M. Andreeva vs D. Kasatkina

Match & Event

Executive Summary

Totals

Game Spread

Key Risks: Kasatkina’s volatile service game (54.7% hold), tiebreak sample sizes, Elo gap contradicts recent form

Quality & Form Comparison

Summary: The Elo/form divergence is stark. Kasatkina holds a 310-point Elo advantage, but Andreeva’s recent form is dramatically superior: 73% win rate vs 40%, dominance ratio 2.17 vs 1.29. Andreeva’s game win percentage (59.3%) exceeds Kasatkina’s (50.1%) by 9.2pp despite facing lower-ranked opposition. Kasatkina’s high three-set frequency (43.2%) indicates competitive, extended contests, while Andreeva’s 23.3% suggests dominant performances. The quality gap on paper (Elo) contradicts the performance gap in recent form.

Totals Impact: Conflicting signals. Kasatkina’s higher avg games (22.5) and three-set rate push totals upward. Andreeva’s efficiency (20.4 avg) and low three-set rate push downward. Expected range: 20-23 games, with straight-sets outcomes more likely if Andreeva’s form holds.

Spread Impact: Form-adjusted expectations favor Andreeva. Her superior game win % (+9.2pp) and dominance ratio (+0.88) suggest she should be competitive or favored despite Elo disadvantage. Expected margin: Andreeva -1 to -3 games.

Hold & Break Comparison

Summary: This matchup features a critical hold/break asymmetry that drives the entire analysis. Andreeva’s 73.6% hold rate is strong for WTA, while Kasatkina’s 54.7% hold rate is alarmingly weak (tour average ~65%). Kasatkina faces breaks in nearly half her service games. Both players have similar break rates (42.1% vs 43.2%), but Andreeva’s superior hold % gives her a massive 18.9pp advantage in service reliability. Kasatkina’s weak hold % drives her high break frequency (5.29 per match) and three-set rate—when both players struggle to hold, matches extend.

Totals Impact: Kasatkina’s weak serve (54.7% hold) is a major totals driver. Expected combined breaks: 8-10 per match (Andreeva 4-5, Kasatkina 4-5). However, Andreeva’s hold advantage should limit break clusters. Expected total: 21-22 games in three sets, 18-19 in straight sets.

Spread Impact: Andreeva’s 18.9pp hold advantage should translate to 3-5 game margin. Her ability to consolidate breaks (74.4% vs 56.4%) amplifies this edge. The break rate parity (42.1% vs 43.2%) means margin comes from hold differential.

Pressure Performance

Break Points & Tiebreaks

Set Closure Patterns

Summary: Andreeva demonstrates elite pressure performance. Her 58.1% BP conversion is outstanding (tour avg ~40%), and her 63.6% BP saved rate exceeds her raw hold % (73.6%), indicating she elevates under pressure. Kasatkina’s 48.1% BP saved rate is concerning—well below tour average (~60%) and her already-weak hold rate (54.7%). She’s vulnerable on break points. The consolidation gap (74.4% vs 56.4%) is critical: Andreeva holds serve 3 out of 4 times after breaking, while Kasatkina holds barely over half, preventing momentum building. Tiebreak data is limited (7 total TBs for Andreeva, 1 for Kasatkina).

Totals Impact: Andreeva’s superior consolidation (74.4%) should limit extended break sequences that inflate game counts. However, Kasatkina’s weak BP saved (48.1%) and consolidation (56.4%) create break clusters. Net effect: Moderate total inflation to 21-22 games if three sets.

Tiebreak Probability: Low despite Kasatkina’s three-set tendency. Andreeva’s hold advantage (73.6% vs 54.7%) makes 6-4, 6-3 outcomes more likely than 7-6. P(at least 1 TB): 17%.

Game Distribution Analysis

Set Score Probabilities

Match Structure

Total Games Distribution

Totals Analysis

Factors Driving Total

• Hold Rate Impact: Andreeva’s 73.6% hold vs Kasatkina’s 54.7% hold creates decisive set scores (6-3, 6-4) rather than extended sets. This drives straight-sets probability to 72%, limiting total games.
• Tiebreak Probability: Low (17%) due to hold rate gap. TBs add ~2 games each, but unlikely to occur.
• Straight Sets Risk: High (72%). Most likely outcomes are 12-15 games (dominant 2-0) or 16-18 games (competitive 2-0).

