WNCAAB Desk Strategy Spec

Extracted from Sports Desk Merged Spec (2026-03-15)

Full merged reference: spec-final.md

OVERVIEW

The WNCAAB (Women's College Basketball) desk detects mispricings between Kalshi alt line prices and Pinnacle sharp book fair value for women's college basketball games. Same core thesis as all desks: Pinnacle is the anchor, prediction markets lag.

WNCAAB has less data coverage, wider lines, and thinner liquidity compared to NCAAB. This creates both opportunity (larger mispricings when they exist) and risk (fewer data sources to validate edges, harder to execute).

CORE EDGE DETECTION

How It Works

Get Pinnacle's odds for a WNCAAB game (moneyline, spread, total)
Remove the vig to get the "fair" probability
Compare that fair probability to Kalshi's price for the same market
If Kalshi's price is off by more than 4 cents after fees, that's a tradeable edge

De-Vig Method

2-Way Markets (Spreads, Totals, Moneylines) — Multiplicative method:

total_implied = implied_home + implied_away
novig_home = implied_home / total_implied
novig_away = implied_away / total_implied

Kalshi Fee Impact

Kalshi charges 7% fee on profit.

Minimum net edge: 4 cents after fees. Consider raising to 5 cents for WNCAAB given thinner data — model confidence is lower.

Sweet spot: implied probabilities between 35% and 65%.

Position Sizing (Kelly Criterion)

Use 1/4 Kelly for WNCAAB bets:

bet_size = min(bankroll * full_kelly * 0.25, bankroll * 0.05)

Hard Caps:

Maximum single bet: 5% of bankroll
Maximum per game: 10% of bankroll
Maximum daily WNCAAB exposure: 20% of bankroll (lower than men's due to less data confidence)

WNCAAB-SPECIFIC CONSIDERATIONS

Data Scarcity

WNCAAB has significantly less public analytics coverage than NCAAB:

No KenPom equivalent with the same depth
BartTorvik covers WNCAAB but with less granularity
Fewer rating models available
Less historical data for backtesting

Implication: Model confidence is inherently lower. Compensate by:

Requiring larger minimum edges (consider 5 cents vs 4 for men's)
Reducing position sizes
Being more conservative on deep tail bets

Efficiency Ratings

Source	What It Provides	Priority
HerHoopStats	Women's college basketball analytics, efficiency ratings	PRIMARY
BartTorvik	WNCAAB team ratings (less detailed than men's)	PRIMARY
ESPN	Power rankings, BPI where available	SECONDARY

Talent Parity

WNCAAB has even larger talent gaps than NCAAB:

Dominant programs (UConn, South Carolina, etc.) create 30-40 point spreads
Model is most reliable for competitive matchups
For spreads > 20 points, do not bet — model accuracy drops significantly

Market Characteristics

Thinner liquidity: Fewer traders, wider bid/ask spreads
Larger mispricings: When edges exist, they're often bigger than NCAAB
Slower market adjustment: Kalshi prices update more slowly for WNCAAB
Less Pinnacle coverage: Pinnacle may not offer lines for all WNCAAB games — only bet when Pinnacle anchor exists

Tournament Play (March Madness)

Women's tournament has expanded to 68 teams (same as men's)
Public interest has grown significantly — Kalshi volume increasing
Same tournament considerations as NCAAB: neutral sites, motivation asymmetry, single-elimination variance

WNCAAB DATA SOURCES

Source	What It Provides	Frequency	Priority
Pinnacle (via Odds API)	Sharp odds: ML, spread, total	When available	Critical
Kalshi	Market prices	Real-time API	Critical
HerHoopStats	Efficiency ratings, team stats	Daily	High
BartTorvik	WNCAAB team rankings	Daily	High
ESPN API	Scores, game state, injury reports	Continuous	High

Rating Models

Model	Source	Type
HerHoopStats	herhoopstats.com	Efficiency ratings
BartTorvik WNCAAB	barttorvik.com	Team rankings
ESPN rankings	espn.com	Power rankings (when available)

Average available models. Exclude NULL silently.

