The Tamer Who Matters

Taming is one of SotA's most loved and most argued-about schools. The engine under it is genuinely strong: 199 tamable creatures across 14 families, per-family stat templates that scale from their owner's skills, a real focus commitment for fielding a pet, pet gear slots, and a full command set. But nearly all of a pet's power is passive, the handler's rotation is three buttons, and the balance targets we publish say plainly that the pet should not be most of the tamer. This document measures the school against the genre that invented it, and proposes twenty-eight changes built on one deal: less power that plays itself, more power in the handler's hands.

Executive summary

Ultima Online shipped the ancestor in 1997: Animal Taming with loyalty and feeding, Veterinary as its own craft, the dragon-tamer power problem, and then twenty years of retrofits ending in the 2017 pet revamp that priced pet power in training points and choices. WoW ran the same arc faster — happiness, family abilities, talent trees, Bestial Wrath's burst window, taming challenges. FFXI put pet abilities on the owner's hotbar. Guild Wars 2 made the pet a swappable second half of its ranger. City of Heroes built a whole archetype out of paying upkeep for henchmen you actively command. The genre's verdict is consistent: a pet class is interesting exactly in proportion to what the owner does per fight.

Measured against that bar, SotA's taming has a strong skeleton and an inverted power budget. The community's own headline thread says both halves out loud: pets are too strong, and taming is too low-interaction — "solo dungeons with a click every ~90 seconds." Our published balance targets already draw the line this page designs to: a tamer's solo clear rate at most 1.25x the median solo school, and the pet at most 60% of the tamer's total damage — the tamer must matter. The gaps cluster in three places:

1. Passive power. A grandmaster's pet reaches level ~114 because two bonuses are applied after the level cap instead of under it (an honest bug, and our headline fix); mastery multiplies stat lines linearly all the way to skill 200; a second innate layer stacks flat survivability on top; and the one "burst" skill has 100% uptime. None of it asks the tamer to press anything.
2. Interaction. Three owner actives (heal, frenzy, resurrect). No pet abilities the owner triggers, no stances with tradeoffs, no positioning commands, no capture gameplay, no bond. Fourteen families that mostly differ by model.
3. Accountability. A pet's kills pay its owner in full whether or not the owner does anything at all. Pet death costs a five-minute timer; collar recovery exceeds 100% at skill 50; an ignored pet fights exactly as well as a handled one.

The deal that runs through all twenty-eight proposals: Tier 1 removes a slice of the always-on power, in the open, with the math shown. Tier 2 sells roughly half of it back as active play — a pet ability bar, real stances, positioning commands, a revived Veterinarian, sidegrade gear in the two empty gear tiers that are already sitting in the data. Tier 3 adds the keystone: Bond, the system that routes the top of a pet's performance through actually handling it. A tamer who plays their pet lands inside the published bands and feels stronger than today in the moments that matter. A pet left to play itself simply isn't very good anymore.

Where we are today

The pieces in place are real: taming necklaces that carry a stored creature, a 10-second field tame with collars as the consumable, a summon gate by creature power, a pet status panel, a pet character sheet with three gear slots, persistent per-pet command preferences, chat commands (/pet, /petstate), and three owner actives. The pet fights from a behavior tree that follows, leashes, teleports to you when stranded, defends you when you're hit, and picks its own targets when set to aggressive.

• 1Three actives. Tame Heal, Tame Frenzy, Resurrect Creature. Everything else the pet ever does is automatic. The community thread's "a click every ~90 seconds" is not an exaggeration; it is the rotation. Proposals S10–S13.
• 2Aggressive is free. The strongest mode — the pet picks its own targets forever — costs nothing and trades nothing. Proposals S11, S24.
• 3The power is invisible. The pet's real strength is assembled from a stat pipeline (below) the game never explains, and one stage of it is plainly bugged. Proposals S1–S4.
• 4Dormant depth everywhere. Two taming skills exist in the data but can't be bought (Animal Handling, Veterinarian); four pet stats are declared and never used (Excitement, Enrage, Growth, Reduction); two whole pet gear tiers are empty folders; the injured-pet retreat behavior is wired but inert; Taunt Training only functions for ten of fourteen families. Proposals S9, S14, S16, S18, S21, S22, S24.
• 5The floor is cracked. Fourteen of the forty-six open taming and pet reports are pathing: pets stuck in barns, warping onto furniture, refusing elevators, escaping to the neighbor's lot. Command-driven play needs pets that reliably arrive. Proposal S23.

