Game plan

Karolina Muchova v Marta Kostyuk

Every number combines what Karolina Muchova does well with what Marta Kostyuk allows, each measured against the tour average and shrunk toward it when the sample is thin. Flip perspective

Forecast

Karolina Muchova wins, best of 3 72%90%: 54%–85% · best of 5: 77%

Serve points won 62.5% / 58.0% Karolina / Marta · tour 56.7%

Strengths only, no similarity priors 66%serve 62.1% / 58.8%

Each player's serve and return strength is fitted against every opponent they were charted against, so a record built on weak opponents counts for less. Both were charted enough in the last three seasons, so those carry the most weight. The result is then nudged by Karolina Muchova's record against Marta Kostyuk's tactical lookalikes and in their charted head-to-heads (lookalikes: +0.5 on serve, +4.7 on return vs expectation (954 points); head-to-head: +2.3 on serve, −2.9 on return vs expectation (387 points)). A game-by-game Markov chain turns point odds into match odds; the 90% range covers the uncertainty in the two strengths, not the nudges. Charted matches lean toward big events, so treat this as a scouting estimate, not a betting line.

Head to head, per 100 shots

Career	Karolina	Marta
Direction choice	−0.01 ±0.05 better than 52%	−0.06 ±0.05 better than 43%
Shot selection	−0.17 ±0.09 better than 30%	+0.13 ±0.09 better than 60%
Execution	+0.22 ±0.28 better than 70%	−0.34 ±0.29 better than 41%
Points left on the table	2.88 ±0.08 lower than 18%	2.69 ±0.07 lower than 36%

Each player's career against an average tour player in the same position, adjusted for opponent strength, with a 90% margin (shots clustered by match). Points left on the table is the gap to the best-value direction for the same stroke, so lower is better. Percentiles are within each player's own tour. A side is highlighted only when the gap is larger than the margin on the difference.

Structural compatibility

Expected edge per 100 rally shots from style alone: Karolina Muchova −1.13, Marta Kostyuk −0.87. Each player's shot mix weighted by their own skill with each shot and by how much the other gives up against it. This is why some rankings gaps don't hold in a given matchup.

Serve plan

The share of points the server wins when a first serve lands in that direction (hover a rate for its 90% interval; ± is the 90% margin). "Matchup" combines the server's rate with how this returner handles that serve. "Optimal" is the mix that wins most against this returner once they start reading a habit, at the response measured across the tour, and only within the range servers' habits actually vary. The gain over the current mix is how exploitable that mix is.

Karolina Muchova serving

Deuce court

1st serve	Now	Karolina wins	v Marta	Matchup	Optimal
Wide	40%	68%	63%	66.1%±4.0	40%
Body	22%	58%	57%	57.8%±5.6	7% ▼
T	38%	74%	68%	74.6%±4.0	53% ▲

Optimal v Marta Kostyuk: +1.3±0.8 per 100 first serves (faults included) over the current mix. Serving T every time would read +7.1 per 100 first serves in before the returner adjusts.

Ad court

1st serve	Now	Karolina wins	v Marta	Matchup	Optimal
Wide	48%	68%	64%	66.4%±4.2	63% ▲
Body	14%	57%	58%	58.0%±6.8	0% ▼
T	39%	67%	61%	64.1%±4.3	37% ▼

Optimal v Marta Kostyuk: +0.5±0.8 per 100 first serves (faults included) over the current mix, inside the 90% margin. Serving wide every time would read +2.0 per 100 first serves in before the returner adjusts.

Marta Kostyuk serving

Deuce court

1st serve	Now	Marta wins	v Karolina	Matchup	Optimal
Wide	36%	70%	65%	69.7%±3.9	51% ▲
Body	24%	60%	58%	60.5%±5.4	9% ▼
T	40%	72%	69%	72.4%±4.1	40%

Optimal v Karolina Muchova: +1.2±0.8 per 100 first serves (faults included) over the current mix. Serving T every time would read +3.9 per 100 first serves in before the returner adjusts.

Ad court

1st serve	Now	Marta wins	v Karolina	Matchup	Optimal
Wide	47%	73%	66%	73.1%±3.8	62% ▲
Body	26%	58%	54%	55.7%±5.9	11% ▼
T	27%	64%	64%	63.4%±5.0	27%

Optimal v Karolina Muchova: +1.2±0.7 per 100 first serves (faults included) over the current mix. Serving wide every time would read +7.1 per 100 first serves in before the returner adjusts.

Return plan

Value of each return, in points per 100 returns against an average return of the same serve (direction, court, surface): the tour's result with that return, the returner's own edge with it, and what this server gives up when it comes back to that side. Returns with no charted direction are left out, so values compare with each other rather than with zero. Depth isn't a choice here: missed returns have no depth. Serve quality isn't charted, so a block through the middle partly reflects the serve that forced it.

