Home / Tier list

Tier list & value

Every number is computed by the build script from local game files, not copied. Re-run the data pipeline after a patch and the tables update themselves.

What this table measures, and what it does not.There are two denominators: power per gold (√(L1 total DPS × L1 total HP) ÷ cost) when gold is tight, and power per plot (same score without dividing by cost) when gold is not the constraint, because one plot holds one card. Neither measures positioning, range value, AOE, control, economy, support adjacency or policy synergy. Those exclusions are listed below — read that section before acting on the numbers.

Troop tier list (38 rankable troops)

Tier	Troop	Cost	Units	L1 DPS	DPS/gold	L1 HP	HP/gold	Max DPS	Atk/s	Growth	Max multiplier	HP ×	Atk ×	Max HP	Range	Crit†
S	Imp	2	13	48.8	24.38	169	84.5	707.9	0.86	6.60×	14.52×	3.00×	4.00×	507	10	5%
S	Skeleton	2	13	37.1	18.57	182	91.0	250.7	0.71	5.63×	6.75×	4.69×	2.25×	853	10	0%
S	Ghost	2	13	27.3	13.65	182	91.0	117.9	0.67	4.32×	4.32×	4.32×	1.44×	786	10	5%
S	Exploder	5	6	78.0	15.60	396	79.2	526.5	1.00	5.63×	6.75×	4.69×	2.25×	1856	1.1	0%
S	Hound	2	9	18.9	9.45	216	108.0	138.4	0.78	5.63×	7.32×	4.32×	1.56×	933	4	5%
S	Leech	2	9	18.9	9.45	216	108.0	138.4	0.78	5.63×	7.32×	4.32×	1.56×	933	4	5%
S	Soldier	2	9	18.9	9.45	216	108.0	138.4	0.78	5.63×	7.32×	4.32×	1.56×	933	4	5%
A	Piercing Demon	6	9	56.7	9.45	468	78.0	265.8	0.78	5.63×	4.69×	6.75×	1.00×	3159	5	5%
A	Replicator	2	9	11.3	5.63	198	99.0	58.8	0.58	4.75×	5.23×	4.32×	1.44×	855	4	5%
A	Dwarf	2	9	8.3	4.13	162	81.0	60.5	0.47	5.63×	7.32×	4.32×	1.56×	700	4	1%
A	Wizard	3	9	21.6	7.20	135	45.0	198.4	0.20	6.56×	9.19×	4.69×	3.06×	633	1	0%
A	Scapegoat	2	6	6.0	3.00	180	90.0	18.0	1.00	3.00×	3.00×	3.00×	1.00×	540	4	0%
A	Trapper	2	5	12.0	6.00	90	45.0	73.6	0.48	5.58×	6.13×	5.07×	1.69×	456	10	0%
