The comparison & result
One policy attempted to fold whenever another seat declared riichi. The current harness prefers a tile seen in any player’s discards, rather than checking safety against each declarer. That is a limited defensive heuristic, so this comparison should not be read as a test of expert defensive play. The normal policy retained its usual decision-making.
The result concerns those two policies and the opponents they faced. It supports weighing your hand’s progress and value against the risk, rather than assuming that either always folding or always pushing is correct.
At your table
- Consider distance from ready, wait quality, hand value and safe discards.
- The experiment does not identify the best discard in a particular human game.
Check your table’s scoring rules before applying the finding. Read the Japanese Riichi reference.
How it was tested
Published experiment: 100,000 simulated Riichi hands, reflex folding against riichi versus the normal policy, with alternating seat assignments.
The players were software policies, not human participants. This report describes the published experiment; it does not establish the best move for every possible hand or opponent.
Results without a published uncertainty interval should be read as point estimates. Similar results do not prove two choices are equivalent.
Read the Lab’s method and limitationsPublished by Hé Fēn Mahjong. Ask about this finding or suggest a correction.