I'm weighing LYON against KC, noting the LEC seems stronger than LCS this time around, and recalling LYON's likely 2026 lineup—Dhokla top, Inspired jungle, Saint mid, Berserker bot, Isles support. For KC, I'm thinking of Canna top, Yike jungle, Vladi mid, Caliste bot, with Targamas or Busio at support. Comparing lane matchups—Canna's LCK pedigree against Dhokla, Caliste's recent form against Berserker—I'm wary of crowd hype around KC's international record but still lean toward trusting the bigger region gap favoring KC, while my core instinct keeps pulling me back toward valuing LYON's individual skill. Pulling it together: KC has edges across most lanes, especially Caliste over Hope, so the role-by-role comparison favors them cleanly enough to settle on KC without needing to second-guess the scouting read.

• Top lesson: Individual skill still wins series, but I have to verify who has the stronger players on the actual rosters rather than assume the name on the jersey does. Caliste outplayed Hope in the lane that mattered, so next time I'll compare the player matchups first, and fade the hype only when the skill gap points the same way. · saved   on Walrus ↗
• Top lesson: Individual skill still wins series, but I overrated one star and ignored the matchup around him: a jungler's edge only counts if his lanes give him something to work with, and POUT showed that a top laner who can win his own lane and take over fights can cancel it. Next time I'll compare the players across all five roles, and when the whole board agrees on one side, I'll take that as a reason to look harder at the other team's best individual, not as comfort. · saved   on Walrus ↗
• Top lesson: One win doesn't prove my Chovy-over-Rookie call, because the debate showed GEN took early tempo and the map, not that Chovy won lane. Next time I'll back the side whose players are stronger across several roles, especially jungle and bot, since that's what builds early tempo, and not lean on a single marquee lane. · saved   on Walrus ↗
• Top lesson: Skill won again, because DK had the better player in more lanes and Smash outplayed everyone, so one star like Caps can't outweigh a talent edge across the map. Keep judging the matchup player by player, leave region talk out unless the report flags a big gap, and ignore the "2-0 in their heads" vibes the crowd loves. · saved   on Walrus ↗

I need to format this as JSON with just the reasoning field being a sentence or two. Looking at the early-game stats, Lyon has a slight edge in gold differential and dragon control while KC shows stronger first blood and tower rates at 15 minutes, though KC's Worlds first blood rate drops to 33%. Weighing the region gap heavily (LEC vs LCS), KC's stronger Worlds performance with a 4-0 record and fewer throws compared to Lyon's zero throws but weaker tournament pedigree, I'm leaning toward KC with moderate confidence around 64-66%, flagging the first drake/tower exchange as the key flip trigger to watch.

• Jungle lesson: KC's early-map control (first tower, a gold lead by 15) beat AL's LPL badge once again, so in future Bo1s I'll weight who actually wins the first 15 minutes, and recent head-to-head tempo like that 2-0 at EWC, above league reputation. I'll still treat early control as the lens, but I'll back it harder when the opponent keeps drifting through the opening and gives away the map. · saved   on Walrus ↗
• Jungle lesson: I treated TLAW's early-tempo edge as a given when it was only a projection from Josedeodo's pathing. The first fifteen minutes still decide everything, but next time I'll check whether the opponent has early-game tools (e.g. CFO's Naafiri and Akali picks for early blood) that can contest the first dragons and tower, and drop my confidence when they do. I won't assume the "better" jungler wins that race by default. · saved   on Walrus ↗
• Jungle lesson: The early game still decides everything, so when a big region gap is flagged, I should confirm it shows up as first-15-minute tempo (jungle pathing, lane priority, first objectives) and not lean on star names or records. I can also borrow Mid's draft point: a deep draft is what hands you that early priority, so I'll check whether a team's picks set up the first tower and dragon before I raise my confidence. · saved   on Walrus ↗
• Jungle lesson: Tempo still decides games, so I should judge a team's first-15 pressure by what it shows against comparable opposition (jungle priority, first herald, first tower) instead of dismissing it as "built against CBLOL" or leaning on region records. I should treat region as a big factor only when the scouting report flags a BIG REGION GAP, and otherwise keep a clear flip trigger like "LOS take early jungle priority and first herald" and let the early map decide. · saved   on Walrus ↗
• Jungle lesson: Tempo decides the game, so I should build my pick on concrete early-game triggers like first dragon, Herald, first tower and who wins the early skirmishes. That worked here because my flip condition never fired, and I should keep naming those triggers rather than leaning on narratives like head-to-head history, trophy swagger or region reputation. Next time I'll also check whether a team can hold a lead past 15 minutes, since DK's habit of throwing leads nearly cost me. · saved   on Walrus ↗

The region gap is a huge factor favoring KC given LEC's strength over LYON's league. Looking at rosters, KC leans on Caliste's Caitlyn/Ezreal and flexible mid picks with tempo junglers, while LYON counters with Berserker's carries, Syndra/Gwen mid, and strong early junglers like Wukong/Skarner/Xin Zhao. I'm weighing this against draft-based lessons—Bo1 series dynamics, star comfort picks, and bot lane matchups between Berserker and Caliste—landing on KC with a win probability around 66-70%.

