Tennis Totals & Handicaps Analysis

N. Mejia vs N. Basilashvili

Match & Event Information

Executive Summary

Key Factors:

• Model projects 22.8 total games (95% CI: 19-26) vs market line of 18.5
• P(Over 18.5) = 82% vs no-vig market 52.5% → 29.5pp edge on Over
• Both players have weak holds (~75%) → high break frequency → 52% chance of three sets
• Mejia projected -1.2 game margin, but market line is -5.5 (Basilashvili favored) → model disagrees with market direction
• Pass on spread due to conflict with market and insufficient edge alignment

Critical Insight: The market totals line at 18.5 is significantly low compared to the model’s 22.8 expectation. Both players’ modest hold percentages (75%) and high break rates create a game-rich environment. The 52% three-set probability further supports higher totals.

Quality & Form Comparison

Summary: Despite identical Elo ratings (1200), Mejia shows a stronger recent track record (44-31 vs 33-31) and significantly higher dominance ratio (1.42 vs 1.13), suggesting he’s winning games more comfortably than Basilashvili. Both players are stable in form with nearly identical three-set frequencies (~37%). The Elo parity indicates a competitive match, but Mejia’s superior recent performance and game dominance give him a slight quality edge.

Totals Impact: Both players average similar three-set frequencies (37%), supporting a balanced match structure projection. Basilashvili’s higher average total games (24.4 vs 23.0) suggests he plays longer matches, potentially pushing this match toward the higher end of expectations.

Spread Impact: Mejia’s dominance ratio advantage (+0.29) and superior recent record suggest a modest game margin edge. However, the Elo parity limits spread confidence—this appears to be a close matchup on paper despite Mejia’s better recent form.

Hold & Break Comparison

Summary: The hold percentages are nearly identical (75.2% vs 74.6%), indicating both players struggle somewhat to dominate on serve—typical of players at this ranking level. The key differential is on the return: Mejia breaks 29.1% of the time versus Basilashvili’s 25.7%, a meaningful 3.4pp edge. This translates to Mejia averaging 4.12 breaks per match compared to Basilashvili’s 3.82. Both players have modest tiebreak sample sizes (11 and 13 TBs respectively), but Mejia’s 63.6% TB win rate substantially exceeds Basilashvili’s 46.2%.

Totals Impact: With both players holding around 75%, we expect frequent breaks (8-9 per match combined) and fewer tiebreaks. The relatively low hold rates push toward shorter set scores (6-3, 6-4 range) rather than extended 7-5 or 7-6 sets. Basilashvili’s historical average (24.4 games) provides the upper anchor.

Spread Impact: Mejia’s 3.4pp break rate advantage is the primary spread driver. Converting 4.12 breaks per match versus 3.82 creates a modest game margin edge. Combined with his 2.1pp game win percentage advantage, Mejia should win slightly more games, but the narrow hold differential (0.6pp) limits the expected margin.

Pressure Performance

Break Points & Tiebreaks

Set Closure Patterns

Summary: The clutch statistics reveal an interesting paradox. Basilashvili converts break points at an elite 64.2% (well above the 40% tour average), but saves them at a below-average 57.7%. Mejia is the inverse: solid BP saved (63.5%) but converts at a strong 56.2%. This creates a high-break-rate environment when opportunities arise. In tiebreaks, Mejia dominates on serve (63.6% vs 46.2%) while Basilashvili excels on return (53.8% vs 36.4%). The set closure patterns are nearly identical except for serving for match, where Basilashvili’s 87.0% far exceeds Mejia’s 78.6%.

Totals Impact: The moderate consolidation rates (76%) and breakback rates (24-25%) suggest volatile sets with breaks being traded. Neither player locks down sets after gaining breaks, which pushes toward more games per set. However, both are reasonably efficient serving for sets (83-85%), preventing excessive 7-5 scorelines.

Tiebreak Probability: With both players holding only ~75%, tiebreaks are relatively unlikely (estimated 12-15% per set). When they occur, the matchup is fascinating—Mejia holds serve better in TBs but Basilashvili returns better. This volatility slightly widens the total games confidence interval.

Game Distribution Analysis

Set Score Probabilities

Match Structure

Total Games Distribution

Analysis: The game distribution reflects a competitive match with narrow hold advantages. The most likely set scores are 6-4 (54% combined) and 6-3 (40% combined), consistent with ~75% hold rates. The 52% three-set probability (vs 48% straight sets) indicates a coin-flip match structure—neither player dominates. This elevated three-set frequency is a primary driver pushing expected totals above market lines.

