About KSplit
A lineup-driven MLB strikeout analytics platform built to answer a question traditional projections rarely address clearly: what does the entire strikeout distribution look like for a pitcher in a specific matchup?
Most projections reduce pitcher strikeouts to a single number. In reality, strikeouts behave as a distribution shaped by pitcher tendencies, lineup composition, and workload dynamics. KSplit was built to model that distribution directly.
The platform analyzes pitcher strikeout ability against left- and right-handed hitters, integrates hitter strikeout tendencies within the confirmed lineup, and evaluates how those interactions shape the probability of outcomes across the entire strikeout range. The result is a structured distribution that shows not just the most likely outcome, but how the right tail develops and where volatility enters the profile.
This framework allows users to move beyond simple over/under thinking and instead understand the risk structure of a strikeout prop. Some matchups produce stable distributions centered around the median. Others produce fragile distributions where the right tail becomes difficult to access. KSplit exists to identify those differences before the game starts.
KSplit began as a modeling project focused on a problem that became increasingly obvious in the modern betting and fantasy landscape. As sportsbooks and DFS platforms expanded player prop markets, strikeout lines became one of the most widely bet outcomes in baseball. Yet most public projections treated strikeouts as a point estimate rather than a probability distribution. That approach often hides the structure that actually determines whether a strikeout prop has value.
Instead of asking what a pitcher is projected for, the more useful question became:
Answering that question required a model designed specifically for strikeout dynamics rather than repurposing generalized projection systems. KSplit was developed to make those dynamics visible.
KSplit models pitcher strikeouts at the plate-appearance level and builds the distribution from the matchup outward.
Pitchers often perform differently against left- and right-handed hitters. These splits form the foundation of the strikeout probability engine.
Each projected batter contributes a strikeout probability shaped by their historical performance against that pitcher's handedness.
Strikeout opportunity depends heavily on how long a pitcher remains in the game. KSplit incorporates workload expectations to anchor the volume of plate appearances.
Instead of producing a single projection, the model generates a full probability distribution across possible strikeout outcomes, allowing mean, median, and right-tail probabilities to be evaluated simultaneously.
KSplit was created by an independent analyst focused on probability modeling and matchup-driven baseball analytics. The project grew out of an effort to better understand how pitcher strikeout props actually behave in real game environments.
After logging and analyzing hundreds of historical matchups, the modeling framework evolved into a structured system for evaluating strikeout distributions rather than point projections. What started as a personal modeling tool gradually developed into the platform that now powers KSplit.
KSplit is designed for bettors, analysts, and baseball fans who want a deeper understanding of pitcher strikeout dynamics. By focusing on probability distributions instead of single projections, the platform aims to provide a clearer view of risk, volatility, and upside in pitcher strikeout markets.
Baseball outcomes are inherently uncertain. The objective of KSplit is not to eliminate that uncertainty, but to structure it in a way that can be understood and analyzed before the first pitch is thrown.
Learn how the model works on the Methodology, walk through the columns on the How to Guide for Today's Dashboard, or look up any metric in the Glossary.
Ready to explore the live data? Create an account or sign in.