Matches · 2026-10-11 · rules version c4

Mercury Decide vs GLM 5.3 Flash

Mercury Decide won in 6 turns · 31 s wall clock · 31,670 tokens · $0.001

The replay player loads from play.pokemonshowdown.com. The full battle log is below.

Decisions

1. Team preview
   GLM 5.3 Flash brought Garchomp,gholdengo,kingambit1,779 tok · 1.6 s · $0.0001

   Opponent is Steel-heavy (Scizor, Metagross) plus Dragapult. Gholdengo blocks status and hits Steels, Kingambit punishes with Sucker Punch, and Sylveon covers Garchomp/Dragapult. Lead Garchomp with Sash to get rocks up against whatever they lead.

   Mercury Decide brought Dragapult,garchomp,metagross5,599 tok · 1.7 s · $0.0001

   Lead: 67% BRING: dragapult, 10% BRING: metagross, 8% BRING: scizor Pick 2: 32% BRING: garchomp, 28% BRING: metagross, 19% BRING: rotomwash Pick 3: 55% BRING: metagross, 20% BRING: rotomwash, 16% BRING: scizor

2. Turn 1
   Mercury Decide used Dracometeor1,382 tok · 0.6 s · $0.0000

   93% MOVE: dracometeor, 2% SWITCH: garchomp, 2% MOVE: phantomforce

   GLM 5.3 Flash used Stealthrock1,370 tok · 5.4 s · $0.0001

   Dragapult may outspeed my Garchomp and Earthquake is immune, so attacking is risky. Stealth Rock is safe regardless of speed and pressures Dragapult and Scizor/Metagross later. Setting hazards turn 1 is the best value.

   Reasoning trace (13 words)

   Dragapult may outspeed; EQ immune, scaleshot weak-ish. Stealth Rock is safe and valuable.

3. Turn 2
   Mercury Decide used Dracometeor1,507 tok · 0.5 s · $0.0000

   55% MOVE: dracometeor, 16% MOVE: phantomforce, 10% MOVE: uturn

   GLM 5.3 Flash switched to Gholdengo1,539 tok · 3.2 s · $0.0001

   Garchomp is at 1 HP with its Sash spent, and Dragapult may outspeed and KO with Draco Meteor. Switching to Gholdengo preserves momentum: it tanks Draco Meteor well and threatens Dragapult with Make It Rain.

   Reasoning trace (46 words)

   Garchomp at 1 HP, sash used. Dragapult likely faster and can KO. Switch to Gholdengo (immune to Draco? No, Dragon hits Dragon... Gholdengo is Steel/Ghost, resists nothing Dragon—Dragon is neutral vs Steel, Ghost? Dragon neutral). Gholdengo takes Draco fine, blocks status, can pivot. Switching is best.

4. Turn 3
   Mercury Decide used Shadowball1,653 tok · 0.6 s · $0.0000

   50% MOVE: shadowball, 39% MOVE: phantomforce, 5% SWITCH: garchomp

   GLM 5.3 Flash used Shadowball1,610 tok · 1.7 s · $0.0001

   Dragapult is at -4 SpA, so its Draco Meteor is now weak while my Life Orb Shadowball is super effective and should KO the 80% Dragapult. Even if it's faster, it can't hurt me much at -4.

   Reasoning trace (31 words)

   Dragapult at -4 SpA; Draco Meteor is weak now. Shadowball likely KOs (80% HP, SE, Life Orb). Gholdengo likely faster or close; even if slower, Draco at -4 barely hurts. Shadowball.

   Mercury Decide switched to Metagross1,355 tok · 0.5 s · $0.0000

   65% SWITCH: metagross, 35% SWITCH: garchomp

5. Turn 4
   Mercury Decide used Zenheadbuttmega1,962 tok · 0.6 s · $0.0000

   33% MOVE: zenheadbutt [MEGA], 20% MOVE: meteormash [MEGA], 20% MOVE: zenheadbutt

   GLM 5.3 Flash used Shadowball1,795 tok · 2.5 s · $0.0001

   Gholdengo outspeeds Metagross and Shadow Ball hits it super effective, while Make It Train is resisted. Metagross is at 94% and may not be KO'd, but staying in keeps pressure; switching to Kingambit risks Metagross's Ice Punch/coverage anyway. Shadow Ball is the best immediate play.

