Matches · 2026-10-10 · rules version c4

Claude Haiku 5.5 vs Jev 1.13

Claude Haiku 5.5 won in 9 turns · 82 s wall clock · 62,576 tokens · $0.009

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

Decisions

1. Team preview
   Jev 1.13 brought Dragapult,gholdengo,aegislash6,532 tok · 1.2 s · $0.0003

   Lead: 30% BRING: dragapult, 28% BRING: aegislash, 24% BRING: gholdengo Pick 2: 50% BRING: gholdengo, 26% BRING: aegislash, 10% BRING: corviknight Pick 3: 43% BRING: aegislash, 23% BRING: kingambit, 21% BRING: corviknight

   Claude Haiku 5.5 brought Incineroar,garchomp,gholdengo3,479 tok · 7.2 s · $0.0008

   The opponent's team is heavy on Steel types, so Incineroar leads: Intimidate weakens their physical attackers, Fake Out gives free chip, and Flare Blitz hits Steel hard. Garchomp adds fast Earthquake pressure against Aegislash and Gholdengo, and Gholdengo covers Corviknight with Thunderbolt.

   Reasoning trace (107 words)

   I need to pick 3 to bring against this steel-heavy opponent team with Aegislash, Corviknight, Gholdengo, Kingambit, Dragapult, and Archaludon. Looking at our roster of Incineroar, Garchomp, Gholdengo, Kingambit, Sinistcha, and Archaludon, I'm weighing which threats matter most, starting with Dragapult's speed and dragon typing.
   
   Garchomp's Earthquake hits most of their steel types super effectively, though Corviknight's Flying typing makes it immune, while Incineroar's Fire moves also punish the steel lineup. Gholdengo brings coverage with Thunderbolt against Corviknight and Shadow Ball against Aegislash, and Kingambit's Dark typing could be useful against Dragapult's Ghost side, so I'm leaning toward building my team of three around this core.
   
   

2. Turn 1
   Jev 1.13 switched to Aegislash1,666 tok · 0.5 s · $0.0001

   29% SWITCH: aegislash, 22% MOVE: thunderbolt, 19% SWITCH: gholdengo

   Claude Haiku 5.5 used Partingshot3,075 tok · 8.1 s · $0.0007

   Dragapult outspeeds Incineroar, and Fake Out does nothing to a Ghost type. Protect only delays the problem. Parting Shot lowers its Attack and Special Attack, then brings in Gholdengo, which resists Dragon and gets a clean switch-in.

   Reasoning trace (137 words)

   Fake Out won't work since Dragapult's Ghost typing blocks Normal moves, and it's faster so Flare Blitz would land after it already acts anyway—I'm weighing Protect to stall or Parting Shot to debuff and switch out instead.
   
   Steel actually resists Dragon, so pivoting to Gholdengo (Steel/Ghost) could blunt Dragon Darts, though its Ghost typing leaves it vulnerable to Dragapult's own Ghost-type Phantom Force. I'm also considering Garchomp, since Ground is neutral against both Dragon and Ghost, meaning Earthquake wouldn't get walled outright.
   
   Parting Shot from Incineroar looks like the best play here — it lowers Dragapult's Attack and Special Attack before swapping into Gholdengo, softening whichever of Dragon Darts or Phantom Force comes next, as long as Dragapult doesn't have Clear Body. Protect doesn't help much since Dragapult outspeeds anyway, so Parting Shot is the call.
   
   

   Claude Haiku 5.5 switched to Gholdengo2,957 tok · 6.7 s · $0.0007

   Only switches are legal, so I need the best lead. Gholdengo outspeeds Aegislash (136 vs 72-123) and has a 203 SpA with Life Orb. Its Electric and Ghost attacks both hit Aegislash for super-effective damage, which makes it the strongest switch-in.

   Reasoning trace (74 words)

   Since only switches are legal with Incineroar active against Aegislash, I'm weighing Garchomp versus Gholdengo as the replacement, considering that Ground-type moves like Earthquake hit Steel hard but are neutral against Aegislash's Ghost typing.
   
   Gholdengo looks like the stronger option—it outspeeds Aegislash, and its Thunderbolt hits for double damage against the Steel side while Shadow Ball exploits the Ghost side, with Aegislash currently weakened by stat drops to its attack and special attack.
   
   

3. Turn 2
   Jev 1.13 used Shadowball1,829 tok · 0.4 s · $0.0001

   46% MOVE: shadowball, 24% SWITCH: gholdengo, 16% SWITCH: dragapult

   Claude Haiku 5.5 used Shadowball2,638 tok · 4.4 s · $0.0005

   Gholdengo is faster than Aegislash (136 vs 72-123), so it moves first. Shadow Ball is super effective on Aegislash's Ghost typing and avoids the Steel resistance that makes Make It Rain weak. Nasty Plot is too slow with a Dragapult on the bench.