How a pet's power is actually assembled

Every tamed creature is rebuilt from a per-species template whose stats scale from its owner. The pipeline runs in four stages, and understanding it is understanding every Tier 1 card:

flowchart LR
  A["Species template\n178 tamed prefabs\nstat formulas + base level"] --> B["Level assembly\ncap applied FIRST,\nbonuses added AFTER"]
  B --> C["Innate layer\nall-pets package\n+ family package"]
  C --> D["Owner transfer\ntraining skills + 3 gear slots\n+ focus reservation"]
  D --> E["The pet that fights"]
  style B stroke:#a33d2e,stroke-width:2.5px
  style E stroke:#c9a55a,stroke-width:2px

Stage two is the bug: the cap from Summon Tamed is applied to the template level first, and then Combat Training's +25 and the collar's +5 are added on top, so at grandmaster the "cap" of 89 produces level 114–119 pets — and stages three and four both scale off that escaped level. Stage three is the layer most players never see: every tamed pet carries a hidden package (+2 health per level, rising damage resistance, a flat 45 damage absorption), and tank families add the same shapes again on top. The Tier 1 cards operate on exactly these stages, in the open, with before/after tables.

Lessons: Ultima Online to modern

UO's Animal Taming was the genre's first real pet class, and it shipped with real friction: taming was slow and failure-prone, pets had loyalty that decayed and demanded feeding, a disloyal pet simply ignored orders or went wild, dead pets lost skills, and Veterinary was its own craft for keeping the beast alive. The friction was the identity — a grandmaster tamer with a pair of white wyrms had visibly earned them and visibly worked to keep them. UO also invented the problem every successor inherited: the best pets trivialized content, and the tamer's own contribution shrank toward walking. UO's eventual answer, the 2017 pet revamp, is the most instructive single change in the genre: every pet got a training budget, power got priced in points, and taking more power meant choosing it — the pet you built, not the pet you found.

What each game contributes

Ultima Online: loyalty and feeding made neglect visible; Veterinary made pet care a skill; control slots priced power in capacity; bonding made the individual animal the investment; the 2017 revamp turned raw pet power into a spendable budget. WoW: the happiness era proved engagement scaling works (and its removal proved players notice when a pet class stops asking anything of them); family abilities and specs gave every species a job; Bestial Wrath is the model burst window — enormous, brief, and yours to time. EverQuest: the deepest command vocabulary in the genre; a mage's pet is an instrument you play. FFXI Beastmaster: Ready moves are the direct precedent for our flagship — pet abilities the owner triggers off a charge economy. GW2: the F2 beast skill and mid-fight pet swap; also the cautionary tale, because for years the pet mostly played itself and rangers asked for exactly the control we're proposing. City of Heroes: Masterminds pay endurance upkeep and get bodyguard mode and tiered commands in exchange — power rented through attention. ARK: capture as gameplay (weaken, then commit) and imprinting — bond built by interaction; also the proof that full breeding is an entire game's worth of scope. UO freeshards (Outlands above all): pet ability trees, affinity depth, named rares — the community's living answer to "what would UO taming look like done today."

Capability matrix

● native · ◐ partial or era-limited · ○ absent. *UO shards = the modern freeshard canon, Outlands above all. "SotA today" = the unity-6.5 branch, 2026-07.

The twenty-eight proposals

Tier 1 is the rebalance, published with its math. Tier 2 is the toolkit that hands the removed power back as things you do. Tier 3 is the keystone systems and the Episode 2 decision card. Every tuning number on this page is an illustrative starting point: final values ship through the combat balance program's test-shard changesets and its weekly measurement windows, against the published bands (pet at most 60% of the tamer's damage; solo clear at most 1.25x the median school).