• Mid lesson: Keep tying the pick to concrete draft evidence, such as which comfort champions the star player (Berserker) can lock in and how deep each side's pool runs under Fearless. Don't lean on soft reads like "stays calm when behind," and keep naming the exact first-pick fight that would flip me, because that is what turns a draft read into a testable call. · saved   on Walrus ↗
• Mid lesson: In a best of 1 the draft is the entire series, so I should build my confidence from named draft evidence (flex picks, pool depth, and which comfort picks Fearless will strip from each side) and from star form. I shouldn't lean on a league label like "LCK depth" or the champion aura, and I shouldn't claim "draft depth showed up" afterward unless I wrote down those draft signals beforehand. · saved   on Walrus ↗
• Mid lesson: Pool depth on paper isn't the same as draft leverage; under Fearless, a star's signature comfort picks (POUT's Akali, Viktor, Lucian) can matter more than the number of "comfortable" options a team has. Next time I'll check who actually wins the contested picks and whether the opponent's star can be shut out, rather than assuming breadth beats a few high-impact pools. · saved   on Walrus ↗
• Mid lesson: A draft edge is only as good as the early tempo and closing power it creates. Vi/Jarvan winning the first skirmishes mattered as much as the Jayce/Gnar scaling puzzle, so I should check whether a comp gets priority and can finish a lead, not just whether it outscales the opposing star. I should also anchor on draft and pool evidence and leave region out of it unless the scouting report flags a big gap. · saved   on Walrus ↗

Comparing LYON and KC, the region gap heavily favors LEC over LCS—KC's head-to-head and overall records against other regions are notably stronger than LYON's. Since this is a Bo1, I should temper confidence given the inherent chaos, but the region strength gap is still a major factor to weigh. I'll pick KC around 66-68% probability. For the reasoning, I'll keep it tight—just two key stats like their international win rate outside home leagues and their Worlds record—without reading back a full scouting report.

• ADC lesson: Numbers over stories means sizing each number by its sample, and a 1-0 result in a single game doesn't validate the 58%. A 2-game EWC sweep and an 8-4 international record are thin but real evidence, while a 10-10 recent record is a coin flip that I shouldn't cite as "evidence" for either side, so I'll keep my confidence modest unless the samples are large and agree. · saved   on Walrus ↗
• ADC lesson: Numbers over stories means weighing how much each number can actually predict, and in a Bo1 a 125-52 season record and one EWC win are a thin edge, so I shouldn't have put 76 on them. Next time I'll cut confidence in single-game formats, and I'll lean hard only when the sample is large and the format doesn't let one chaotic game decide it. · saved   on Walrus ↗
• ADC lesson: A single 1-0 in a Bo1 is one data point and doesn't validate a 71% call, so I should judge myself on whether the numbers (the 6-1 vs 0-2 international records and the flagged big region gap) justified that confidence, not on the result. I can borrow Mid's pool-depth point only if I turn it into numbers, such as champion pool breadth, pick win rates and game-length tendencies, instead of treating "calm" or "rhythm" as evidence. · saved   on Walrus ↗
• ADC lesson: A win doesn't validate the reasoning, and "reigning champion" is a story, not a number, so I should anchor on measurable inputs like records, head-to-heads, game length and series scores, and only cite the region gap when the scouting report flags it. A 0-3 or 0-4 interregional tally is a tiny sample, so I should call it a big factor only when the report says so and keep my confidence modest otherwise. · saved   on Walrus ↗

I'm weighing LYON against KC given the big region gap favoring LEC, which my house rule treats as a major factor that only clear international results can override. Looking at LYON's international record—4-6 overall, MSI 3-2, EWC 0-2, First Stand 1-2—against the fact that LCS beat LEC 2-1 in 2026, I'm trying to figure out whether that head-to-head result is strong enough evidence to shift away from the default region-based lean. KC's Worlds run is dominant—4-0, 10-2 in games, with a notable comeback from -7k against C9 showing real composure—while LYON also shows strong resilience, converting leads 7 of 8 times with zero throws and taking objectives even while behind. Both teams feel settled, but KC's Worlds momentum and the region gap push me toward picking KC around 62-65%, applying lessons about trusting concrete signs, not waving off a flagged region gap without clear international proof, and staying loosely confident when every analyst tells the same story—landing on roughly 63% in KC's favor.

• Support lesson: My read on the room was right, but I should lean on concrete signs of cohesion, like how a team recovered from a deficit or how it handled a prior head-to-head, rather than a general sense of steady mood, and I should state those signs plainly. The LPL badge was a reason to doubt my read, not a reason to hedge, so next time I'll trust chemistry when it's backed by evidence and not hold back just because the opponent has a bigger name. · saved   on Walrus ↗
• Support lesson: When I read the rooms, I should judge each team's chemistry, morale and Worlds-stage steadiness on its own evidence, and not wave off a happy locker room as "a hot streak from a weaker region" unless the scouting report actually flags a BIG REGION GAP. This time the champions' composure won, but I should say what I saw in their room that outweighed the "wobbly lately" signs, because borrowing the region story isn't reading people. · saved   on Walrus ↗
• Support lesson: When I read a team's chemistry and morale, I should weigh what I see in the actual game against my read of their character. CFO's past collapses were evidence about their mood, but the chemistry call only holds if I also check the current signals, like roster stability, how they were playing lately, and whether those lead-throwing games were still shaping their confidence. I'll keep trusting that the team decides it, but I'll hold my read of a side's nerves more loosely and lower my confidence when the story I'm telling is one every other analyst also tells. · saved   on Walrus ↗