A	Boar	3	6	22.2	7.41	96	32.0	96.1	0.59	4.68×	4.32×	5.07×	1.44×	487	1	5%
B	Thief	5	9	38.9	7.78	90	18.0	182.3	0.40	4.69×	4.69×	4.69×	1.56×	422	4	20%
B	Garrison	2	9	9.0	4.50	45	22.5	60.8	1.00	5.63×	6.75×	4.69×	2.25×	211	4	0%
B	Archer	4	9	12.4	3.09	126	31.5	98.9	0.98	5.86×	7.99×	4.69×	1.56×	591	10	10%
B	Warlord	4	9	15.5	3.87	99	24.8	72.9	0.31	4.50×	4.71×	4.69×	1.44×	464	10	10%
B	Gigafunghi	2	9	30.0	15.00	9	4.5	202.5	0.48	5.63×	6.75×	4.69×	2.25×	42	4	0%
B	Sporeling	2	9	30.0	15.00	9	4.5	202.5	0.48	5.63×	6.75×	4.69×	2.25×	42	4	0%
C	Executioner	7	9	14.4	2.06	207	29.6	62.2	1.33	4.32×	4.32×	4.32×	1.44×	894	10	20%
C	Raptor	5	6	12.6	2.52	96	19.2	85.0	0.50	6.75×	6.75×	6.75×	2.25×	648	1	5%
C	Frog	2	3	3.4	1.69	42	21.0	22.8	0.71	5.63×	6.75×	4.69×	2.25×	197	10	5%
C	Small Slime	2	3	3.4	1.69	42	21.0	22.8	0.71	5.63×	6.75×	4.69×	2.25×	197	10	5%
C	Orbiter	9	9	10.6	1.18	234	26.0	60.5	0.43	4.75×	5.70×	4.32×	1.44×	1011	10	10%
C	Lab RatNo level cap	2	2	3.4	1.68	26	13.0	10.5	0.33	3.12×	3.12×	3.12×	1.04×	81	10	1%
C	Elf	6	3	7.2	1.19	90	15.0	40.8	0.15	4.75×	5.70×	4.32×	1.44×	389	10	10%
C	Paladin	6	3	6.3	1.05	90	15.0	42.5	0.25	4.95×	6.75×	3.63×	2.25×	327	1	5%
C	Slime	3	3	3.4	1.13	42	14.0	14.6	0.71	4.32×	4.32×	4.32×	1.44×	181	10	5%
C	Golem	6	3	5.4	0.90	90	15.0	49.6	0.20	6.56×	9.19×	4.69×	3.06×	422	1	0%
C	Shaman	6	6	4.8	0.80	78	13.0	50.6	0.31	7.03×	10.55×	4.69×	2.25×	366	1	0%
C	Ogre	3	3	2.1	0.72	42	14.0	9.4	0.28	3.99×	4.39×	3.63×	1.21×	152	4	5%
D	Mob	5	3	4.9	0.98	42	8.4	21.3	0.31	4.32×	4.32×	4.32×	1.44×	181	4	5%
D	Demon	7	1	6.8	0.98	56	8.0	58.5	0.84	7.61×	8.57×	6.75×	1.69×	378	1	5%
D	Warper	11	5	2.9	0.27	110	10.0	14.9	0.06	4.68×	5.07×	4.32×	1.69×	475	1	0%
D	Warlock	6	3	0.7	0.11	81	13.5	3.5	0.13	5.15×	5.23×	5.07×	1.44×	411	1	0%
D	Ballista	11	3	2.1	0.19	48	4.4	9.8	0.08	4.68×	4.68×	5.07×	1.44×	243	10	20%
D	Defender	3	3	0.0	0.00	96	32.0	0.0	1.00	4.50×	3.00×	6.75×	1.00×	648	10	0%

