kt Rolster vs Dplus Kia: Aurora Risk Shapes LCK G3

TL;DR: In LCK Game 3, kt Rolster hold a 52% draft-model edge over Dplus Kia's 48%, but ShowMaker's Aurora is the defining risk: the champion has only a 40% LCK WR over 76G, and he has played it in just 2G. Dplus Kia need that volatile mid pick to unlock their scaling core.

Compositions: kt Rolster vs Dplus Kia

kt Rolster drafted a front-to-back teamfight composition around PerfecT's Gnar, Cuzz's Cho'Gath, Bdd's Ryze, Jiwoo's Ashe and Effort's Seraphine. Ashe arrows and Seraphine's engage create reliable setup for Ryze's follow-up, while Cho'Gath gives the team objective control. Their early game should prioritize safe lanes, poke and pick pressure before grouping around dragons.

Dplus Kia answer with Siwoo's Ambessa, Lucid's Naafiri, ShowMaker's Aurora, Smash's Yunara and Career's Lulu. This is a more skirmish-heavy draft that wants Naafiri access into the backline, then Yunara-Lulu scaling to carry later fights. Aurora is the unusual element: rather than a stable mid pick, Dplus Kia are betting ShowMaker can survive Ryze's pressure and find flank angles in mid-game fights.

Key Picks and Stats

Top lane is kt Rolster's clearest statistical lane edge. Gnar owns a 55.4% global WR over 1093G and a 57.6% LCK WR over 99G; PerfecT is 83.3% on Gnar in 6G with a 5.0 KDA. Against Ambessa, Gnar posts a 60.7% global WR over 163G and a 68.8% LCK WR over 16G. Siwoo's Ambessa has only a 38.9% LCK WR over 95G, while his own record is 33.3% in 6G with a 2.0 KDA.

Bdd's Ryze brings a 55.1% LCK WR over 138G, plus a 56.2% record in 16G and 6.4 KDA for Bdd. ShowMaker's Aurora is at 39.5% in LCK over 76G, despite his 50.0% mark in 2G and 9.2 KDA. Ryze has a 54.9% global WR over 71G against Aurora, making mid Dplus Kia's highest-variance lane.

Bot lane is less decisive. Ashe has a 58.5% LCK WR over 65G, while Jiwoo is 100.0% in 1G with a 10.0 KDA. Yunara's LCK WR is 47.3% over 110G, though Smash is 58.3% in 12G with a 6.3 KDA. Effort's Seraphine is 72.7% in 11G with a 7.1 KDA, but Seraphine has only a 35.7% LCK WR over 14G versus Lulu.

Draft Edge

Last night's forecast expected Dplus Kia to value Jarvan IV and kt Rolster to take Karma if available, yet neither appears in the shown five picks. The expected bans cannot be confirmed without a ban list. Aurora is the departure from that forecast, and it is a deliberate upside play rather than a standard meta selection.

kt Rolster have cleaner engage, stronger top counter-pick data and better cross-map reliability through Ashe-Ryze. Dplus Kia win if Lucid's Naafiri disrupts Bdd before Ryze can roam, allowing Yunara and Lulu to reach their scaling fights.

Polymarket Market

Polymarket prices Game 3 at 50% for kt Rolster and 50% for Dplus Kia, directly more cautious on kt Rolster than the 52% model. The current series market is 78% kt Rolster to 22% Dplus Kia, versus 22% and 78% before the match. The supplied series delta is -1.0 percentage points, which conflicts with those displayed series snapshots.

The market is far more optimistic about kt Rolster winning the series than this individual game. That makes sense after kt Rolster won Game 1 11-15 in 38:40 and Game 2 13-15 in 43:10, but the Game 3 price respects Dplus Kia's Yunara-Lulu scaling and Aurora's potential to swing a single map.

Prediction

The model remains kt Rolster 52% — Dplus Kia 48%. I would keep that number unchanged: kt Rolster's 59% head-to-head signal and lane data offset Dplus Kia's superior 0.621 Elo and 0.540 season WR. Two close kt wins create momentum, but a long series can also punish a team that repeatedly relies on late-game recoveries.

FAQ

A: It can, but the numbers favor Bdd: Ryze has a 54.9% global WR over 71G against Aurora, while Aurora has a 39.5% LCK WR over 76G.

A: Gnar holds a 68.8% LCK WR over 16G against Ambessa, and PerfecT is 83.3% on the champion in 6G.

*Data: draftlol live draft model + Polymarket, draft close.*