Karolina Muchova returning

1st serve to the forehand

Return	Now	Tour	Own	v Marta	Value
FH through the middle	41%	+4.2	+1.1	+0.7	+6.0±1.8
FH slice through the middle	18%	−6.7	+3.0	−0.7	−4.4±2.2
FH crosscourt	15%	+5.3	−0.7	−3.9	+0.7±3.5
FH down the line	14%	+1.5	−2.9	−2.7	−4.1±3.8
FH slice crosscourt	9%	−6.6	+2.2	−3.3	−7.7±2.7

Lean FH through the middle: +5.9±1.4 per 100 returns v the current mix (1182 returns charted)

1st serve to the backhand

Return	Now	Tour	Own	v Marta	Value
BH through the middle	46%	+6.0	−0.4	−0.2	+5.5±1.6
BH crosscourt	20%	+7.7	+0.9	−1.5	+7.2±2.7
BH slice through the middle	14%	−6.2	+0.2	−2.3	−8.3±2.2
BH down the line	12%	+2.2	−1.2	−2.8	−1.8±3.9
BH slice crosscourt	5%	−4.2	−0.8	−3.6	−8.5±3.0

Lean BH crosscourt: +5.4±2.3 per 100 returns v the current mix (1188 returns charted)

2nd serve to the forehand

Return	Now	Tour	Own	v Marta	Value
FH through the middle	43%	−3.2	−1.4	+0.3	−4.2±2.4
FH crosscourt	28%	+0.5	−0.4	−2.3	−2.1±3.8
FH down the line	21%	−0.6	−1.9	−2.4	−4.9±4.7
FH slice through the middle	5%	−15.2	+0.1	+0.3	−14.8±2.1
FH slice crosscourt	3%	−14.9	+1.3	−1.3	−14.9±2.0

Lean FH crosscourt: +2.5±3.1 per 100 returns v the current mix (247 returns charted, inside the 90% margin)

2nd serve to the backhand

Return	Now	Tour	Own	v Marta	Value
BH through the middle	40%	−2.6	+1.0	−0.1	−1.7±2.1
BH crosscourt	31%	+1.5	−0.4	−2.8	−1.7±3.0
BH down the line	20%	−0.5	−0.3	−1.3	−2.1±4.9
FH inside-in	5%	+0.7	−1.9	−2.3	−3.4±3.9
FH through the middle	3%	−2.7	+1.7	+0.3	−0.6±2.0

Lean FH through the middle: +1.2±2.5 per 100 returns v the current mix (582 returns charted, inside the 90% margin)

Marta Kostyuk returning

1st serve to the forehand

Return	Now	Tour	Own	v Karolina	Value
FH through the middle	41%	+4.2	+0.6	−0.1	+4.7±1.7
FH crosscourt	19%	+5.3	+0.5	−1.4	+4.4±3.3
FH slice through the middle	17%	−6.7	+1.4	−1.7	−7.0±2.1
FH down the line	12%	+1.5	+4.8	−0.7	+5.6±3.7
FH slice crosscourt	8%	−6.6	+0.9	−2.5	−8.2±2.7

Lean FH down the line: +4.1±3.4 per 100 returns v the current mix (1346 returns charted)

1st serve to the backhand

Return	Now	Tour	Own	v Karolina	Value
BH through the middle	52%	+6.0	−0.2	+0.1	+5.9±1.6
BH crosscourt	29%	+7.7	−2.1	+0.3	+5.9±2.4
BH down the line	10%	+2.2	−1.1	+0.2	+1.3±4.1
BH slice through the middle	4%	−6.2	−0.1	−1.1	−7.4±2.5
BH slice crosscourt	3%	−4.2	−0.7	−2.1	−7.0±3.0

Lean BH crosscourt: +1.7±2.0 per 100 returns v the current mix (969 returns charted, inside the 90% margin)

2nd serve to the forehand

Return	Now	Tour	Own	v Karolina	Value
FH through the middle	41%	−3.2	+1.5	+0.5	−1.2±2.4
FH crosscourt	28%	+0.5	−1.0	−3.9	−4.4±3.8
FH down the line	25%	−0.6	+0.7	−1.0	−0.9±4.7
FH slice through the middle	4%	−15.2	+0.6	±0.0	−14.6±2.0
FH slice crosscourt	2%	−14.9	+0.6	+0.1	−14.2±1.9

Lean FH down the line: +1.9±3.8 per 100 returns v the current mix (362 returns charted, inside the 90% margin)

2nd serve to the backhand

Return	Now	Tour	Own	v Karolina	Value
BH through the middle	46%	−2.6	−2.6	−0.7	−5.8±2.0
BH crosscourt	30%	+1.5	−1.1	−1.3	−1.0±2.9
BH down the line	19%	−0.5	+1.0	−2.0	−1.5±4.8
FH inside-out	2%	+1.4	±0.0	−1.0	+0.3±3.4
FH through the middle	1%	−2.7	−0.2	+0.5	−2.3±1.8

Lean BH crosscourt: +2.4±2.4 per 100 returns v the current mix (471 returns charted, inside the 90% margin)

Rally plan

Edge, in points per 100 shots: the hitter's skill with the shot (own) plus how much the receiver usually gives up against it (theirs), both measured against the tour average on hard. Each player's hard record is shrunk toward their all-surface one, so a thin sample on it moves the numbers only a little.