A model estimate for the 2023 calendar. The game decides entry in compiled code; the files supply its inputs, and this model fills in who enters. Treat these as planning ranges, not guarantees. The ranges get tighter as observed values are added to calibration.json.
NbWildcard. Qualifiers = qualifying draw ÷ 2^QualifNbRound, where QualifDraw −1 means the main-draw size and −2 means half of it. These reproduce 104 direct places at Slams and 12 qualifiers at 96-draw 1000s.TypeRank order (Slams, 1000s, 500s, 250s, Challengers, ITF). Tiers with TopPresence ≥ 0.9 (Slams, 1000s) take everyone who gets in. For other weeks, each rank enters with a weekly probability derived from the game's own ranking export: events played in 52 weeks, minus the Slams and 1000s that rank reaches, spread over the free weeks. Entrants go to the best level they can get into.TopPresence as the weekly entry probability.| Level | Events | Main-draw cutoff rank (25th / median / 75th pct.) | ≈ Points at median | Qualifying cutoff rank (median) | ≈ Points |
|---|---|---|---|---|---|
| Grand Slam | 4 | 104 / 104 / 104 | 359 | 232 | 223 |
| 1000 | 9 | 49 / 79 / 79 | 464 | 127 | 315 |
| 500 | 14 | 77 / 131 / 140 | 310 | 233 | 222 |
| 250 | 38 | 109 / 155 / 239 | 288 | 354 | 93 |
| Challenger / 125 | 150 | 257 / 356 / 526 | 93 | 610 | 25 |
| Futures / ITF | 66 | 487 / 645 / 832 | 22 | 779 | 12 |
| Satellite / ITF low | 34 | 495 / 767 / 1110 | 13 | 880 | 7 |
| Level | Events | Main-draw cutoff rank (25th / median / 75th pct.) | ≈ Points at median | Qualifying cutoff rank (median) | ≈ Points |
|---|---|---|---|---|---|
| Grand Slam | 4 | 110 / 110 / 110 | — | 245 | — |
| 1000 | 10 | 51 / 88 / 88 | — | 142 | — |
| 500 | 15 | 56 / 56 / 56 | — | 128 | — |
| 250 | 23 | 152 / 165 / 165 | — | 293 | — |
| Challenger / 125 | 91 | 158 / 227 / 304 | — | 391 | — |
| Futures / ITF | 119 | 358 / 438 / 555 | — | 652 | — |
| Satellite / ITF low | 117 | 540 / 623 / 755 | — | 781 | — |
Cutoffs vary by week: a Challenger in a Slam's second week, or in a week with five Challengers, is easier to enter. Each event page shows its own estimate.
Record observed values from the game in calibration.json. After registration closes, open a tournament's draw or entry list and read Last direct acceptance. With two or more observations at a level, that level is rescaled by the median observed ÷ model ratio.
No observations yet: the model is uncalibrated.