The cumulative distribution shows:

• 54% chance of 23+ games
• 77% chance of Under 24.5
• Only 18% chance of Under 20.5

This distribution strongly supports the Over on low totals lines (18.5-20.5) while cautioning against high totals (23.5+).

Totals Analysis

Model vs Market

Totals Probabilities by Line

Key Drivers

Over Drivers:

1. Weak Hold Rates (75%): Both players struggle to hold serve, creating a break-heavy environment
2. High Three-Set Probability (52%): Coin-flip match structure adds ~10 games when going three sets
3. Modest Consolidation (76%): Neither player locks down leads, extending sets
4. Player Averages: Mejia 23.0 avg, Basilashvili 24.4 avg → both exceed market line

Under Drivers:

1. Low TB Probability (15%): 75% hold rates suppress tiebreaks, preventing bonus games
2. Straight Sets Scenario (48%): If either wins 2-0, totals could land 19-21 games

Expected Total Games: 22.8 (95% CI: 19-26)

The 4.3-game gap between the model’s 22.8 fair line and the market’s 18.5 line represents a massive inefficiency. Even in the worst-case scenario (straight sets with minimal breaks), the match should reach 19-20 games. The model’s 82% Over probability vs the no-vig market’s 52.5% produces a robust 29.5pp edge.

Handicap Analysis

Model vs Market

Spread Coverage Probabilities

Model View (Mejia as favorite):

Market View (Basilashvili -5.5):

Key Drivers

Mejia Margin Drivers:

• Break rate edge: +3.4pp (29.1% vs 25.7%)
• Game win percentage: +2.1pp (52.1% vs 50.0%)
• Dominance ratio: 1.42 vs 1.13
• Recent form: 44-31 vs 33-31

Basilashvili Margin Reducers:

• Serving for match: 87.0% vs 78.6% (elite closer)
• BP conversion: 64.2% vs 56.2% (capitalizes on chances)
• Elo parity: Both 1200 rating

Expected Game Margin: Mejia -1.2 games (95% CI: -5 to +3)

The model fundamentally disagrees with the market on the favorite. The model projects Mejia as a slight favorite (-1.2 games) based on superior break rates and recent form, while the market favors Basilashvili at -5.5 games. This directional conflict creates uncertainty. Even if we consider Basilashvili +5.5 (the inverse of the market line), the model’s 18% probability of Mejia covering -5.5 suggests only ~25% probability of Basilashvili covering -5.5, well below the no-vig market’s 55%.

Given the Elo parity, wide confidence interval, and market disagreement, we PASS on the spread market.

Head-to-Head

No head-to-head data available in the briefing file.

Without H2H context, we rely on the player priors and statistical profiles. The Elo parity (both 1200) and similar playing styles (weak holds, high break rates) support the model’s projection of a close, competitive match.

Market Comparison

Current Market Lines

Totals:

• Line: 18.5
• Over: 1.81 (55.2% implied)
• Under: 2.00 (50.0% implied)
• No-vig: Over 52.5% / Under 47.5%
• Vig: 5.2%

Spread:

• Line: Basilashvili -5.5
• Basilashvili -5.5: 1.72 (58.1% implied)
• Mejia +5.5: 2.10 (47.6% implied)
• No-vig: Basilashvili 55.0% / Mejia 45.0%
• Vig: 5.7%

Model vs Market Summary

Totals: The model identifies a massive edge on Over 18.5. The market appears to be significantly underpricing the total games expectation.

Spread: The model and market disagree on the favorite, creating uncertainty. The model’s Mejia -1.2 projection conflicts with the market’s Basilashvili -5.5 line, suggesting potential mispricing, but the directional conflict and wide confidence interval warrant caution.

Recommendations

Totals: Over 18.5

Rationale:

• Model projects 22.8 total games vs market line of 18.5 → 4.3-game gap
• Both players hold only ~75%, creating frequent breaks (8-9 per match)
• 52% three-set probability adds significant game volume
• Even straight-sets scenarios (6-3, 6-4) typically reach 19-20 games
• Player historical averages (23.0 and 24.4) both exceed market line
• 29.5pp edge is among the largest edges in our analysis framework

Risk Factors:

• Dominant straight-sets win (6-2, 6-2) would land at 16 games
• Model error on three-set probability could reduce totals

Verdict: The edge is substantial and supported by multiple statistical angles. Strong Over 18.5 bet at 2.0 units.