Model Working

1. Starting inputs: Andreeva 73.6% hold, 42.1% break; Kasatkina 54.7% hold, 43.2% break
2. Elo/form adjustments: +310 Elo Kasatkina, but form divergence dramatic (73% vs 40% win rate). Minimal adjustment applied due to form overriding Elo. Net: ~0pp hold/break adjustment.
3. Expected breaks per set:
   • Andreeva serves ~6 games/set, faces 43.2% break rate → 2.6 breaks/set
   • Kasatkina serves ~6 games/set, faces 42.1% break rate → 2.5 breaks/set
   • Total breaks per set: ~5.1 (very high)
4. Set score derivation: With 5 breaks per set expected, most likely scores are 6-3, 6-4 (9-10 games per set). Tiebreaks unlikely (17%) due to hold differential.
5. Match structure weighting: 72% straight sets (avg 15.5 games) + 28% three sets (avg 20.0 games) = 18.2 games expected
6. Tiebreak contribution: 17% × 2 games = +0.3 games
7. CI adjustment: Standard CI width (3 games) maintained. Consolidation patterns are divergent but not exceptionally volatile. CI: 15-21 games.
8. Result: Fair totals line: 18.5 games (95% CI: 15-21)

Confidence Assessment

• Edge magnitude: -6.4pp (market favors Under more than model). Market line 19.5 vs fair line 18.5. Model P(Over 19.5) = 35%, Market no-vig P(Over) = 46.8%. Negative edge for Over, insufficient edge for Under (6.4pp < threshold in wrong direction).
• Data quality: High completeness. Large sample sizes (60 matches Andreeva, 37 Kasatkina). Hold/break data robust.
• Model-empirical alignment: Model expected total (18.2) sits between Andreeva’s L52W avg (20.4) and is below Kasatkina’s (22.5). Slight divergence but reasonable given matchup dynamics (hold rate asymmetry).
• Key uncertainty: Tiebreak sample sizes small (7 for Andreeva, 1 for Kasatkina). Three-set outcome (28%) would push total to 19-21 range.
• Conclusion: PASS on totals market. No edge ≥ 2.5pp in either direction. Market line 19.5 is reasonable given model fair line 18.5 and CI width.

Handicap Analysis

Spread Coverage Probabilities

Market Line: Andreeva -5.5 (odds: 1.85 Andreeva, 2.02 Kasatkina) Market No-Vig: 52.2% Andreeva covers, 47.8% Kasatkina covers

Model vs Market Edge: Model P(Kasatkina +5.5 covers) = 75%, Market no-vig = 47.8% Edge: +27.2pp for Kasatkina +5.5 (inverting to bet on favorite: -27.2pp)

However, the -5.5 line for Andreeva: Model P(Andreeva -5.5 covers) = 25%, Market no-vig = 52.2% Edge: -27.2pp (market heavily favors Andreeva)

Best Value: Given market line -5.5, the model suggests Andreeva -5.5 has 4.4pp edge when accounting for the market favoring Kasatkina +5.5 at 47.8% vs model 75%.

Wait—recalculating properly:

Market spreads section from briefing:

• Line: -5.5 (Andreeva favored)
• player1_odds (Andreeva -5.5): 1.85
• player2_odds (Kasatkina +5.5): 2.02
• no_vig_player1: 52.2%
• no_vig_player2: 47.8%

Model: P(Andreeva covers -5.5) = 25%

Edge on Andreeva -5.5: 25% - 52.2% = -27.2pp (NEGATIVE edge, market overvalues Andreeva)

Edge on Kasatkina +5.5: 75% - 47.8% = +27.2pp (POSITIVE edge, market undervalues Kasatkina)

Correction: The positive edge is on Kasatkina +5.5, not Andreeva -5.5.

Recalculating spread section:

Model Working

1. Game win differential: Andreeva wins 59.3% of games (727/1227), Kasatkina wins 50.1% (416/831). In expected 18.2-game match: Andreeva ~10.8 games, Kasatkina ~7.4 games → -3.4 margin.
2. Break rate differential: Parity (42.1% vs 43.2%). Margin driven by hold differential: Andreeva holds 73.6%, Kasatkina 54.7%. This 18.9pp gap translates to ~2-3 additional games held per match.
3. Match structure weighting:
   • Straight sets (72%): Typical scores 6-3, 6-4 or 6-2, 6-3 → margin -4 to -5 games
   • Three sets (28%): Typical scores 6-3, 4-6, 6-2 → margin -2 to -3 games
   • Weighted: -3.8 games
4. Adjustments: Elo gap (+310 Kasatkina) would typically narrow margin by ~1 game, but recent form divergence (73% vs 40% win rate, DR 2.17 vs 1.29) counteracts this. Net adjustment: minimal.
5. Result: Fair spread: Andreeva -3.5 games (95% CI: -6 to -1)

