Statcast MCP Server

Instructions

Implementation Reference

• src/statcast_mcp/server.py:1190-1219 (handler)

  Implementation of the batter_percentile_ranks MCP tool, which fetches data from pybaseball and formats it.

  @mcp.tool()
  def batter_percentile_ranks(year: int, player_name: str | None = None) -> str:
      """Statcast percentile ranks for hitters vs the league (exit velo, barrel%, xwOBA, etc.).
  
      Each metric is 0–100 where higher is better for that stat. Use to answer
      "How elite is this hitter on Statcast?" or compare league-wide.
  
      Args:
          year: Season year (e.g. 2024).
          player_name: Optional. If set, returns only that player's row (e.g. "Aaron Judge").
  
      Qualifying batters: ~2.1 PA per team game (Statcast default).
      """
      from pybaseball import statcast_batter_percentile_ranks as _fn
  
      try:
          data = _fn(year)
      except Exception as e:
          return f"Error fetching batter percentile ranks: {e}"
  
      if player_name:
          try:
              data = _filter_player_rows(data, player_name)
          except ValueError as e:
              return str(e)
          if data.empty:
              return f"No percentile data found for {player_name} in {year}."
  
      return _fmt(data, max_rows=50)