Rankings

Of the tsumo wins by others, the share where the player was dealer and paid a mangan or more. Lowest (luckiest) first. Bars show the rate itself from 0%, shorter being luckier (the league average is about 12%). Aligned with the Paifuya and Tenhou definition on 9/28.

Of riichi wins, the share won with ippatsu, as a difference from the league average. Aligned with the Paifuya and Tenhou definition on 9/28.

How often a draw advanced the hand, minus the chance it would, computed from the hand at that moment (per draw; plus means luckier). The chance comes from the hand the player actually kept: how many of the tiles that would lower its shanten are still among the tiles left to draw (tiles in other players' hands excluded). Good tile efficiency, keeping shapes with more acceptances, goes into the chance, so the difference averages zero whatever the playing style and can be read as luck with the style taken out as far as possible. Of all the luck, it has the largest effect on results. The dealer's first draw is left out as part of the deal. Details in section 4.1 of the methodology.

Of hands with riichi, the share that dealt in (including on the riichi discard).

The share of riichi with 6 or more winning tiles unseen from the player's view (counted as Paifuya counts good-wait riichi).

Of hands with riichi, the share that dealt in after the riichi stood (not on the riichi discard).

The estimate adjusted for the measured chance factors (starting hands, draws, ura-dora).

The share of first calls after which the player is already tenpai with the discard. It shows whether a player calls from far away or close to tenpai, not good or bad.

Right after another player's first riichi, with a closed 1-shanten hand: the share of first discards that were not genbutsu. A push/fold tendency, not good or bad.

Right after another player's first riichi, with a closed hand at 2-shanten or worse: the share of first discards that were not genbutsu. A push/fold tendency, not good or bad.

Win rate × average win (points gained in winning hands, honba and riichi sticks included): income from wins per hand. Calculated as on Mahjong Soul Paifuya.

Deal-in rate × average deal-in (points paid, honba included): loss from deal-ins per hand. Smallest loss first. Calculated as on Mahjong Soul Paifuya.

Win efficiency − deal-in loss: a rough balance per hand from wins and deal-ins only (others' tsumo and draws not included). Calculated as on Mahjong Soul Paifuya.

The average of tiles discarded at the table at the win ÷ 4 + 1: how fast wins come. Fastest first. Counted as on Mahjong Soul Paifuya.

The average points per win. Bars show the points themselves from 0.

The average points per deal-in, lowest first. Bars show the points themselves from 0; shorter means cheaper deal-ins.

Of the hands with riichi, the share where the player declared before anyone else. Not the number of riichi itself.

With no riichi out, in 1- and 2-shanten hands up to the player's 12th discard: the share of discards that, among those keeping shanten lowest, keep the most ukeire. Between players no clear link with results has been found, so this value alone cannot judge discards or skill.

The turn of tenpai in hands that reach tenpai closed by turn 12, adjusted to league-average starting shanten and draw luck, so only the speed of hand building is compared.

The share of riichi declared with the discard that made tenpai. It rises across the league year by year, so it is compared as the difference from each season's league average.

After a first riichi, the share where someone chased with riichi, as a difference from the league average (plus: chased less). No differences between players were found; it depends on the opponents, so it counts as luck.

The average points won or lost in hands with riichi (riichi sticks and honba included).

The average points won or lost in hands with calls.

The share of wins made without riichi or calls. Aligned with the Paifuya and Tenhou definition on 9/28.

1. #1 Yu Suzuki +4.2
2. #2 Hiroshi Daigo +3.1
3. #3 Arisa Date +2.8
4. #4 Futoshi Watanabe +2.8
5. #5 Hisato Sasaki +2.4
6. #6 Kei Nakabayashi +2.2
7. #7 Ken Sonoda +2.0
8. #8 Geki Shimoishi +1.8
9. #9 Akina Mizuhara +1.6
10. #10 Genta Takeuchi +1.6
11. #11 Takaharu Ooi +1.4
12. #12 Sho Shiratori +1.4
13. #13 Tomohiro Honda +1.3
14. #14 Seiichi Kondo +1.2
15. #15 Kenji Katsumata +1.1
16. #16 Shingo Hori +1.1
17. #17 Kosuke Nagai +1.0
18. #18 Saki Kurosawa +1.0
19. #19 Namba Shibukawa +1.0
20. #20 Kazunori Takizawa +0.9
21. #21 Makoto Sawazaki +0.5
22. #22 Maki Asami +0.2
23. #23 Taro Suzuki +0.1
24. #24 Aki Nikaido 0.0
25. #25 Kazuma Ishii -0.2
26. #26 Kotaro Uchikawa -0.2
27. #27 Rio Tojyo -0.2
28. #28 Naoki Setokuma -0.3
29. #29 Go Kobayashi -0.3
30. #30 Daisuke Suzuki -0.5
31. #31 Yoshihiro Matsumoto -0.6
32. #32 Jun Murakami -0.7
33. #33 HIRO Shibata -0.9
34. #34 Yuumi Uotani -0.9
35. #35 Sayaka Okada -0.9
36. #36 Takaya Matsugase -1.2
37. #37 Hiroe Sugawara -1.3
38. #38 Aiko Hinata -1.5
39. #39 Yudai Maehara -1.5
40. #40 Koshin Asakura -1.5
41. #41 Sayaka Kayamori -1.6
42. #42 Satoshi Fujisaki -1.6
43. #43 Takaki Asai -1.7
44. #44 Mari Takamiya -1.8
45. #45 Kanako Maruyama -1.9
46. #46 Masatoshi Sarukawa -2.0
47. #47 Rumi Nikaido -2.3
48. #48 Kana Nakada -2.6
49. #49 Nobuhiro Ishibashi -2.6
50. #50 Masato Hagiwara -7.5

* Ranked by the Estimate, not the actual record, so a lucky short run does not top the table. Bars run from the league average, longer to the right toward the good side (for style metrics, the chosen order). Hover to see the 90% interval of the estimate. Players with close estimates could well be the other way round. Players with 40+ career games, or 10+ games in the season.