diff --git a/evalharness/model/pool.py b/evalharness/model/pool.py
index 275e37c..5d0ddf4 100644
--- a/evalharness/model/pool.py
+++ b/evalharness/model/pool.py
@@ -320,9 +320,8 @@ class AdaptiveGate:
             lvl = max(1, int(self.limit))
 
             # not enough evidence yet at this level: keep measuring.
-            # need = max(MIN_OK, 2x level): rate noise shrinks only with
-            # samples proportional to the concurrency being judged
-            need_ok = max(self.MIN_OK, lvl * 2)
+            # evidence = 2 x level completions, no other floor
+            need_ok = max(2, lvl * 2)
             # dwell fallback scales with the OBSERVED completion cadence:
             # a 60s-per-request bench needs minutes, not 25s, before a
             # single-sample judgment is acceptable