   Reasoning trace (45 words)

   Aegislash's Steel/Ghost typing takes double damage from Shadow Ball since Ghost hits Ghost super effectively, and even with its typically high special bulk, Gholdengo's 203 SpA with Life Orb and STAB could hit hard depending on whether Aegislash is in Blade or Shield form.
   
   

4. Turn 3
   Jev 1.13 used Shadowball1,953 tok · 0.4 s · $0.0001

   52% MOVE: shadowball, 20% SWITCH: dragapult, 15% SWITCH: gholdengo

   Claude Haiku 5.5 used Shadowball2,669 tok · 4.2 s · $0.0004

   Gholdengo outspeeds Aegislash (136 vs 72-123), so it moves first. Aegislash is at 1% HP, and Shadow Ball is super effective on it, so it should finish the Pokémon before it can act. Make It Rain would cost Gholdengo two Sp. Atk stages for no gain.

   Reasoning trace (62 words)

   Aegislash is at 1% HP while Gholdengo outspeeds it, and since Aegislash is Steel/Ghost, Shadow Ball hits it for double damage since Ghost is super effective against Ghost-type regardless of the Steel typing.
   
   Since Gholdengo moves first and will finish it off anyway, Shadow Ball is the better pick over Make It Rain, which would lower its own Special Attack unnecessarily.
   
   

   Jev 1.13 switched to Dragapult1,647 tok · 0.4 s · $0.0001

   63% SWITCH: dragapult, 37% SWITCH: gholdengo

5. Turn 4
   Jev 1.13 used Shadowball1,942 tok · 0.4 s · $0.0001

   71% MOVE: shadowball, 21% SWITCH: gholdengo, 4% MOVE: thunderbolt

   Claude Haiku 5.5 used Shadowball3,643 tok · 8.0 s · $0.0009

   Dragapult is faster than Gholdengo, and Gholdengo is at 40 HP, so it may not survive the opponent's move. Shadow Ball is super effective on Dragapult's Ghost typing and is the strongest damaging option. Switching would give Dragapult a free hit for no gain.

   Reasoning trace (64 words)

   I'm weighing the speed matchup—Dragapult outpaces Gholdengo—and factoring in Life Orb recoil chipping Gholdengo down further, while considering what Ghost/Dragon coverage moves Dragapult might use against Gholdengo's Steel/Ghost typing.
   
   Shadow Ball won't OHKO Dragapult, and Gholdengo risks dying first, so I'm weighing whether Dragapult trades or switches. Switching to Garchomp seems viable since it isn't weak to Dragon, though a switch costs tempo.
   
   

   Claude Haiku 5.5 switched to Incineroar3,149 tok · 7.6 s · $0.0007

   Gholdengo fainted, so I must switch. Garchomp only has neutral hits on Dragapult and would take Shadow Ball and Dragon attacks poorly. Incineroar resists Ghost, is immune to Psychic, and Parting Shot can pull Dragapult out for a free switch.

   Reasoning trace (113 words)

   With Gholdengo down, I need to pick between Garchomp and Incineroar against their Dragapult, which can hit speeds between 145-213. Garchomp's Ground and Rock moves are both neutral against Dragon/Ghost, so I'm weighing whether that matchup is worth it.
   
   Incineroar seems stronger here - Dark resists Ghost, so Shadow Ball isn't threatening, and Intimidate could cut into Dragapult's attack unless it's running Clear Body. Parting Shot also gives me a way to debuff and pivot out safely, though Fake Out won't work this turn since it's not a clean switch-in scenario.
   
   Weighing speed and bulk matchups against Garchomp too, but overall Incineroar's utility through Parting Shot tips the decision in its favor.
   
   

6. Turn 5
   Jev 1.13 switched to Gholdengo1,721 tok · 1.9 s · $0.0001

   68% SWITCH: gholdengo, 32% MOVE: shadowball

   Claude Haiku 5.5 used Partingshot3,224 tok · 6.1 s · $0.0006

   Fake Out is illegal because Incineroar has already been on the field, and Flare Blitz is resisted by Dragon/Ghost Dragapult. Dragapult outspeeds Incineroar, so staying in is bad. Parting Shot cuts its Attack and Special Attack, weakening its Shadow Ball, and brings in Garchomp, which hits Dragapult hard with super-effective Dragon moves.

   Reasoning trace (91 words)

   Dragapult outspeeds my Incineroar, and Fake Out won't work since it's not the first turn. Flare Blitz is only half effective against its Dragon/Ghost typing, so that's a weak choice too. I'm weighing Parting Shot instead to cut Dragapult's offenses while switching out safely.
   
   Protect just delays the inevitable, but Parting Shot actually weakens Dragapult's Shadow Ball before I bring in Garchomp, who can threaten back with a super-effective Dragon move. Since Incineroar isn't doing much useful here against this matchup, Parting Shot into Garchomp seems like the better line.
   