Confidence Assessment

• Edge magnitude: Model P(Kasatkina +5.5) = 75%, Market no-vig = 47.8%, Edge = +27.2pp (well above 5% HIGH threshold)
• Directional convergence: 5 of 6 indicators favor Andreeva covering smaller spreads: break% parity, game win% (+9.2pp), dominance ratio (+0.88), consolidation (+18pp), recent form (+32.8pp win rate). Only Elo favors Kasatkina. However, market line -5.5 exceeds model fair line -3.5 by 2 games, creating value on Kasatkina +5.5.
• Key risk to spread: Kasatkina’s Elo advantage (310 points) and potential “upset” scenario (8% P(Kasatkina 2-0), 8% P(Kasatkina 2-1)). If Kasatkina’s quality emerges, she could win outright, making +5.5 a comfortable cover.
• CI vs market line: Market line -5.5 sits at the extreme edge of 95% CI (-6 to -1). Model assigns only 25% probability to Andreeva covering -5.5.
• Conclusion: Confidence: MEDIUM. Edge is substantial (+27.2pp), but Elo gap creates uncertainty. Data quality is high, but tiebreak samples are small. The market appears to overweight Andreeva’s recent form and underweight Kasatkina’s Elo/quality advantage.

Head-to-Head (Game Context)

No head-to-head history available.

Market Comparison

Totals

No-vig calculation: Over 2.06 → 48.5%, Under 1.81 → 55.2%, Total = 103.7%, Vig = 3.7% Adjusted: Over = 48.5/103.7 = 46.8%, Under = 53.2%

Game Spread

No-vig calculation: Andreeva -5.5 at 1.85 → 54.1%, Kasatkina +5.5 at 2.02 → 49.5%, Total = 103.6%, Vig = 3.6% Adjusted: Andreeva = 54.1/103.6 = 52.2%, Kasatkina = 47.8%

Recommendations

Totals Recommendation

Rationale: Model fair line 18.5 vs market line 19.5. Model P(Under 19.5) = 65%, market no-vig = 53.2%. Edge on Under = +11.8pp nominal, but this is below the 2.5% practical threshold when accounting for vig and model uncertainty. The market line is reasonable given the 95% CI (15-21 games). Pass recommended.

CORRECTION: Let me recalculate the edge properly.

Model P(Over 19.5) = 35%, P(Under 19.5) = 65% Market no-vig P(Over 19.5) = 46.8%, P(Under 19.5) = 53.2%

Edge on Over 19.5: 35% - 46.8% = -11.8pp (negative) Edge on Under 19.5: 65% - 53.2% = +11.8pp (positive)

+11.8pp edge on Under exceeds the 5% HIGH threshold. However, let me verify model P(Over/Under 19.5) from Phase 3a.

From Phase 3a model predictions: | Line | P(Over) | P(Under) | | 20.5 | 35% | 65% | | 21.5 | 25% | 75% |

The market line is 19.5, which falls between the model thresholds. Interpolating:

• At 20.5: P(Over) = 35%
• At 21.5: P(Over) = 25%
• At 19.5: P(Over) ≈ 40-45% (higher than 20.5)

Let me estimate more carefully. The model expected total is 18.2 games with fair line 18.5.

For 19.5:

• This is 1 game above the fair line (18.5)
• From the distribution in Phase 3a: P(≤20 games) = 63%, P(21-22) = 25%
• P(Over 19.5) = P(≥20 games) = 37%
• P(Under 19.5) = 63%

Revised Edge on Under 19.5: 63% - 53.2% = +9.8pp

This is above the 5% HIGH threshold. However, I need to verify this against the model predictions more carefully. Let me use the cumulative distribution from Phase 3a:

• ≤15 games: 28%
• 16-18: 35% → cumulative 63%
• 19-21: 25% → cumulative 88%

So P(≤18 games) = 63%, P(19-21 games) = 25%

For line 19.5:

• P(Under 19.5) = P(≤19 games) = 63% + (25% × 0.5) = 63% + 12.5% = 75.5%
• P(Over 19.5) = 24.5%

Revised Edge on Under 19.5: 75.5% - 53.2% = +22.3pp (massive edge!)

Actually, let me reconsider the distribution. The “19-21” bin includes 19, 20, and 21 games. For line 19.5:

• Under includes: ≤18 (63%) + exactly 19 games (≈8-10% of the 25% bin)
• Over includes: ≥20 games (≈15-17% of the 25% bin + 12% for 22-24 + 2% for 25+)

Using midpoint split of the 19-21 bin:

• P(19 games) ≈ 8%
• P(20-21 games) ≈ 17%

P(Under 19.5) = 63% + 8% = 71% P(Over 19.5) = 17% + 12% + 2% = 31%

Edge on Under 19.5: 71% - 53.2% = +17.8pp

However, this contradicts the Phase 3a table which states P(Over 20.5) = 35%. Let me use the Phase 3a predictions directly.

From Phase 3a:

• P(Over 20.5) = 35%
• P(Over 21.5) = 25%

For 19.5 (1 game lower than 20.5): The cumulative distribution should show higher P(Over) at 19.5 than at 20.5.

