VALORANT Champions Shanghai: T1 and 100 Thieves Advance on Opposite Scripts

On the night of October 9 in Shanghai, on the map Summit, T1 entered the second half trailing 3-9. Paper Rex had controlled the first half almost completely. But when I opened the individual stat sheet after the match, one line of data made me pause longer than the scoreline did: Ham "iZu" Woo-Joo was the only T1 player with a positive KDA on the deciding map, posting a 1.15 rating and 290 ACS. A team that won a match while only one individual played above average. After six years of following sports through the lens of data, I have learned that wins like this usually tell a different story than the scoreboard suggests. T1 won, but the way they won leaves more to discuss than it resolves.

Before going into detail, I need to state my method. This analysis draws on the result sheets and individual data published from the VALORANT Champions Shanghai playoff round, along with the bracket schedule. Two BO3 series are examined: T1 versus Paper Rex, and 100 Thieves versus Nongshim RedForce. Both ended 2-1, meaning both went to a deciding map. I have no data on the competitive patch version, no full roster information, and no agent pick rates. Any conclusion about the meta is therefore speculative, and I will flag that limit at each point. The five maps that appeared across these two series were Lotus, Ascent, Split, Sunset, and Summit.

The first notable data point sits on Summit, where T1 came back from a 3-9 deficit to win 13-11. Paper Rex did not merely lead at halftime; they also won the second-half pistol. Losing a map after leading 9-3 and taking the second-half pistol is the kind of defeat that rarely comes from pure skill. It usually reflects a problem in mid-round adaptation or a loss of composure at the decisive moment. As an observer, I have seen similar collapses in football, when a team dominates the first half and then falls apart in the final twenty minutes. That pattern repeats often enough to become a signal rather than an accident.

The second data point lies in the player crowned MVP. iZu played the Initiator role, and 290 ACS is a fairly high mark for that role, suggesting an aggressive, entry-supporting style rather than a passive one. But the more telling line is "the only player with a positive KDA." Read in reverse, it means T1's other players were all individually negative on the deciding map. T1 won on a single point of support. This is a single-carry dependency profile, and it differs in nature from a team that wins through its system.

In the other series, the picture was nearly the opposite. 100 Thieves beat Nongshim RedForce 2-1, with the deciding map Ascent ending 13-1. Matthew "Cryocells" Panganiban, a Sentinel player, took MVP with a 1.41 rating and 253 ACS. A Sentinel reaching such a high ACS usually reflects a map-control and anchoring style rather than one that rushes into duels. The way 100 Thieves closed the series at 13-1 shows they shut the match down through structure, not through a moment of individual brilliance. Both teams advanced, but one lives on an individual's moment, the other on the discipline of a system. That difference will determine who lasts longer in the next round.

Ascent appeared in both series with two completely opposite results: 13-6 for Paper Rex and 13-1 for 100 Thieves. A single map producing two such divergent scorelines suggests it depends heavily on execution rather than on the strength of a fixed strategy. This observation carries medium confidence, because two data points are not enough to conclude. But it is enough for me to place Ascent in the high-variance group, where a mistake in map veto can cost more than a mistake in a duel.

The 2026 pandemic season was the period when I learned to read data most seriously. When tournaments were suspended, I spent three months at home collecting figures from 380 matches of a football season and calculating pressing metrics. I realized that a strong correlation never means causation. That lesson applies directly to the regional picture in Shanghai. Four of the six remaining teams come from the Americas: 100 Thieves, G2, NRG, and LOUD. Only T1 represents the Pacific, and only Team Vitality represents EMEA. Both are the fourth seeds of their regions.

The easy conclusion for fans is that the Americas are dominating. But the data does not allow going that far. The survival of fourth seeds while higher seeds fell can come from two sources: the variance of the BO3 format, or genuine underperformance by the regions' top teams. I cannot separate these two possibilities with the available data. A region reduced to its fourth seed usually implies the top three seeds were eliminated earlier, but that may reflect a brutal bracket draw rather than real decline. The Americas' concentration could be a meta advantage, or simply luck in the bracket draw. At low confidence, I choose not to assert either side.

The double-elimination format complicates the picture further. G2 faces Team Vitality and NRG faces LOUD in the upper bracket on October 10, with the winners reaching the upper-bracket final. The losers of those two series drop into the lower bracket the very next day, October 11. T1 and 100 Thieves are already in the lower bracket, meaning they face exactly those teams that just dropped down. This is a familiar friction point of lower-bracket scheduling: the upper-bracket loser gets no recovery time, and the waiting team gets no time to study a new opponent. Both sides have their preparation compressed into a single night.

For T1, the risk is structural. They live on a single point of support, and in the lower bracket one loss ends everything. iZu cannot carry every deciding map. For 100 Thieves, the risk lies in a sharp jump in opponent quality: they will meet the loser of the G2 - Vitality pair, both stronger than Nongshim RedForce. And Paper Rex left a warning lesson for every remaining team: leading 9-3 is not insurance. Every sheet of numbers is a cut, and every cut is a story.

From Busan, where I follow matches on screen at dawn, there is one thing I always remind myself: a player's value is only an equation missing its unknowns. The same rating figure can carry entirely different meanings depending on whether it was produced in a favorable game state or during a comeback. iZu posted a 1.15 rating in a match where his teammates all played below average; Cryocells posted a 1.41 rating in a match where the whole team played above average. Two figures identical in unit of measure, but different in nature.

Looking toward October 11, the signal I am watching is not the scoreline but how these two teams handle the first half. If T1 again lets its opponent take the lead and has to lean on one individual, the dependency profile will become clearer. If 100 Thieves keeps shutting down deciding maps through structure, they are building a credible profile for a deep run. The lower bracket does not reward the strongest team; it rewards the team that makes the fewest mistakes on the decisive night. And as I always say, the abacus never sleeps, but the game does.