Tier 1 · Rebalance, in the open — data and tuning only

Tier 2 · The handler's toolkit — client work

Tier 3 · The big systems

Considered and rejected

• Breeding. Lineage, stat inheritance, and a breeding economy are an entire game's worth of scope, pointed at a school this program is tuning down.
• Multiple combat pets. Directly against the power goal, doubles the load on the buggiest part of the system (pathing), and we have no control-slot budget to price it with. S17's quick swap delivers the flexibility without the army.
• Feeding-based happiness. The alchemy proposal's Handler's Ration already owns pet food. Bond deliberately runs on actions, not consumables, so the two never collide.
• Pet permadeath. Too punitive for this game's covenant with its players. S15's Wounded state captures the intent at survivable cost.
• Stat-bearing rares or store items. Both re-add passive power through the back door; both cards (S19, S22) are locked to cosmetic and sidegrade for exactly that reason.
• Corpse-claim experience. Banking part of the kill's experience on the corpse punishes the same behavior S25 already prices, and taxes everyone else's moment-to-moment play to do it.
• A PvP pet overhaul. Real, and its own initiative; this page is scoped to the PvE economy of the school.
• Folding spell summons into taming. They share the pet machinery, and several cards here (stances, commands) will benefit them for free, but merging the schools is a mage-identity question this page does not own.

Sequencing

flowchart LR
  S1["S1 cap fix"] --> S18["S18 sidegrade gear"]
  S1 --> S6["S6 focus retune"]
  S10["S10 ability bar"] --> S13["S13 coordinated strike"]
  S10 --> S24["S24 Bond"]
  S11["S11 stances"] --> S24
  S12["S12 commands"] --> S24
  S24 --> S27["S27 training points"]
  S8["S8 family jobs"] --> S10
  S14["S14 veterinarian"] --> S15["S15 wounded"]
  style S1 stroke:#a33d2e,stroke-width:2px
  style S10 stroke:#c9a55a,stroke-width:2px
  style S24 stroke:#c9a55a,stroke-width:2px

Appendix

Already built — credited, not re-proposed

This page deliberately builds on what shipped recently rather than re-asking for it: the per-pet no-stun-attacks option, tamer credit stamped on filled taming necklaces, the pet damage panels in the combat analytics window, the pet sheet's real attack-damage display, and the deco-pet wander fix. If you have asked for one of those before: it exists, go enjoy it.

Where the numbers come from

Constraints honored

• Every tuning number is illustrative until it passes the balance program's measurement gates; nothing on this page ships by vibes.
• No card adds unbounded per-pet or per-player state; the two cards that save anything carry their byte cost in their notes.
• No card re-proposes shipped work, collides with the alchemy page's pet meal, or moves power onto store items.
• The engagement cards are tuned so that any real play — including pure support handling — never trips them; only literal zero-input play does.

Test surface

Existing pet suites cover the no-stun option, PvP disposition, the pet sheet's damage math, tamer credit, and wander leashing. Each shipping card adds its pin: the S1 cap-order test, S16's threshold-and-sign test, S5's duration-under-cooldown asset pin, S25's share-and-floor logic, S24's meter transitions, and S4's family sweep — pure-logic tests in the same style the school's recent fixes already use.

Sources

Shipping game data (species templates, skill sheet, innate and rune assets, gear archetypes, the taming matrix); our published balance targets and balance change log; the public bug tracker's taming and pet reports; the community wishlist survey, including the taming-redux and monetize-taming threads; and the genre record — UO's publish history and 2017 pet revamp notes, UO Outlands' wiki, WoW's hunter-pet history from Classic happiness through Mists taming challenges, EverQuest and FFXI Beastmaster references, GW2's ranger and soulbeast documentation, City of Heroes Mastermind guides, and ARK's taming and imprinting documentation.

General comments

No account needed. Every proposal above also has its own Discussion box for talking about that one alone; this thread is for the plan as a whole. Comments are plain text, held to basic decency, and may be trimmed.