Sorted by power per gold. Max DPS and multiplier use the level formula below. Baseline 2026-11-06.When gold is not the constraint, use the per-plot table.

† Crit is split in two, and only one half is verified.The base CriticalChance is a client field; the run bonus is additive(base + 0.001 × run index n) and matches 8/8 local save samples. The “+5 percentage points per level” reading (wiki.gg: Archer 10% → 15% → 20%) has no decidable sample, so per-level crit stays unproven. It is a reference column only and is not used in the radar or growth curves.

Tier thresholds

S Clearly ahead at the same costA Strong and build-definingB Average; needs supportC Low marginal value or nicheD Clearly behind at the same cost

S ≥ 30 · A ≥ 15 · B ≥ 8 · C ≥ 3 · D < 3, in units of √(DPS×HP) per gold. Thresholds are arbitrary reading aids; the ranking itself is continuous.

When gold is not the constraint: rank by plot

The gold denominator stops working once gold is solved.Only gold can multiply itself: Loan +99 (20 stacks = 1980), Vault +3/year/level, Burglary gold on kill, Salvage 27x refund. Plots are the thing you cannot buy: fixed at start, +3 per Exploration/Wilderness, and one card per plot. Late-game scarcity is plots > card slots > levels > waves > gold.

Tier	Plot rank	Troop	Cost	Units	Per plot L1	Per plot L3	Max HP	HP ×	Atk ×	Growth	Gold rank	Move	Max multiplier	Range
S	1	Exploder	5	6	176	989	1856	4.69×	2.25×	5.63×	#4	↑3	6.75×	1.1
S	2	Piercing Demon	6	9	163	916	3159	6.75×	1.00×	5.63×	#8	↑6	4.69×	5
S	3	Imp	2	13	91	599	507	3.00×	4.00×	6.60×	#1	↓2	14.52×	10
S	4	Skeleton	2	13	82	462	853	4.69×	2.25×	5.63×	#2	↓2	6.75×	10
S	5	Hound	2	9	64	359	933	4.32×	1.56×	5.63×	#5	—	7.32×	4
S	6	Leech	2	9	64	359	933	4.32×	1.56×	5.63×	#6	—	7.32×	4
S	7	Soldier	2	9	64	359	933	4.32×	1.56×	5.63×	#7	—	7.32×	4
S	8	Wizard	3	9	54	354	633	4.69×	3.06×	6.56×	#11	↑3	9.19×	1
A	9	Ghost	2	13	70	305	786	4.32×	1.44×	4.32×	#3	↓6	4.32×	10
A	10	Thief	5	9	59	277	422	4.69×	1.56×	4.69×	#15	↑5	4.69×	4
A	11	Orbiter	9	9	50	247	1011	4.32×	1.44×	4.75×	#25	↑14	5.70×	10
A	12	Archer	4	9	39	242	591	4.69×	1.56×	5.86×	#17	↑5	7.99×	10
A	13	Executioner	7	9	55	236	894	4.32×	1.44×	4.32×	#21	↑8	4.32×	10
A	14	Raptor	5	6	35	235	648	6.75×	2.25×	6.75×	#22	↑8	6.75×	1
A	15	Replicator	2	9	47	224	855	4.32×	1.44×	4.75×	#9	↓6	5.23×	4
A	16	Boar	3	6	46	216	487	5.07×	1.44×	4.68×	#14	↓2	4.32×	1
A	17	Dwarf	2	9	37	206	700	4.32×	1.56×	5.63×	#10	↓7	7.32×	4
A	18	Warlord	4	9	39	184	464	4.69×	1.44×	4.50×	#18	—	4.71×	10
A	19	Trapper	2	5	33	183	456	5.07×	1.69×	5.58×	#13	↓6	6.13×	10
B	20	Demon	7	1	20	149	378	6.75×	1.69×	7.61×	#34	↑14	8.57×	1
B	21	Golem	6	3	22	145	422	4.69×	3.06×	6.56×	#30	↑9	9.19×	1
B	22	Shaman	6	6	19	136	366	4.69×	2.25×	7.03×	#31	↑9	10.55×	1
B	23	Elf	6	3	25	126	389	4.32×	1.44×	4.75×	#27	↑4	5.70×	10
B	24	Paladin	6	3	24	118	327	3.63×	2.25×	4.95×	#28	↑4	6.75×	1
B	25	Garrison	2	9	20	113	211	4.69×	2.25×	5.63×	#16	↓9	6.75×	4
B	26	Scapegoat	2	6	33	99	540	3.00×	1.00×	3.00×	#12	↓14	3.00×	4
B	27	Gigafunghi	2	9	16	92	42	4.69×	2.25×	5.63×	#19	↓8	6.75×	4
B	28	Sporeling	2	9	16	92	42	4.69×	2.25×	5.63×	#20	↓8	6.75×	4
C	29	Warper	11	5	18	84	475	4.32×	1.69×	4.68×	#35	↑6	5.07×	1
C	30	Frog	2	3	12	67	197	4.69×	2.25×	5.63×	#23	↓7	6.75×	10
C	31	Small Slime	2	3	12	67	197	4.69×	2.25×	5.63×	#24	↓7	6.75×	10
C	32	Mob	5	3	14	62	181	4.32×	1.44×	4.32×	#33	↑1	4.32×	4
C	33	Slime	3	3	12	51	181	4.32×	1.44×	4.32×	#29	↓4	4.32×	10
C	34	Ballista	11	3	10	49	243	5.07×	1.44×	4.68×	#37	↑3	4.68×	10
C	35	Ogre	3	3	10	38	152	3.63×	1.21×	3.99×	#32	↓3	4.39×	4
C	36	Warlock	6	3	7	38	411	5.07×	1.44×	5.15×	#36	—	5.23×	1
D	37	Lab Rat	2	2	9	29	81	3.12×	1.04×	3.12×	#26	↓11	3.12×	10
D	38	Defender	3	3	1	2	648	6.75×	1.00×	4.50×	#38	—	3.00×	10

