Tennis Match Moneyline (full match) model

Tennis Match Moneyline Model — User Guide

What this tool does

• Prices the full match moneyline (A vs B) for Best-of-3 or Best-of-5.
• Uses Service Games Won% (SGW) and Return Games Won% (RGW) as core inputs, adjusted for surface and conditions.
• Runs a Monte Carlo simulation of full matches to output:
  • P(A wins match), P(B wins match)
  • Fair American odds for both sides
  • Optional “value” vs your sportsbook odds

Key inputs

• Preset: Loads ATP/WTA surface baselines for SGW/RGW and volatility (Hard, Clay, Grass, Indoor).
• Format: Best of 3 or Best of 5. Favorites widen in Bo5.
• Surface: Affects base hold rates and tiebreak tendency (faster → higher holds).
• Players A/B names: For labeling only.
• SGW%: Service Games Won% by surface (last 12–24 months preferred).
• RGW%: Return Games Won% by surface.
• Volatility proxy: Aces + double-faults per service game (higher tends to increase randomness).
• Conditions:
  • Altitude/Indoor boost: Faster conditions slightly increase holds.
  • Wind: Reduces holds; lowers tiebreak rates.
  • Injury/fitness nudge: Small shift toward the fitter player.
• Advanced (optional):
  • Second-serve edges: A’s 2nd-serve points won on serve vs B’s 2nd-serve return vs 2nd (and vice versa).
  • Conversion nudge: Slight bias on big points toward returners (+) or servers (−).
  • Fatigue/Travel: Negative favors A (fresher A), positive favors B (fresher B).
• Monte Carlo controls:
  • Volatility: Match-to-match variability; higher for big servers or unstable form.
  • Correlation: Shared swings (e.g., wind affects both); increases set-to-set co-movement.
  • Samples: Number of simulations (higher = smoother results).
• Book Odds (optional): Enter American (e.g., -135, +120) or decimal (e.g., 1.74, 2.20) to see value edges.

How to use it (quick workflow)

1. Choose a preset (ATP/WTA + surface).
2. Confirm or enter SGW% and RGW% for both players (surface-specific if possible).
3. Set format (Bo3/Bo5) and surface if different from the preset.
4. Adjust conditions: Altitude/Indoor and Wind; add Injury if relevant.
5. Optional: Add second-serve numbers, a conversion nudge, and fatigue if you have reliable info.
6. Click Calculate to get:
   • P(A wins), P(B wins)
   • Fair American odds for A and B
7. Enter book odds to see “value” badges showing your edge in probability points.

Interpreting outputs

• P(A wins match): No-vig probability that A wins, given inputs.
• Fair odds: Derived from those probabilities; compare to book odds.
• Value badges: Show the difference between your fair probability and the book’s implied (e.g., +3.1 pts ≈ 3.1% edge).

Best practices for inputs

• Use recent, surface-weighted SGW/RGW (12–24 months). If only overall is available, start from a preset and nudge ±1–2%.
• Keep Injury/Fatigue adjustments modest unless well-supported.
• Wind matters: moderate/strong wind can cut holds and narrow favorite edges.
• Bo5 inflates edges for favorites; if switching Bo3 → Bo5, re-check value.

Common scenarios and how to adjust

• Big server favorite on fast court:
  • Altitude/Indoor: Slightly faster
  • Wind: Calm
  • Expect higher P(A wins), especially in Bo5; volatility can be set a bit higher.
• Elite returner vs shaky second serve:
  • Use second-serve nudges to depress the opponent’s hold.
  • Consider a small “returners favored” conversion nudge.
• Grind on slow clay with wind:
  • Set Wind to Light/Moderate; no Altitude boost.
  • Lower tiebreak propensity; favorites’ edge may shrink.

Troubleshooting

• Outputs look too favorite-heavy:
  • Reduce Altitude/Indoor; add Wind.
  • Check that SGW% isn’t pulled from a fast-surface sample for a clay match.
• Outputs look too returner-heavy:
  • Ensure RGW% is realistic for the surface/time window.
  • Set Conversion to Neutral if you’re unsure.
• Performance:
  • Lower samples to 10–15k while tuning; raise to 30–80k to finalize.