes hunting for information regarding a June 1999 game between the San Diego Padres and Houston Astros. I finally stumbled upon the answer. In this game, Astros manager Larry Dierker had collapsed in the dugout, resulting in a suspended game that was finished at a later date. This was just one of several instances where bizarre circumstances had resulted in little or no information provided in the boxscore.
Below represent just some of our findings. In years 1998, 2001, 2003, and 2004, and over the ten year span collectively, we found that the Top Ten teams played 1% longer games than the middle ten. The bottom ten teams in the majors played 1% shorter games than the middle tier teams. An interesting result, yes but 1% is an insignificant percentage that can be explained away in a number of different ways. One example might be to argue that the top tier teams have been built to win with offense in recent years. And that may be true. People also might criticize the data because of the rule changes that have been implemented to try to speed up the game during the last decade. Still, our research reveals some fascinating bits of information that require further exploration.
During our study, we also discovered random points of interest. For example, in 2005, the Yankees and Red Sox averaged seven percent longer games than the average of all the top tier teams that year. Another fascinating tidbit: I was right about Indians games. In their last pennant-winning year, 1997, Cleveland averaged almost 3 1/2 hours a contest. During their 94 loss campaign of 2003, the Indians average just over three hours a game.
1996 - 2005
--------------------WINS---------AVG MINS/54 OUTS-----INDEX TO AVG--RUNS SCORED
TOP TEN----------9,410----------------179----------------------1.01------------8,365
MIDDLE TEN------8,078----------------176----------------------1.00-----------7,840
BOTTOM TEN-----6,469----------------174----------------------0.99-----------6,963
AVERAGE------------------------------176---------------------------------------7,723
Copyright, 2006
One situation that we failed to fully address in our initial research is the fact that the Yankees may play ten minute longer games than a team like the Devil Rays. But these two teams play against each other multiple times over the course of the season. So what kind of influence does this matchup have on each's game duration averages? I believe the next step to be looking at when top ten teams play each other and when bottom ten teams play each other. Analyzing game duration in these instances could reveal more significant data that might support my thesis. Other variables that we have taken into account in our study but not appearing in this summary include: average minutes per game extra innings or otherwise, runs scored by the visitor, runs scored by home team, runs allowed by home and visitor, and comparisons of the American League game versus National League game duraiton. If you have any feedback, don't hesitate to chime in with your thoughts.
Well, I promised way back when that I was working on a study, comparing win/loss record versus game duration in Major League Baseball. I first had the idea a couple of years ago when the Indians went from being a perennial contender to a bottom feeder. Suddenly, 7 p.m. games were ending before 10 p.m., when just a few years previous, Indians games wouldn't conclude till 10:30 or later. With the help of my brother-in-law, we embarked on a mission to see what we might discover exploring game duration and any possible correlation to win/loss record. We decided to go as far back as 1996 because 1995 was a shortened season and 1994 was the lockout year. Using the last ten years gave us a pretty unique glance on how pitching and hitting have influenced win/loss records on post-expansion era baseball.
The process was rather painful. My brother-in-law downloaded the boxscore of every game, putting the information into a spreadsheet. I compiled a win/loss record for each team, noting overall record and yearly record. Having gathered data for over 16,000 games, I went back through, searching to fill in the blanks for games with either no information or skewed data. At times, this process was frustrating to say the least. (In the table below, we decided to normalize the games to 54 outs so that we could eliminate skewed results that variables like rain-shortened contests would have caused)
It was still necessary to find out the circumstances around these games so that we could have the option of including them in research down the road. Just to give you an idea of how tedious the process was...one day, I spent thirty-five minutes hunting for information regarding a June 1999 game between the San Diego Padres and Houston Astros. I finally stumbled upon the answer. In this game, Astros manager Larry Dierker had collapsed in the dugout, resulting in a suspended game that was finished at a later date. This was just one of several instances where bizarre circumstances had resulted in little or no information provided in the boxscore.
Below represent just some of our findings. In years 1998, 2001, 2003, and 2004, and over the ten year span collectively, we found that the Top Ten teams played 1% longer games than the middle ten. The bottom ten teams in the majors played 1% shorter games than the middle tier teams. An interesting result, yes but 1% is an insignificant percentage that can be explained away in a number of different ways. One example might be to argue that the top tier teams have been built to win with offense in recent years. And that may be true. People also might criticize the data because of the rule changes that have been implemented to try to speed up the game during the last decade. Still, our research reveals some fascinating bits of information that require further exploration.
During our study, we also discovered random points of interest. For example, in 2005, the Yankees and Red Sox averaged seven percent longer games than the average of all the top tier teams that year. Another fascinating tidbit: I was right about Indians games. In their last pennant-winning year, 1997, Cleveland averaged almost 3 1/2 hours a contest. During their 94 loss campaign of 2003, the Indians average just over three hours a game.
1996 - 2005
--------------------WINS---------AVG MINS/54 OUTS-----INDEX TO AVG--RUNS SCORED
TOP TEN----------9,410----------------179----------------------1.01------------8,365
MIDDLE TEN------8,078----------------176----------------------1.00-----------7,840
BOTTOM TEN-----6,469----------------174----------------------0.99-----------6,963
AVERAGE------------------------------176---------------------------------------7,723
Copyright, 2006
One situation that we failed to fully address in our initial research is the fact that the Yankees may play ten minute longer games than a team like the Devil Rays. But these two teams play against each other multiple times over the course of the season. So what kind of influence does this matchup have on each's game duration averages? I believe the next step to be looking at when top ten teams play each other and when bottom ten teams play each other. Analyzing game duration in these instances could reveal more significant data that might support my thesis. Other variables that we have taken into account in our study but not appearing in this summary include: average minutes per game extra innings or otherwise, runs scored by the visitor, runs scored by home team, runs allowed by home and visitor, and comparisons of the American League game versus National League game duraiton. If you have any feedback, don't hesitate to chime in with your thoughts.
# posted by Nate @ 8:48 PM 
