MSI Draft: Hanwha’s Lulu Bet vs LYON’s Lucian-Milio Edge

Hanwha Life Esports turned the whole draft with Delight on Lulu, a pick sitting at 44.4% WR over 705G this season and only 39.1% vs Milio over 64G globally. That signals a deliberate bet on protecting Gumayusi’s Kog'Maw and enabling Zeka’s Yone through extended front-to-back fights, but it also means Hanwha are accepting a lane matchup where Isles’ Milio has repeatedly found answers. If that lane does not hold, LYON (2024 American Team) get a very direct path to snowball through bot priority.

Compositions

LYON (2024 American Team) drafted a cleaner mid-game skirmish and teamfight setup: Dhokla on Gwen gives side-lane pressure, Inspired on Skarner provides reliable engage, Saint on Viktor supplies zone control, and Berserker with Lucian plus Isles on Milio gives early bot priority with strong follow-up. This composition wants early dragon access, first move from bot, and controlled mid-game fights where Skarner can force Hanwha’s carries into Viktor gravity fields and Lucian burst windows.

Hanwha Life Esports drafted a more volatile but dangerous scaling shell. Zeus on Rumble and Kanavi on Vi give the engage layer, Zeka on Yone is the reset threat, and Gumayusi’s Kog'Maw with Delight’s Lulu is the hypercarry core. Their best games come when Vi and Rumble buy enough time for Kog'Maw to free-hit and for Yone to enter after cooldowns are spent. The risk is that their lanes are less naturally stable, especially if Lucian-Milio gets to dictate tempo.

Key Picks and Stats

Top lane is close on paper. Dhokla’s Gwen owns a 55.4% global WR over 271G, while Zeus’ Rumble sits at 51.2% over 935G. The direct matchup is dead even at 50.0% over 8G for both sides, and the model’s lane matchup sample is tiny as well. Gwen matters more later as a split-push and anti-frontline threat, while Rumble is stronger when Hanwha can force grouped fights first.

Jungle is a meaningful LYON lever. Inspired’s Skarner is only 0.0% in 1 MSI game with a 2.0 KDA, but the broader pool is much better: 51.5% global WR over 274G, plus 63.2% vs Vi over 19G. Kanavi’s Vi has the stronger MSI sample at 72.7% over 11G, though his personal MSI mark is 50.0% in 2G with 6.4 KDA. In raw matchup terms, Skarner into Vi is favorable for LYON.

Mid lane is steadier than flashy. Saint’s Viktor is 49.7% global over 437G, 60.0% at MSI over 10G, and 25.0% in 4 MSI games for Saint personally with 5.9 KDA. Zeka’s Yone is 52.4% global over 208G, but only 25.0% at MSI over 4G and 50.0% in 2 MSI games for Zeka with 2.7 KDA. The direct Viktor-Yone matchup is 50.0% over 16G globally, so execution matters more than theory here.

Bot lane is where the draft swings. Berserker’s Lucian is 50.4% global over 514G and 87.5% at MSI over 8G, even though the direct Lucian vs Kog'Maw number is only 25.0% over 8G. Isles’ Milio is 50.4% global over 379G, 83.3% at MSI over 6G, and 56.2% vs Lulu over 64G. Across the Rift, Gumayusi’s Kog'Maw is 47.2% global over 36G, and Delight’s Lulu brings the surprise profile: huge sample, weak return, and a poor matchup into Milio.

Against the pre-draft read, Hanwha’s broader draft flexibility still showed up, and Vi indeed made the final composition. What changed the picture was Lulu: last night’s expected ban conversation made Lulu look like a denial target, yet Hanwha ended up embracing it themselves rather than pivoting to a safer support.

Draft Edge

I give LYON (2024 American Team) a small draft edge because their win conditions are more direct. Lucian-Milio can create lane priority, Skarner has favorable data into Vi, and Viktor plus Gwen scale well enough that LYON do not need to hard-force early chaos. Hanwha Life Esports still have the scarier late-game carry pattern if Gumayusi is untouched, but that outcome depends on a lower-floor support matchup and cleaner engage timing from Kanavi and Zeus.

Polymarket Market

Polymarket is much colder on LYON than the draft model: Game 3 is 28% for LYON (2024 American Team) and 72% for Hanwha Life Esports, while the Series Now market is 22% to 78%. Those figures are not identical, so this is not the deciding-map mirror-price case; the market is slightly more optimistic about LYON in this specific game than in the series as a whole by 6 percentage points. Series pre-match odds are not provided here, so there is no grounded way to measure how far the live series number has moved from open. The likely market logic is simple: Hanwha’s baseline team strength remains respected, but the draft-specific lane data, especially Lucian-Milio into Kog'Maw-Lulu and Skarner into Vi, gives LYON more immediate upset paths than the overall series price implies.

Prediction

The model opens at 55% for LYON (2024 American Team) and 45% for Hanwha Life Esports. I trim that slightly to 52% for LYON (2024 American Team) and 48% for Hanwha Life Esports because Hanwha’s Kog'Maw-Lulu-Yone core can flip the map on two clean fights, and the 1-1 series state keeps pressure high on whichever bot lane misplays first. Even so, LYON’s draft is easier to pilot from minute 1 to minute 25, and that matters in a Game 3 where one early objective sequence can decide side control.