top of page
Writer's pictureBruin Sports Analytics

Is Stealing Bases an Effective Strategy for Teams to Win?

By: Nathan Kim and Aiden Lee


Source: Yahoo Sports

A sensational young player Elly De La Cruz from Cincinnati Reds is dominating the MLB with his baserunning and not showing signs of a sophomore slump. He has 31 stolen base bases in his 55 game appearances (5/28/2024). His stolen base stat is currently ranked in 1st place in MLB. This outstanding stolen base pace can reach approximately 100 stolen bases for the 2024 season. His performance can recall the memory of Rickey Henderson who is the all-time stolen base leader in MLB (1406). Moreover, a similar story happened in the 2023 MLB season. Esteury Ruiz, who was a below-average batter (OPS last season = 0.654), stole 67 bases, which is a significant increase in stolen base stats compared to recent MLB seasons. Then, what were the changes that impacted MLB sprinters?


One of the reasons for the increase in stolen bases and their success rate is the introduction of new rules and modifications in rules. First, the pitch clock, which is the new rule, started to be part of MLB baseball. Since the season of 2023, pitchers can have 15 seconds, at maximum, in between the pitches when there isn’t any runner and 20 seconds if there is at least one runner on base. Due to the pitch clock and limited time between pitches, pitchers are not able to care for the runners as usual. Moreover, the pitchers have a limited number of pickoffs. More specifically, pitchers can attempt pickoff or step-off from the mound for a maximum of twice per batter. Therefore, pitchers are more likely to struggle to keep the runner close to the base. Runners have another advantage due to limited pickoffs. If pitchers make two pickoffs or step-offs, runners can run and keep a large distance between them and the base with fewer concerns. Additionally, the base size increased as well. The distance between each base since the 2023 season has been 4.5 inches shorter than the 2022 season. Obviously, runners can reach the next base in a shorter time.


As people can clearly observe significant changes in a number of stolen bases from the 2022 to 2023 season, players attempt to steal bases more with higher success rates. It is obvious that MLB teams, media, and fans are now more interested in stealing bases. However, stolen base statistics is still a controversial topic in baseball sabermetrics. There are many articles and analyses that exhibit a strong or moderate correlation with a team’s winning percentages and popular statistics such as home runs, ERA, and total runs. Nevertheless, we are wondering whether the stealing bases positively influence teams to win or have no impact. Therefore, the research question we have decided to research is “Does stealing bases help teams to win?”

Source: mlb.com

Data Collection


For the data analysis, we referenced MLB.com and Statcast. We collected data from the 2022 and 2023 seasons to make a direct comparison. Moreover, we gathered data from each team to easily link with the winning percentage. The categories that we checked are the difference in number of stolen bases, the difference in stolen base success rate, the difference in caught stealing rate, and the difference in winning percentage. Then, we measured the correlation coefficient between the two variables, one variable regarding stolen base against winning percentage. When the correlation coefficient is near zero, then we can not claim an association between the two variables. In contrast, as the correlation coefficient gets closer to positive 1 or negative 1, there is a stronger correlation between the two variables. We referenced linear, quadratic, and exponential correlation coefficients. In other words, we measured linear, quadratic, and exponential correlations between the two variables. 


Here are the key definitions of the tracking data:

  • OPS (On-base Plus Slugging): It is a statistic that combines the on-base percentage and the slugging percentage which is commonly used to evaluate a batter’s performance. It can show a player’s power, batting eye, and contact skills simultaneously.

  • Stolen base success rate: It is a success rate calculated by the number of stolen bases divided by stolen base attempts. 


Data Analysis


Between the 2022 season and 2023 seasons, there were significant differences in trends regarding the stolen base success rate and a number of stolen bases. In 2023, there were 3503 stolen bases, in total. This is a drastic increase from the 2022 season with 2486 stolen bases. In terms of percentage, the number of stolen bases increased by 40 percent. Additionally, the average success rate increased from 74.9 percent to 79.9 percent. This statistic signifies that new adjustments of MLB including new rules helped the runners. 

By checking the correlation coefficient, we attempted to determine whether there is a meaningful association between two variables, one variable regarding the stolen base against the winning percentage. In general, we could notice that there is insufficient evidence to claim that there is a correlation between the difference in a number of stolen bases, stolen base success rate, and caught stealing percentage against winning percentage. We checked linear, quadratic, and exponential correlation. Yet, in most cases, we could only obtain a weak correlation or correlation coefficient close to zero. In one case we could see a weak association was a linear correlation between the difference in stolen base success rate and the difference in winning percentage. The correlation coefficient r is 0.3809634, which indicates a weak correlation, but not necessarily a strong correlation. Another case is the exponential link between the difference in stolen base success rate and the difference in winning percentage, with a correlation coefficient of 0.38092. Other than these two cases, most of the correlation coefficient values are nearly zero or insignificant. Therefore, excluding the stolen base success rate, it is hard to claim that stealing bases or catching stolen bases attempts help the team to increase the winning percentage, which is the ultimate goal of the team.

We collected the data for changes in the caught stealing rate from 2022 to 2023. Then, we drew a regression plot to find a relationship between changes in the caught stealing rate and changes in the winning percentage of teams. The scatterplot did not show a correlation between these two variables. Thus, we can conclude that the increase in the caught stealing rate from 2022 to 2023 did not have an obvious impact on the winning percentage of the teams.

