fix(detector): suppress niche-only signals under a min trade size (#100)

The size anomaly detector currently emits a base 0.2 confidence signal
whenever a market is "niche" (low volume OR niche-prone category with
unknown volume) — even for $5 trades. In production this floods the
alert pipeline with nothing-burgers, because the niche-prone category
set covers `science / tech / finance / other`, which matches a huge
chunk of Polymarket's long tail.

This adds a `DEFAULT_NICHE_MIN_TRADE_SIZE = $500` floor that applies
ONLY to the niche-only path: if a trade exceeds the volume or book
thresholds, the guard does not block it. So real anomalies still come
through, but tiny niche trades get filtered out before reaching the
risk scorer.

The threshold is configurable via `niche_min_trade_size` in
`SizeAnomalyDetector.__init__`.

Tests added:
- niche-only below floor → suppressed
- niche-only at/above floor → still emits 0.2 base
- small trade with high volume_impact → guard does not block

Co-authored-by: schrodinger01 <schrodinger01@users.noreply.github.com>

src/polymarket_insider_tracker/detector/size_anomaly.py

@@ -17,6 +17,9 @@ logger = logging.getLogger(__name__)
 DEFAULT_VOLUME_THRESHOLD = 0.02  # 2% of daily volume
 DEFAULT_BOOK_THRESHOLD = 0.05  # 5% of order book depth
 DEFAULT_NICHE_VOLUME_THRESHOLD = Decimal("50000")  # $50k daily volume
+# Niche-only path requires real trade size; below this, suppress the base
+# 0.2 confidence so we don't flood alerts with low-value niche trades.
+DEFAULT_NICHE_MIN_TRADE_SIZE = Decimal("500")  # USDC notional
 
 # Niche market categories - markets in these categories with low specificity
 # are more likely to have insider information value
@@ -59,6 +62,7 @@ class SizeAnomalyDetector:
         volume_threshold: float = DEFAULT_VOLUME_THRESHOLD,
         book_threshold: float = DEFAULT_BOOK_THRESHOLD,
         niche_volume_threshold: Decimal = DEFAULT_NICHE_VOLUME_THRESHOLD,
+        niche_min_trade_size: Decimal = DEFAULT_NICHE_MIN_TRADE_SIZE,
     ) -> None:
         """Initialize the size anomaly detector.
 
@@ -67,11 +71,15 @@ class SizeAnomalyDetector:
             volume_threshold: Threshold for volume impact (default 0.02 = 2%).
             book_threshold: Threshold for book impact (default 0.05 = 5%).
             niche_volume_threshold: Volume below which market is niche ($50k).
+            niche_min_trade_size: Minimum trade notional ($) to allow a
+                niche-only signal. Below this, niche-only path is suppressed
+                to avoid flooding alerts with low-value trades.
         """
         self._metadata_sync = metadata_sync
         self._volume_threshold = volume_threshold
         self._book_threshold = book_threshold
         self._niche_volume_threshold = niche_volume_threshold
+        self._niche_min_trade_size = niche_min_trade_size
 
     async def analyze(
         self,
@@ -129,6 +137,18 @@ class SizeAnomalyDetector:
         exceeds_volume = volume_impact > self._volume_threshold
         exceeds_book = book_impact > self._book_threshold
 
+        # Niche-only signals require a minimum trade size; otherwise we'd
+        # flood alerts with every tiny trade in any niche-prone category.
+        niche_only = is_niche and not exceeds_volume and not exceeds_book
+        if niche_only and trade_size < self._niche_min_trade_size:
+            logger.debug(
+                "Trade %s niche-only but size %s < min %s, skipping",
+                trade.trade_id,
+                trade_size,
+                self._niche_min_trade_size,
+            )
+            return None
+
         if not exceeds_volume and not exceeds_book and not is_niche:
             logger.debug(
                 "Trade %s does not exceed thresholds: volume=%.4f, book=%.4f",

tests/detector/test_size_anomaly.py

@@ -588,6 +588,93 @@ class TestAnalyzeMethod:
         assert signal.is_niche_market is True
         assert signal.confidence == 0.2  # niche_base
 
+    @pytest.mark.asyncio
+    async def test_analyze_niche_only_below_min_trade_size_skipped(
+        self,
+        mock_metadata_sync: AsyncMock,
+        sample_metadata: MarketMetadata,
+    ) -> None:
+        """Niche-only trades below the min trade size are suppressed."""
+        mock_metadata_sync.get_market.return_value = sample_metadata
+        detector = SizeAnomalyDetector(mock_metadata_sync)
+
+        tiny_trade = TradeEvent(
+            market_id="market_abc123",
+            trade_id="tx_tiny",
+            wallet_address="0xabc",
+            side="BUY",
+            outcome="Yes",
+            outcome_index=0,
+            price=Decimal("0.5"),
+            size=Decimal("100"),  # $50 notional, below default $500 floor
+            timestamp=datetime.now(UTC),
+            asset_id="token_123",
+            event_title="Niche tiny trade",
+        )
+
+        signal = await detector.analyze(tiny_trade)
+        assert signal is None
+
+    @pytest.mark.asyncio
+    async def test_analyze_niche_only_at_min_trade_size_emits(
+        self,
+        mock_metadata_sync: AsyncMock,
+        sample_metadata: MarketMetadata,
+    ) -> None:
+        """Niche-only trades at or above the min trade size still emit."""
+        mock_metadata_sync.get_market.return_value = sample_metadata
+        detector = SizeAnomalyDetector(mock_metadata_sync)
+
+        ok_trade = TradeEvent(
+            market_id="market_abc123",
+            trade_id="tx_ok",
+            wallet_address="0xabc",
+            side="BUY",
+            outcome="Yes",
+            outcome_index=0,
+            price=Decimal("0.5"),
+            size=Decimal("2000"),  # $1000 notional, above default $500 floor
+            timestamp=datetime.now(UTC),
+            asset_id="token_123",
+            event_title="Niche ok trade",
+        )
+
+        signal = await detector.analyze(ok_trade)
+        assert signal is not None
+        assert signal.is_niche_market is True
+        assert signal.confidence == 0.2
+
+    @pytest.mark.asyncio
+    async def test_niche_min_trade_size_does_not_block_real_anomalies(
+        self,
+        mock_metadata_sync: AsyncMock,
+        sample_metadata: MarketMetadata,
+    ) -> None:
+        """A trade that exceeds volume/book thresholds is never blocked by the niche guard."""
+        mock_metadata_sync.get_market.return_value = sample_metadata
+        detector = SizeAnomalyDetector(mock_metadata_sync)
+
+        small_but_high_impact_trade = TradeEvent(
+            market_id="market_abc123",
+            trade_id="tx_small_impact",
+            wallet_address="0xabc",
+            side="BUY",
+            outcome="Yes",
+            outcome_index=0,
+            price=Decimal("0.5"),
+            size=Decimal("200"),  # $100 notional, below niche floor
+            timestamp=datetime.now(UTC),
+            asset_id="token_123",
+            event_title="Small but high-impact",
+        )
+
+        # Provide small daily_volume so volume_impact exceeds 2% threshold
+        signal = await detector.analyze(
+            small_but_high_impact_trade, daily_volume=Decimal("1000")
+        )
+        assert signal is not None
+        assert signal.volume_impact > 0.02
+
     @pytest.mark.asyncio
     async def test_analyze_no_anomaly(
         self,