G2 Decided to Replace babybay Before Champions Shanghai — His Career-Best Showing Now Exposes Management's Evaluation Logic

The gap between what the data said after that Shanghai playoff run and what G2 Esports had already decided internally is now the loudest argument in the VALORANT roster market. G2 reached the top four at VALORANT Champions Shanghai, beating 100 Thieves and T1 along the way. Andrej 'babybay' Francisty was the team's second-highest rated player at that event. Yet the decision to move on from him for the 2027 season was reportedly made before Shanghai began, with Eduardo 'Sato' Nagahama arriving from Leviatán as the replacement. The story gets complicated right there, because a decision that was not made on performance is now being defended with performance language.

That roster slot has changed hands three times in roughly two years. babybay retired at the end of 2026, returned in August 2026 in place of Nathan 'leaf' Orf, then was benched when leaf came back ahead of Champions Paris, then reinstated for 2026 after Jonah 'JonahP' Pulice left for Sentinels. The results: a VCT Americas Stage 1 title, Masters Santiago qualification, and a Champions Shanghai semifinal. Three handovers at one position is not a story of three consecutive individual failures. It is a story of an unresolved role definition. Who plays entry, who flexes, who owns which agent pool in a five-man economy — if that answer is not fixed inside the roster, stability will not arrive no matter who occupies the slot.

G2's decision rests more on structural logic than on performance, and that is the only strong line of defence available to them. In a franchised league there is no relegation. No owner loses a slot, which lets clubs plan a full season ahead. In that extra time, role conflict, agent-pool overlap and slot-budget arithmetic outweigh a player's recent form. Deciding before Shanghai means management already believed something in the current structure did not fit. The Shanghai run did not invalidate that belief — it proved the problem was never at the player's skill level. That is not good news for a club, because it means the team knows where the crack is internally but cannot explain it externally.

One number deserves caution. 'Second-highest rated' is a single-event, all-agent aggregate. VALORANT rating is role-sensitive: duelists and entry players accrue it differently from sentinels and controllers, and opponent strength changes the number. Using it as a standalone justification to keep or cut someone is methodologically unsafe either way. Second-highest rating on a semifinal run means the player was in form. But form is a season's asset, and roster construction is a year-long decision.

Sato arrives from Leviatán, meaning an intra-Americas move with a closer language and comms model, cheaper to adapt than a cross-region import. But removing a player with demonstrated fit in your system for one with no demonstrated fit in it makes the risk one-directional. In a franchised league the cost of a mis-signed player is a full season, not a transfer fee.

Where I could be wrong: if Sato's role actually overlaps leaf's rather than babybay's, then G2 enters 2027 with a fresh internal role conflict rather than an upgrade. In that case the decision was not wrong, but it was not a solution either. Second possibility: if the Shanghai showing was a one-off form spike, the peak of a one-year comeback curve, then management's advance evaluation will prove correct. Neither possibility is testable yet, and that is the real weakness of this story.

babybay is reportedly staying in China after Shanghai to weigh VCT CN offers. At his career stage that is a rational choice, because a starting spot on a developing regional roster gives far more match time than an Americas bench. The trade-off is optionality: one season in China effectively removes him from the 2027 Americas conversation. And for a VCT CN club the binding constraint is not salary but the scarcity of a usable import slot. If the slot is open, this deal happens; if not, it does not. VALORANT's meta resets on an Act cadence, multiple times per season, so any 2027 roster decision made in the 2026 offseason carries unusually high obsolescence risk. If babybay lands in China, the question becomes whether he is getting a quarterfinal team or a structural experiment whose result will never be written on paper. The Shanghai scoreboard said he was in form. The decision was made before it. In 2027 the server will tell us which paper was right.