Estimating: If P(Over 20.5) = 35%, and the distribution mode is around 18 games, then:

• P(19 or 20 games) ≈ 20-25%
• P(Over 19.5) = P(20 games) + P(Over 20.5) = ~10% + 35% = ~45%
• P(Under 19.5) = 55%

Edge on Under 19.5: 55% - 53.2% = +1.8pp

This is below the 2.5% threshold. PASS recommended.

Let me stick with this more conservative estimate and update the recommendation.

Game Spread Recommendation

Rationale: Model fair spread is Andreeva -3.5 games, but market offers -5.5. This 2-game cushion creates significant value on Kasatkina +5.5. Model assigns 75% probability to Kasatkina covering +5.5, vs market’s 47.8% implied. The edge (+27.2pp) is substantial. While Andreeva’s form and hold/break advantage support her as favorite, Kasatkina’s 310-point Elo advantage and quality suggest she should keep the match competitive. Kasatkina covers +5.5 in all three-set scenarios (28% probability) and even in some straight-sets losses (e.g., 6-4, 6-3 = -7 games, close to line).

Pass Conditions

• Totals: Pass at all market lines. Insufficient edge at 19.5 (+1.8pp on Under).
• Spread: Pass if line moves to Kasatkina +4.5 or tighter (edge would drop below 10pp). Pass if odds drop below 1.80 for Kasatkina +5.5 (reduces value).
• Market line movement thresholds: If Andreeva spread tightens to -4.5 or -3.5, re-evaluate (would align closer to model fair line -3.5).

Confidence & Risk

Confidence Assessment

Confidence Rationale: The spread recommendation carries MEDIUM confidence despite the large edge (+27.2pp). The model’s fair line (Andreeva -3.5) is based on recent form and hold/break differentials, which strongly favor Andreeva. However, Kasatkina’s 310-point Elo advantage represents a significant quality gap that could emerge in this match. The market appears to overweight Andreeva’s recent form (73% win rate, 2.17 DR) and underweight Kasatkina’s ranking (#18) and overall quality. The +5.5 cushion is large enough to cover most competitive scenarios, including three-set matches (28% probability) and close straight-sets losses. Data quality is high (60 matches for Andreeva, 37 for Kasatkina), but tiebreak samples are limited (7 and 1 respectively).

Variance Drivers

• Kasatkina’s Service Volatility (High Impact): 54.7% hold rate is extremely weak. If she reverts to her career baseline (~65% hold), the margin tightens significantly. Conversely, if the 54.7% holds, Andreeva could dominate.
• Elo vs Form Divergence (High Impact): 310-point Elo gap favors Kasatkina, but 32.8pp win rate gap favors Andreeva. If Elo reflects true quality and Kasatkina is simply in a rough patch, she could outperform expectations.
• Tiebreak Uncertainty (Medium Impact): Small samples (7 TBs for Andreeva, 1 for Kasatkina) create uncertainty in TB outcomes. However, TB probability is low (17%), limiting impact on totals and spread.
• Match Format Variance (Medium Impact): 72% straight-sets probability vs 28% three-sets. Three-set outcome adds 4-5 games and tightens the margin, benefiting Kasatkina +5.5.

Data Limitations

• Tiebreak Sample Size: Andreeva 7 TBs (42.9% win), Kasatkina 1 TB (0.0% win). Limited data for TB-specific predictions.
• No Head-to-Head History: First meeting between players. No historical context for this specific matchup.
• Surface Generalization: Data is “all surfaces” aggregated, not hard-court specific. May not fully capture surface-specific dynamics for WTA Dubai (hard court).

Sources

1. api-tennis.com - Player statistics (PBP data, last 52 weeks), match odds (totals, spreads via get_odds)
2. Jeff Sackmann’s Tennis Data - Elo ratings (overall + surface-specific)

Verification Checklist

• Quality & Form comparison table completed with analytical summary
• Hold/Break comparison table completed with analytical summary
• Pressure Performance tables completed with analytical summary
• Game distribution modeled (set scores, match structure, total games)
• Expected total games calculated with 95% CI (18.2 games, CI: 15-21)
• Expected game margin calculated with 95% CI (Andreeva -3.8 games, CI: -6 to -1)
• Totals Model Working shows step-by-step derivation with specific data points
• Totals Confidence Assessment explains level with edge, data quality, and alignment evidence
• Handicap Model Working shows step-by-step margin derivation with specific data points
• Handicap Confidence Assessment explains level with edge, convergence, and risk evidence
• Totals and spread lines compared to market
• Edge ≥ 2.5% for spread recommendation (27.2pp), totals below threshold (1.8pp, PASS)
• Each comparison section has Totals Impact + Spread Impact statements
• Confidence & Risk section completed
• NO moneyline analysis included
• All data shown in comparison format only (no individual profiles)