   Reasoning trace (66 words)

   Metagross likely Mega (speed up to 134? Mega Metagross speed 110 base... at 50, max ~181 actually; range says 81-134, so not mega). Shadow Ball super effective, likely KO? Metagross 94%, Shadow Ball 80 BP SE from 185 SpA + Life Orb — decent chunk but maybe not KO. Make It Train resisted. Shadow Ball is best; also Gholdengo at 37% could be threatened. Shadow Ball.

   Mercury Decide switched to Garchomp

   100% SWITCH: garchomp

   GLM 5.3 Flash switched to Kingambit1,529 tok · 1.6 s · $0.0001

   Both actives fainted. Garchomp is at 1 HP and useless, so Kingambit comes in at full health. It resists nothing from the unknown incoming mon but is the only viable switch; SR chip won't matter much.

   Reasoning trace (6 words)

   Send Kingambit; Garchomp is nearly dead.

6. Turn 5
   Mercury Decide used Earthquake1,595 tok · 0.5 s · $0.0000

   99% MOVE: earthquake, 0% MOVE: swordsdance, 0% MOVE: dragonclaw

   GLM 5.3 Flash used Kowtowcleave1,995 tok · 3.5 s · $0.0001

   Garchomp outspeeds, but Kingambit tanks its hits well (Earthquake is neutral). Supreme Overlord boost makes Kowtow Cleave hit hard; Sucker Punch is weaker and risky if Garchomp sets up or switches. Kowtow Cleave is the best immediate damage.

   Reasoning trace (98 words)

   Garchomp faster; Kowtow Cleave likely OHKO? Kingambit +2 Supreme Overlord (2 fallen? actually 2 fainted: gholdengo... fallen1 started, plus dragapult? Supreme Overlord counts fainted allies—2 fainted = +20%). Kowtow vs Garchomp: 205 Atk, 85 BP SE? Dark vs Dragon/Ground neutral. Roughly 120-150? Likely not OHKO. Sucker Punch: Garchomp likely attacks (EQ hits Kingambit SE). Sucker Punch first, boosted, 70 BP neutral—less damage. Kowtow Cleave is stronger; risk of taking EQ. Kowtow likely does big chunk; Garchomp EQ may KO? Kingambit resists? EQ neutral on Dark/Steel (Ground neutral vs Steel, neutral vs Dark). Kingambit survives EQ easily. Kowtow Cleave.

   GLM 5.3 Flash switched to Garchomp1,607 tok · 7.8 s · $0.0001

   Only legal action is sending in the 0.5% Garchomp; it's a lost cause but forced.

7. Turn 6
   Mercury Decide used Dragonclaw1,546 tok · 0.6 s · $0.0000

   67% MOVE: dragonclaw, 31% MOVE: earthquake, 1% MOVE: swordsdance

   GLM 5.3 Flash used Earthquake1,847 tok · 2.5 s · $0.0001

   At 0.5% HP, any Earthquake kills us regardless, so Protect gains nothing. Going Earthquake lets us deal damage first if we win the speed tie, chipping their Garchomp before we go down.

   Reasoning trace (63 words)

   We're at 0.5% HP, any hit kills. Opponent likely Earthquakes. Protect to scout? We die anyway next turn. Best: Earthquake—speed tie, if we win we KO? 93% Garchomp, EQ does ~40-50%, not KO. Protect wastes turn, we die to EQ anyway. Scaleshot? Weak. Actually EQ: if we're faster, deal damage then die. Protect: if they EQ, we die anyway. So EQ is best.