LYON Ignore 36% Odds and Push G2 to the Brink at MSI

TL;DR: With G2 Esports already down 0-1 in the series, Game 2 was their chance to steady MSI 2026. Instead, LYON (2024 American Team) shrugged off a mere 36% market chance, converted a slim 50% draft-model edge into a decisive 18-7 win, and moved one step closer to taking full control.

Key Takeaways

• LYON (2024 American Team) turned a pre-game 36% market price into an 18-7 win, proving public doubt missed how well their composition could front-to-back once fights became organized.
• Inspired on Lee Sin finished 6/2/9 with a +1039 GoldDiff@15, and that jungle lead gave LYON the tempo to crack open lanes before G2’s skirmish setup could breathe.
• Saint piloted Viktor to 6/0/5 with a +1029 lane advantage at 15, showing that the live draft model’s 50% nod to LYON did materialize through scaling and control.

Early Game

G2 entered this map needing an equalizer, but the first phase already hinted that the balance would not hold for long. The draft conversation before the game centered on whether LYON’s cleaner teamfight shape could survive G2’s sharper mid-jungle punish windows, and on stage that theory quickly became practical reality. The live model had favored LYON at 50%, and unlike the broader market, that edge was not imaginary.

The early pressure started in solo lanes and spread outward. Dhokla’s Gnar built a hefty +1791 GoldDiff@15 over BrokenBlade’s Yasuo, turning top side into a permanent threat. In the middle of the map, Saint’s Viktor never gave Caps on Cassiopeia room to dictate the pace, quietly building that +1029 advantage and keeping G2 reactive instead of proactive.

G2 did find some neutral-objective value, taking 3 dragons to 1, but that stat hid the larger flow of the game. While they were collecting drakes, LYON were taking space, kills, and structures. By the time the map widened, the North American side had already built the kind of gold lead that makes every later engage simpler.

The Turning Point

The game truly snapped in LYON’s favor when their engage core started arriving first and cleaner to the same fights G2 wanted to force. Isles on Leona ended only 0/3/13, but those numbers tell the story of initiation: he was there for almost everything important, creating the windows that let his carries unload safely.

Behind that setup, the jungle difference became brutal. Inspired’s Lee Sin at 6/2/9 was not just farming a lead; he was accelerating every lane that already looked stable. Once he got moving, G2’s intended punish pattern with Pantheon and Cassiopeia never really arrived on schedule. SkewMond finished 2/2/4, yet the -1039 GoldDiff@15 showed how far behind the tempo had drifted.

That was the moment the draft edge became visible in real time. LYON had been projected with only a slight advantage on paper, but their top-side stability, strong engage, and double-threat scaling from Viktor and Varus translated directly into the server. G2’s answer with Ziggs pressure and skirmish timing never turned into sustained tower damage; they ended with just 1 tower despite reaching 52.1k gold.

Closing Out

Once LYON were ahead, they closed with the discipline of a team that understood exactly how its composition should sound and feel in the final minutes. They did not need a Baron at all. Instead, they won the map through repeated fight control, stacking up 7 towers to 1 and finishing the game in 30:40 with 61.2k gold against 52.1k.

The cleanest closing image came from the two biggest carries. Saint remained deathless at 6/0/5, while Dhokla’s 5/1/7 on Gnar made every flank feel dangerous before it even began. On the bottom side, Berserker’s Varus only posted 1/1/6, but he did exactly what LYON needed: scale, stay alive, and fire through the frontline once the engage landed.

For G2, the damage was not just the loss but the shape of it. They kept enough objective control to suggest a possible comeback route, yet every crucial fight tilted the other way. In a series where they needed calm, LYON brought the clearer voice.

Polymarket Market

From a retrospective view, the market did not read this game correctly. Pricing LYON at just 36% underestimated how often a small draft edge can look much larger when execution is clean. The public leaned toward G2’s brand strength and perceived mid-jungle punch, but the match showed LYON’s actual advantages: a steadier top matchup, better first access on engage, and superior scaling around Viktor and Varus. In that sense, the live draft model at 50% was much closer to reality than the market was. At series level, the number staying at 72% after the win suggests expectations were already heavily tilted toward LYON, which means Game 3 becomes less about surprise and more about whether G2 can break the script.

Match Stats

FAQ

*Odds via Polymarket, 2026-07-10 09:46 UTC.*