We gathered the data for the difference in the number of stolen bases from 2022 to 2023. Then, we drew a regression plot to find an association between changes in the number of stolen bases and changes in the winning percentage of teams. The scatterplot did not show a correlation between these two variables. Thus, we can notice that the increase in the number of stolen bases from 2022 to 2023 did not have a meaningful impact on the team’s winning percentage. 

We collected the data for the difference in the stolen base success rate from 2022 to 2023 as well. Then, we draw a regression plot to find a relationship between changes in the stolen base success rate and changes in the winning percentage of teams. The linear correlation coefficient of this plot is 0.3809634. The scatterplot showed a weak positive linear relationship between these two variables. We can claim that the increase in the stolen base success rate from 2022 to 2023 impacted the winning percentage positively to some extent. However, the correlation is not significant enough to explain the strong relationship between the difference in stolen base success rate and the increase in winning percentage.    

We accumulated the data for the changes in the stolen base success rate between the 2022 and 2023 seasons. However, for this plot, we selected the data only for teams that had above-average differences in stolen base success rates throughout this period. This plot shows the relationship between changes in the stolen base success rate and changes in the winning percentage only for teams who improved their stolen base success rate more than the league average. The scatterplot did not represent a correlation between these two variables. Thus, we can conclude that the increase in the stolen base success rate from 2022 to 2023 did not have obvious effects on the winning percentage of teams. 

We collected the data for the number of stolen base changes from 2022 to 2023. Yet, we selected the data only for teams that have above-average differences in stolen bases this time. The plot shows the link between the difference in the number of stolen bases and the difference in the winning percentage only for teams who improved their stolen base numbers more than the league average. The scatterplot did not show the correlation between these two variables. Therefore, we can claim that the increase in the stolen bases from 2022 to 2023 did not have an apparent impact on the team’s winning percentage. 


As we gather up all of the information on the correlation coefficient from scatterplots, it is plausible to claim that there is an inference between stealing bases and the team’s performance. The difference in stolen base success rates and the winning percentage difference from 2022 to 2023 presented some degree of correlation out of the five scatterplots. Even though the correlation has a small value between the two variables (r = 0.3809634), it is still a meaningful outcome as other correlation values are near zero. It is not convincing to conclude that a higher stolen base success rate induced by a change in MLB rules directly supports teams to win, but it tends to indirectly increase the winning percentage by at least a small amount. If we compare scatterplots between the stolen base success rate of 30 teams and the stolen base success rate of above-average success rate teams (17 teams), we can conclude a different trend among above-average teams and below-average teams. 17 above-average teams did not show a correlation between success rate and winning percentage.  This indicates that as teams get better in stolen base rate above average level, there are no benefits to the team’s performance. However, as teams have below-average stolen base success rates, their performance will be negatively impacted because the scatterplot of all MLB teams showed a weak positive correlation. Thus, the stolen base success rate can contribute to increasing the winning percentage until the average league success rate. In conclusion, we can notice that stealing bases, in general, does not contribute to the victory of the teams. We can only claim that the stolen base success rate helps the win to some limited extent. 


Potential Causes


There are several possible causes of our outcome. One of them is the potential risk of injuries. Stealing bases can easily raise the chance of injuries in wrists, fingers, ankles, shoulders, hamstring, and many other body parts. In many cases, especially if the batter is the core of the lineup, the team will face great harm if the runner is injured. Therefore, though success in stealing bases can increase chances of scoring, in the long term perspective, there may be a greater loss for the team due to the high chance of injuries caused by stealing bases. Players using more energy to steal bases as runners may reduce their energy levels at bat. Despite the difference, which may be merely notable, batting is an activity that requires substantial energy in an instant. Therefore, focusing on baserunning and stealing bases may negatively influence the batting performance. 

Source: ftw.usatoday.com

The trend of the league is against stealing bases as well. Since the late 2010s, the importance of home runs has been emphasized. The total number of home runs in terms of league used to be managed at around 5000 home runs per season. Yet, recently, the total number of home runs increased to 5500 ~ 6000 home runs per season, in general. 


Therefore, if the team focuses on stolen bases, we can assume that the team has a low emphasis on extra-base hits, including home runs, as in many cases, power hitters are not quick in baserunning. In brief, despite the change in rules with increased base sizes, stealing bases may still be out of the trend in Major League Baseball, which has a strong focus on home runs and doubles. 


Conclusion


MLB, where powerful home runs, untouchable fastballs, and dramatic strikeouts dominate the league trends, has confronted a recent and drastic increase in stolen bases. The positive side of this speed-focused baseball play is that players become more active and the atmosphere of the baseball stadium gets more energetic. 

However, the increase in the number of stolen bases and its success rate were not effective enough to result in a dramatic increase in the winning percentage of MLB teams from 2022 to 2023. Through correlation coefficients and plots along with lines of best fit, we can notice that it is hard to establish an association between statistics regarding stolen bases against the difference in winning percentage. We can also check that the difference in data about stolen bases did not support teams to win.


Throughout the research, we also found some potential questions. Later, making distinctions between stolen bases to second base with stolen bases to third base may be meaningful research. Potentially, the importance of stealing bases in the MLB Postseason can be a hot research analysis question as well. Additionally, by witnessing more seasons and further trends, we may be able to draw more meaningful conclusions. 


Sources


244 views

Коментарі


bottom of page