diff --git a/src/coolrl_lost_cities/games/classic/__init__.py b/src/coolrl_lost_cities/games/classic/__init__.py
index 1efc0e3..32227dc 100644
--- a/src/coolrl_lost_cities/games/classic/__init__.py
+++ b/src/coolrl_lost_cities/games/classic/__init__.py
@@ -4,8 +4,6 @@ from .bots import (
     LostCitiesBot,
     available_bot_names,
     build_bot,
-    play_game,
-    run_series,
 )
 from .env import LostCitiesEnv
 from .evaluation import (
@@ -38,10 +36,8 @@ __all__ = [
     "classic_config",
     "evaluate_bot",
     "make_bot_factory",
-    "play_game",
     "play_game_for_evaluation",
     "play_match",
-    "run_series",
 ]
 
 
diff --git a/src/coolrl_lost_cities/games/classic/bots/__init__.py b/src/coolrl_lost_cities/games/classic/bots/__init__.py
index 8d4d901..c497f92 100644
--- a/src/coolrl_lost_cities/games/classic/bots/__init__.py
+++ b/src/coolrl_lost_cities/games/classic/bots/__init__.py
@@ -3,7 +3,6 @@ from __future__ import annotations
 from ..interfaces import BotInput, LostCitiesBot
 from .heuristic import SafeHeuristicBot
 from .passive import PassiveDiscardBot
-from .play import play_game, run_series
 from .random import RandomBot
 from .registry import DEFAULT_BOT, available_bot_names, build_bot
 
@@ -16,6 +15,4 @@ __all__ = [
     "SafeHeuristicBot",
     "available_bot_names",
     "build_bot",
-    "play_game",
-    "run_series",
 ]
diff --git a/src/coolrl_lost_cities/games/classic/bots/play.py b/src/coolrl_lost_cities/games/classic/bots/play.py
deleted file mode 100644
index 29dcda6..0000000
--- a/src/coolrl_lost_cities/games/classic/bots/play.py
+++ /dev/null
@@ -1,61 +0,0 @@
-from __future__ import annotations
-
-from dataclasses import replace
-
-from ..game import GameState, LostCitiesConfig
-from ..interfaces import LostCitiesBot
-
-try:
-    import numpy as np
-except ImportError as exc:  # pragma: no cover
-    raise RuntimeError("numpy is required for Lost Cities bots") from exc
-
-
-def play_game(
-    bot0: LostCitiesBot,
-    bot1: LostCitiesBot,
-    config: LostCitiesConfig,
-    *,
-    seed: int | None = None,
-    max_steps: int = 10_000,
-) -> GameState:
-    game_config = replace(config, seed=seed) if seed is not None else config
-    state = GameState.new_game(game_config)
-    bots = [bot0, bot1]
-    for _ in range(max_steps):
-        if state.terminal:
-            return state
-        action = bots[state.current_player].act(state)
-        state.apply_action(action)
-    raise RuntimeError(f"game exceeded max_steps={max_steps}")
-
-
-def run_series(
-    bot0: LostCitiesBot,
-    bot1: LostCitiesBot,
-    config: LostCitiesConfig,
-    *,
-    games: int = 100,
-    seed: int = 0,
-) -> dict:
-    diffs: list[int] = []
-    wins0 = 0
-    wins1 = 0
-    draws = 0
-    for index in range(games):
-        state = play_game(bot0, bot1, config, seed=seed + index)
-        diff = state.score_diff(0)
-        diffs.append(diff)
-        if diff > 0:
-            wins0 += 1
-        elif diff < 0:
-            wins1 += 1
-        else:
-            draws += 1
-    return {
-        "games": games,
-        "avg_diff": float(np.mean(diffs)) if diffs else 0.0,
-        "wins0": wins0,
-        "wins1": wins1,
-        "draws": draws,
-    }
diff --git a/tests/games/classic/test_bots.py b/tests/games/classic/test_bots.py
index 17634de..735f5b2 100644
--- a/tests/games/classic/test_bots.py
+++ b/tests/games/classic/test_bots.py
@@ -4,9 +4,9 @@ from coolrl_lost_cities.games.classic.bots import (
     LostCitiesBot,
     RandomBot,
     SafeHeuristicBot,
-    play_game,
 )
 from coolrl_lost_cities.games.classic.bots.heuristic import draw_from_discard_action
+from coolrl_lost_cities.games.classic.evaluation import play_game_for_evaluation
 
 
 def test_builtin_bots_implement_lost_cities_bot() -> None:
@@ -15,7 +15,7 @@ def test_builtin_bots_implement_lost_cities_bot() -> None:
 
 
 def test_safe_heuristic_mirror_match_finishes() -> None:
-    state = play_game(
+    state, result = play_game_for_evaluation(
         SafeHeuristicBot(),
         SafeHeuristicBot(),
         LostCitiesConfig(n_colors=3, n_ranks=5, n_handshakes=1, hand_size=5),
@@ -23,6 +23,7 @@ def test_safe_heuristic_mirror_match_finishes() -> None:
         max_steps=200,
     )
     assert state.terminal is True
+    assert result.timed_out is False
 
 
 def test_safe_heuristic_opponent_value_ignores_hidden_hand() -> None: