The Question That Started It All
In 2023, Major League Baseball introduced new rules restricting where defensive players could position themselves on the field. This change effectively banned “The Shift” – a defensive strategy that had become increasingly popular since 2006. At the time, I suspected this ban would be nothing more than a band-aid solution with little real impact on the game. Now, with a full season of data available, we can finally test that hypothesis.
What Was “The Shift” Anyway?
The concept of shifting defensive players isn’t new – teams have been doing it sporadically throughout baseball history. However, the modern version of The Shift was pioneered by the Tampa Bay Rays in 2006 when manager Joe Maddon used advanced statistics (sabermetrics) to devise a strategy against Boston Red Sox slugger David “Big Papi” Ortiz.
The data showed that Ortiz, a left-handed power hitter, pulled nearly every ball to the right side of the field. So the Rays stacked extra defenders on that side, leaving the left side of the infield nearly empty. The strategy worked brilliantly – Ortiz’s batting average dropped from over .300 (2004-2006) to .265 by mid-2006 as other teams copied the approach.
The Shift became controversial because it was primarily used against baseball’s biggest stars and most exciting hitters, making the game feel less dynamic and offensive. Eventually, MLB decided enough was enough and implemented restrictions for the 2023 season.

My Approach to Testing the Impact
To determine whether banning The Shift actually changed anything, I needed to compare player performance before and after the ban. But this presented a challenge – what if other factors (like changes to the baseball itself or new pitching rules) also affected hitting during this period?
My solution was to analyze two groups of players:
Sluggers: The power hitters who were most likely to face The Shift (defined as players with above-average slugging percentage)
Non-sluggers: Regular hitters who rarely, if ever, faced The Shift
If banning The Shift was the primary driver of any performance changes, we should see significant improvements mainly among sluggers, with little change among non-sluggers.
The Data Collection Process
I used the Python library pybaseball to gather statistics from 2006 through 2024, focusing on key offensive metrics: at-bats, hits, doubles, triples, home runs, and walks. I divided this data into two eras:
- “Shift Era”: 2006-2022
- “Post-Shift Era”: 2023-2024
To ensure I was analyzing players who actually faced The Shift regularly, I set minimum at-bat thresholds (400, 600, and 800) and categorized sluggers by how far above average their performance was – from the top 50% down to the elite top 1%.
What the Numbers Revealed
The Initial Results Looked Promising
When I first compared the average performance between the two eras, the results seemed to support the idea that banning The Shift helped hitters. Across the board, offensive numbers were higher in the post-ban period. Problem solved, right?

Not so fast. Averages can be misleading, and I needed to determine whether these differences were statistically significant or just random variation.
The Statistical Reality Check
To get a more complete picture, I used the Kolmogorov-Smirnov test, which compares entire distributions rather than just averages. This test tells us whether two groups of data are fundamentally different or could reasonably come from the same underlying population.
Using a 95% confidence interval (the standard threshold for statistical significance), here’s what I found:
For Non-sluggers: Here’s where things got interesting. The non-sluggers showed statistically significant improvements across nearly all offensive categories – even more consistently than the sluggers did.

For Sluggers: Only the top 16% of power hitters showed statistically significant improvements in most categories (hits, triples, home runs, and walks – but not doubles). The other slugger groups showed mixed results.

The Surprising Conclusion
This finding was the key to understanding what really happened. Since The Shift was rarely used against non-sluggers, it couldn’t be responsible for their improved performance. Yet these “regular” hitters showed more consistent statistical improvements than the power hitters who were supposedly being helped by the ban.
The evidence points to a clear conclusion: something other than banning The Shift was responsible for the improved offensive performance across baseball.
Whether it’s changes to the baseball itself, new rules affecting pitchers, evolving hitting approaches, or other factors, the data suggests that banning The Shift had minimal impact on the game. In fact, sluggers (the players the rule was designed to help) showed less consistent improvement than the players who were never affected by The Shift in the first place.
Final Thoughts
My initial skepticism about the Shift ban appears to have been justified. While offensive numbers did improve after 2023, this improvement affected all types of hitters equally – not just the power hitters who faced The Shift. This pattern strongly suggests that other factors were driving the change.
Sometimes the most interesting finding is discovering that the obvious explanation isn’t the right one. In this case, banning one of baseball’s most visible and controversial strategies appears to have been largely symbolic rather than transformative.
Note: This analysis covers data through 2024. As more seasons pass, we’ll get an even clearer picture of The Shift ban’s true impact – or lack thereof.
To see categories of my sports analytic posts, pick one of the below:
Is this kind of article interesting to you? Let us know in the comments below.
If you want to be added to our newsletter (no worries, we won’t overwhelm you), we often create bonus material only for newsletter subscribers. Just sign up with your email address below.