WHEN NOT TO BET (Hard Rules)

Condition	Reason
No Pinnacle line available	No sharp anchor = no reliable fair value
Spread > 20 points	Model unreliable for extreme mismatches
Deep tail lines (5+ rungs)	Model unreliable
Kalshi market volume < $500	Illiquid
Net edge < 4 cents (consider 5 cents)	Below threshold
Implied probability < 15% or > 85%	Extreme tails

COLLECTION SCHEDULE

[11:00 PM ET prior night]
  - Fetch tomorrow's WNCAAB schedule from ESPN API
  - Check if Pinnacle has lines (only proceed if anchor exists)

[9:00 AM ET — Early Snapshot]
  - Odds API: Pinnacle WNCAAB lines (if available)
  - Kalshi: Pre-game alt line snapshot
  - Calculate no-vig probabilities, run edge scan

[11:00 AM ET — Model Collection]
  - HerHoopStats
  - BartTorvik WNCAAB

[2:00 PM ET — Mid Snapshot]
  - Odds API: Pinnacle update
  - Kalshi: Mid-day snapshot

[30 min before game]
  - ESPN injury check
  - Pre-close polling

[10 min before game — Closing Snapshot]
  - Final closing lines + CLV baseline

[After game → Grade edges, update Brier scores]

SETTLEMENT AND LEARNING

Same framework as NCAAB:

Track every edge with full metadata
Brier scores per market type
Monthly review with performance by spread size bucket

Additional WNCAAB tracking:

Compare model accuracy vs NCAAB (is the thinner data actually hurting?)
Track which games had Pinnacle lines vs which didn't
Monitor if mispricings are systematically larger than NCAAB (expected due to thinner markets)

RISK MANAGEMENT

Control	Threshold	Action
Max loss per trade	2% of capital	Hard cap
Max exposure per day (WNCAAB)	15% of capital	Hard cap
Daily drawdown stop	-5% of capital	Halt all trading
Max per single game	5% of capital	Hard cap

QUICK REFERENCE: KEY CONSTANTS

Parameter	Value
Min net edge	4-5 cents (higher threshold recommended)
Kalshi fee rate	7% on profit
Kelly fraction	0.25 (quarter)
Max single bet	5% bankroll
Max per game	10% bankroll
Max daily exposure	20% bankroll
Max spread to bet	20 points
Min bet size	0.5% bankroll
Implied prob range	15%-85%
Min Kalshi volume	$500

TODO: Upgrade Kalshi REST to WebSocket Feed

Status: NOT BUILT — add when this desk goes live for execution

Current state: All Kalshi data (prices, order books, trades) is fetched via REST API polling on cron schedules. This is fine for edge detection and monitoring, but NOT sufficient for live trade execution.

Why WebSocket matters:

REST polling introduces lag — by the time you see a price, it may already be gone
On tight edges (2-5 cents), even 1-2 seconds of lag can eat the entire edge
WebSocket gives real-time order book deltas and trade notifications
Essential for hitting specific fill prices and monitoring execution quality

What to build when ready:

Connect to Kalshi WebSocket feed (wss://trading-api.kalshi.com/trade-api/ws/v2)
Subscribe to order book channels for active tickers
Stream trade events for real-time freshness (replaces REST trades polling)
Feed live prices into execution engine for precise entry/exit
Update Fill Quality Monitoring to measure slippage vs quoted price

Note: Sports markets use RFQ (Request for Quote) so the visible order book is usually empty — but the trade feed still matters for freshness signals and the WebSocket is required for order submission. Weather markets DO have real visible order books where this upgrade is even more critical.

Added 2026-03-29 — upgrade REST to WS when desk moves to live execution

Source: ~/edgeclaw/results/spec-panel/sports-desk/wncaab-strategy.md