There are a lot of ways to answer the question in the title. I settled on three here:

- What's the most common score for a team?
- What's the most common score for two teams in a game?
- What's the most common margin of victory for a game?

For the data set, I used all MLB games from 2009 - 2018 (taken from https://www.baseball-reference.com/leagues/MLB/).

Most common score for a team is found by collecting each team's score from each game. It is not normally distributed as you can see below:

The stats are:

- Mode (most frequent) - 3
- Median (middle value) - 4
- Average (one you think of...sum divided by count) - 4.4

Most common score for two teams in a game is found by collecting both teams' scores from each game and treating a pair of scores as a single value. The distribution of those is below:

Nothing really interesting jumped out here. 1-score games are most common, but that isn't shocking. Since 2, 3, and 4 were the most common scores above, it's no shock that 3-2 and 4-3 were the most common outcomes.

Nothing really interesting jumped out here. 1-score games are most common, but that isn't shocking. Since 2, 3, and 4 were the most common scores above, it's no shock that 3-2 and 4-3 were the most common outcomes.

Most common margin of victory is found by calculating the difference between the two scores in each game. For this analysis, I used 'home - away', so positive means home team won and negative means away team won. Here's the distribution:

This is the most interesting plot to me. The massive spike at +1 indicates that the home team winning by 1 is by far the most common outcome. Why is that? I don't know for certain, but there are two reasonable explanations that come to mind:

- No matter what happens outside of a home run, if the home team scores in the 9th inning of a game, they can only win by 1. For example, if the bases are loaded and the batter hits a triple, 3 runs should score but only 1 does by rule. Thus, many games that would be home team wins of more than +1 end up as +1. (credit to Ben Ethier for pointing this out)
- The home team can theoretically be more strategic in the bottom of the 9th. I don't know enough about baseball to know how this plays out, but the away team doesn't know how many runs they need to score. Say the game is tied. Away team has a runner on third in the top of the 9th with 1 out. They could sacrifice bunt to score, but is 1 run enough? The home team doesn't have to question this. If tied, sacrifice bunt and win. If down by 12, don't.

Finally...there appears to be a slight home-field advantage in baseball. You can see it pretty clearly in the margin of victory plot. For stats...while the median for both home and away is 4 runs and the mode is 3 runs, the average for home teams is 4.44 runs vs 4.28 runs for the away team. This resulted in the home team winning ~54% of the time.

And that's it. There's probably a lot of cool stuff to do with this type of data so I'll likely do more baseball posts in the future.

## 0 comments:

## Post a Comment