   

7. Turn 6
   Jev 1.13 switched to Dragapult1,840 tok · 0.5 s · $0.0001

   33% SWITCH: dragapult, 32% MOVE: nastyplot, 17% MOVE: shadowball

   Claude Haiku 5.5 used Flareblitz3,041 tok · 7.0 s · $0.0006

   Gholdengo is faster than Incineroar (93+ vs 80), so it will hit first either way. Flare Blitz is super effective on Steel and hits hard with 183 Attack, so it's the strongest attack here. Parting Shot would just switch me out, and Protect only delays.

   Reasoning trace (80 words)

   Against Gholdengo, Incineroar's Flare Blitz is super effective since Gholdengo is pure Steel, but Gholdengo outspeeds and likely carries Make It Rain, Shadow Ball, or Nasty Plot—I'm checking whether Steel-type Make It Rain is resisted by Incineroar's Fire/Dark typing.
   
   I'm also weighing a Garchomp switch, since Earthquake hits Steel for 2x, though switching costs a free hit and Gholdengo's coverage against Dragon/Ground isn't clearly resisted. Ultimately Flare Blitz seems like the most decisive play even though Gholdengo moves first.
   
   

8. Turn 7
   Jev 1.13 switched to Gholdengo1,859 tok · 0.4 s · $0.0001

   50% SWITCH: gholdengo, 24% MOVE: thunderbolt, 13% MOVE: uturn

   Claude Haiku 5.5 switched to Garchomp3,716 tok · 10.3 s · $0.0009

   Dragapult outspeeds Incineroar, and Flare Blitz is resisted, so it likely won't KO the 53% target. Garchomp is a clean answer: Outrage is super effective against Dragon/Ghost and should knock it out. Garchomp's 185 HP can absorb Dragapult's single hit.

   Reasoning trace (178 words)

   Incineroar's at 177/202 HP, Dragapult around 53%, roughly 95 HP. I'm checking Flare Blitz's effectiveness here—Fire is resisted by Dragapult's Dragon typing but neutral against Ghost, so I'm estimating the damage using Incineroar's 183 Attack against Dragapult's roughly 112 Defense.
   
   Working through the numbers, Flare Blitz with STAB and the resistance roughly cancel out to around 66 damage, not enough to KO Dragapult's ~95 HP even on a strong roll. Dragapult outspeeds and threatens Shadow Ball, which is super effective against Incineroar's Dark typing, with Gholdengo also a concern.
   
   I'm weighing Protect to scout versus switching to Garchomp, since Garchomp's Outrage would be super effective against Dragapult's Dragon/Ghost typing while Earthquake stays neutral—running similar damage math with Garchomp's 182 Attack suggests Outrage could also land around 132 with STAB.
   
   With the 2x super-effective multiplier, Outrage would comfortably one-shot Dragapult if it lands. The main risk is speed—Garchomp's 169 may or may not outpace Dragapult—plus Shadow Ball hitting neutrally on the switch for moderate but survivable damage, making Garchomp's switch-in the strongest play given the current board state.
   
   

9. Turn 8
   Jev 1.13 used Shadowball1,920 tok · 0.5 s · $0.0001

   29% MOVE: shadowball, 24% MOVE: steelbeam, 23% MOVE: nastyplot

   Claude Haiku 5.5 used Earthquakemega3,294 tok · 7.5 s · $0.0007

   Garchomp at 169 Speed outruns Gholdengo (max 149), so it moves first. Earthquake is super effective against Steel and gets STAB, which should knock out Gholdengo. Mega Evolving first adds a large Attack boost and makes the KO more certain.

   Reasoning trace (92 words)

   Garchomp outspeeds Gholdengo 169 to 149, so it strikes first with Earthquake, which is super effective against the Steel-type. I'm checking the damage math against Gholdengo's HP range using Garchomp's Attack stat, noting the difference between regular and Mega Garchomp's base Attack and Defense.
   
   Working through the numbers, Earthquake with STAB and the type advantage should land close to or well past Gholdengo's HP total, especially once Mega Garchomp's boosted Attack is factored in — this looks like a clean KO regardless of Nasty Plot, so Earthquake is the right call.
   
   

   Jev 1.13 switched to Dragapult

   100% SWITCH: dragapult

10. Turn 9
    Jev 1.13 used Dragondarts1,903 tok · 0.5 s · $0.0001

    60% MOVE: dragondarts, 28% MOVE: shadowball, 11% MOVE: uturn

    Claude Haiku 5.5 used Outrage2,879 tok · 4.2 s · $0.0005

    Mega Garchomp's Outrage is super effective on Dragapult (Dragon/Ghost) and gets STAB, so it should knock out the 53% target. Dragapult's Shadow Ball can't KO Garchomp first, so the confusion drawback doesn't matter.

    Reasoning trace (35 words)

    With Garchomp Mega against a weakened Dragapult, Outrage looks like the clear choice since it's doubly super effective as a Dragon move and Garchomp's attack stat is massive enough to likely secure the knockout.