I have visited The Negro Leagues Baseball Museum in Kansas City many times in my life. I would highly recommend it to anyone. Last year, the MLB finally incorporated NLB stats – which was outstanding!

Technical Explanation: Below is a k-means cluster of NLB players based on these offensive stats (data from Seamheads). Using a silhouette score, two clusters were deemed optimal. More or less, the clustering model drew a line to clearly call out the best players.

Non-Technical Explanation: The methodology isolates the best NLB players based on comprehensive offense statistics. The players in blue are the “top” players statistically; those in red would be categorized as “good, OK, or not that good”. There was not enough variation in the data to confidently create more than two clusters. The chart below visualizes the similarity of players based on their offensive performance. For example, a player comparatively far away from others would be considered unique in terms of their performance, such as Cool Papa Bell. Additionally, players close together are highly similar. For instance, the most similar player to Turkey Stearns is Oscar Charleston. Lastly, don’t pay much attention to the absolute location of players on the graph – what matters is the relative location of the points. (More or less, we take all the data you see when hovering over a player and compress it onto this graph; so the values on the x-axis and the y-axis are not directly interpretable).

Seeing how distinct some of these stars were is pretty neat! Hover over the points to player stats and appreciate the uniqueness of these great players.