Spread: PASS

Rationale:

• Model and market disagree on the favorite (directional conflict)
• Model projects Mejia -1.2, market favors Basilashvili -5.5
• Wide confidence interval (-5 to +3) indicates high variance
• Elo parity (both 1200) supports uncertainty
• Model P(Mejia -5.5) = 18% vs market no-vig 55% for Basilashvili -5.5

Verdict: The directional disagreement and wide confidence interval create too much uncertainty. PASS on spread market.

Confidence & Risk Assessment

Overall Confidence: HIGH (Totals Only)

Totals Confidence Factors: ✅ Multiple statistical angles support Over (hold rates, three-set probability, player averages) ✅ 4.3-game gap between model and market is substantial ✅ High data quality from api-tennis.com (75 matches for Mejia, 64 for Basilashvili) ✅ 29.5pp edge well above minimum threshold (2.5pp)

Spread Confidence Factors: ❌ Model and market disagree on favorite ❌ Wide confidence interval indicates high variance ❌ Elo parity limits predictive power

Key Risks & Unknowns

Totals Risks:

1. Dominant Straight-Sets Win: If one player wins 6-2, 6-2 or 6-1, 6-3, totals could land 16-18 games
   • Mitigation: Model gives this only 7% probability
2. Model Error on Three-Set Frequency: If match is more lopsided than projected, straight sets (48%) could rise
   • Mitigation: Historical three-set frequencies (37%) support 50-50 projection
3. In-Match Variance: A single tiebreak could add/remove 1-2 games from expectation
   • Mitigation: 15% TB probability is factored into the 22.8 expectation

Spread Risks:

1. Directional Uncertainty: Model favors Mejia, market favors Basilashvili
2. Match Closure Differential: Basilashvili’s 87% serving for match vs Mejia’s 78.6% could swing close sets
3. Wide Confidence Interval: -5 to +3 game margin range indicates high variance

Data Quality Concerns:

• Surface listed as “all” (generic) rather than specific surface
  • Indian Wells is hard court, so assume hard court adjustments apply
• No H2H data available
• Modest tiebreak sample sizes (11 and 13 TBs)

Stake Sizing Justification

Totals: 2.0 units (HIGH confidence)

• Edge of 29.5pp far exceeds 5% threshold for HIGH confidence
• Expected value of +53.3% at 1.81 odds
• Multiple corroborating statistical drivers

Spread: 0 units (PASS)

• Directional conflict between model and market
• Insufficient edge alignment

Sources

Data Collection:

• api-tennis.com (primary stats source)
  • Player profiles and match history (75 matches for Mejia, 64 for Basilashvili)
  • Hold/break percentages from point-by-point data
  • Clutch stats (BP conversion/saved, tiebreak records)
  • Key games (consolidation, breakback, serving for set/match)

Odds Data:

• api-tennis.com (multi-bookmaker aggregation)
  • Totals: 18.5 (Over 1.81 / Under 2.00)
  • Spreads: Basilashvili -5.5 (1.72 / 2.10)
  • Bookmakers: 10Bet, WilliamHill, bet365, Marathon, Betfair, Pinnacle, 1xBet, Betano

Elo Ratings:

• Jeff Sackmann’s Tennis Data (GitHub CSV)
  • Overall Elo: Both 1200
  • Surface-specific Elo: Both 1200 (hard/clay), 1170 (grass)

Analysis Methodology:

• .claude/commands/analyst-instructions.md (Tennis AI framework)
• .claude/commands/report.md (Report generation template)

Collection Timestamp: 2026-03-02T06:22:49+00:00

Verification Checklist

• Match details verified (players, tournament, date, surface)
• Data quality confirmed (HIGH completeness from api-tennis.com)
• Hold/break statistics extracted for both players
• Tiebreak data included with sample sizes
• Recent form and Elo ratings incorporated
• Game distribution model built with set score probabilities
• Expected total games calculated with 95% CI (22.8, range 19-26)
• Expected game margin calculated with 95% CI (Mejia -1.2, range -5 to +3)
• Market odds compared (totals and spreads)
• No-vig probabilities calculated
• Edge calculations completed
  • Totals: +29.5pp edge on Over 18.5
  • Spread: -10.0pp edge (PASS due to directional conflict)
• Confidence levels assigned
  • Totals: HIGH confidence (2.0 units)
  • Spread: PASS (0 units)
• Risk factors identified
• Stake sizes determined
• Totals recommendation: Over 18.5 @ 1.81 odds
• Spread recommendation: PASS
• Report follows template structure
• All statistics sourced and timestamped
• No moneyline analysis included (per market focus)