diff --git a/bsmain/management/commands/showstat.py b/bsmain/management/commands/showstat.py
index 26e04a5..ec8c219 100644
--- a/bsmain/management/commands/showstat.py
+++ b/bsmain/management/commands/showstat.py
@@ -1,8 +1,11 @@
+from datetime import datetime, timedelta
+
 from django.core.management.base import BaseCommand
 from django.utils import timezone
+from django.db.models import Sum, Max
 
 from bugsink.transaction import durable_atomic
-from snappea.models import Task
+from snappea.models import Task, Stat
 from events.models import Event
 
 
@@ -14,9 +17,15 @@ class Command(BaseCommand):
             choices=[
                 "snappea-queue-size",
                 "event_count",
+                "snappea-stats",
                 "digestion_speed",
             ],
         )
+        parser.add_argument(
+            "--task-name",
+            help="Task name to filter by (snappea-stats only)",
+            default=None,
+        )
 
     def handle(self, *args, **options):
         stat = options["stat"]
@@ -28,13 +37,18 @@ class Command(BaseCommand):
             with durable_atomic():
                 print(Event.objects.all().count())
 
+        if stat == "snappea-stats":
+            return self.snappea_stats(options["task_name"])
+
         if stat == "digestion_speed":
+            # NOTE: is this still a valuable stat? snappea_stat for "digest" task is more useful, I'd say. esp. given
+            # the warning
             print("WARNING: when eviction is enabled, the numbers will be wrong")
 
-            for window in [1, 10, 30, 60, 5 * 60, 60 * 60, 24 * 60 * 60]:
-                now = timezone.now()
-                with durable_atomic():
-                    qs = Event.objects.filter(digested_at__gte=now - timezone.timedelta(seconds=window))
+            now = timezone.now()
+            with durable_atomic():
+                for window in [1, 10, 30, 60, 5 * 60, 60 * 60, 24 * 60 * 60]:
+                    qs = Event.objects.filter(digested_at__gte=now - timedelta(seconds=window))
                     digested_in_window = qs.count()
                     # (safe assumption: digestion is done in order; this is "safe enough" (perhaps with a tiny rounding
                     # error) because of the snappea queue, which is a FIFO queue); the rounding error is in the
@@ -65,3 +79,54 @@ class Command(BaseCommand):
 
                     factor = digestion_window / ingestion_window
                     print(f"  ingestion window: {ingestion_window:.1f}s, factor: {factor:.1f}")
+
+    def snappea_stats(self, filter_task_name):
+        now_floor = datetime(*(datetime.now(timezone.utc).timetuple()[:5]), tzinfo=timezone.utc)
+
+        print("""past n minutes  task                                              AVG                  MAX
+                                                  done/s err/s   wall   wait  write   wall   wait  write   backlog""")
+
+        with durable_atomic(using="snappea"):
+            for window in [1, 2, 5, 10, 60, 5 * 60, 24 * 60]:
+                since = now_floor - timedelta(minutes=window)
+                seconds_in_window = 60 * window
+
+                base_qs = Stat.objects.filter(timestamp__gte=since, timestamp__lt=now_floor)
+                if filter_task_name:
+                    base_qs = base_qs.filter(task_name__endswith="." + filter_task_name)
+
+                stats = base_qs.values(
+                    "task_name",
+                ).annotate(
+                    done=Sum("done"),
+                    errors=Sum("errors"),
+                    wall_time=Sum("wall_time"),
+                    wait_time=Sum("wait_time"),
+                    write_time=Sum("write_time"),
+                    max_wall_time=Max("max_wall_time"),
+                    max_wait_time=Max("max_wait_time"),
+                    max_write_time=Max("max_write_time"),
+                    max_task_count=Max("task_count"),
+                )
+                for stat in stats:
+                    for field in ["wall_time", "wait_time", "write_time"]:
+                        stat[field] /= stat["done"]
+
+                    for field in ["done", "errors"]:
+                        stat[field] /= seconds_in_window
+
+                    task_name = stat["task_name"].split(".")[-1]
+
+                    # len("send_project_invite_email_new_user") == 34
+                    print(f"{window:<4}            "
+                          f"{task_name:<34} "
+                          f"{stat['done']:5.1f} "
+                          f"{stat['errors']:5.1f} "
+                          f"{stat['wall_time']:6.3f} "
+                          f"{stat['wait_time']:6.3f} "
+                          f"{stat['write_time']:6.3f} "
+                          f"{stat['max_wall_time']:6.3f} "
+                          f"{stat['max_wait_time']:6.3f} "
+                          f"{stat['max_write_time']:6.3f} " +
+                          (f"{stat['max_task_count']:9d}" if stat["max_task_count"] else "  v. many")
+                          )