Per-plot power = √(max-level total DPS × max-level total HP), without dividing by cost. Thresholds: S ≥ 350 · A ≥ 180 · B ≥ 90 · C ≥ 30.

Side-by-side comparison

Tick 2–3 troops. The dumbbell chart puts every axis on one horizontal percentile scale; the radar is a secondary view. The selection is kept in the URL (?c=archer,orbiter&m=hp,dps).

Axes (tick 1 or more; >5 switches to percentile cells)

Extra axis

Allowed names: hp, dps, spd, walk, units, cost, hpPerCost, dpsPerCost, growth, with + - * / and brackets. No scripts.

Percentile per axis (band = P25–P75)

Archer

Survival62% · 591 HP

Damage64% · 82.4 DPS

Atk speed86% · 0.98 /s

Mobility54% · 1.50 tiles/s

Orbiter

Survival93% · 1,011 HP

Damage38% · 50.4 DPS

Atk speed36% · 0.43 /s

Mobility14% · 1.00 tiles/s

Total HP

Total DPS (no crit)

Archer
Orbiter

Troop	Cost	Units	L1 HP	Max HP	L1 DPS	Max DPS	Atk/s	Growth	HP ×	DPS ×	Speed	Range
Archer	4	9	126	591	11.3	82.4	0.98	5.86×	4.69×	7.32×	1.50	10
Orbiter	9	9	234	1,011	9.6	50.4	0.43	4.75×	4.32×	5.23×	1.00	10

Curves start at level 1 and stop at MaxLevel. A flat line means the multipliers really are 1.0, not missing data. Percentiles are computed against all comparable troops; grey bands mark P25–P75.

Support / economy buildings

Buildings have no independent cost field, so they cannot be scored per gold. Instead they are rated by whether they add a multiplier or an additive stat.

S Clearly ahead at the same cost

Building	Client field	Why
Vault	`currencyPerLevel = 3`	Settles +3 gold per level each year. Base yearly income is only 9, so a level-3 vault doubles it; the only reliable compound economy building.
Codex	`CardsPerLevel = 1`	+1 hand size per level. Hand size is a raw resource, not a stat: one more card is one more multiplication opportunity.
Farm	`unitsPerEffect = 1`	Gives one adjacent troop +1 unit each year. Unit count is a multiplier in the level formula, worth far more than +2% stats.
Shrine	`initialCost = 9; costDecrementPerLevel = 3`	Spend 9/6/3 gold for a random level-up; the only way to buy levels directly with gold. Levels are the main source of power.

A Strong and build-defining

Building	Client field	Why
Demon's Altar	`FlatStatBoostPerAlly = 0.1; FlatSpeedBoostPerAlly = 0.002`	Each allied death gives a permanent +10% boost. The more allies you have, the faster it snowballs; core of the blood imp swarm.
Quarry	`damageBoost = 1.02; costPerLevel = 9`	All towers and castles gain +2% damage per year, and each level costs only 9 gold. The only growth curve for tower builds, but it is linear.
Blacksmith	`strengthPercentage = 1.02; projectilesPerLevel = 2`	Adjacent units gain +2% damage per year and +2 projectiles per level. Projectiles are a count term, better than the 2%.
Beacon	`speedBoost = 1.02`	Adjacent units gain +2% attack speed per year. Attack speed multiplies DPS like damage, but adjacency is restrictive.
Temple	`statBoost = 1.09`	Random plot, random stat, x1.09, one-shot but large; effectively a free small level.
Dispenser	`boostPerCoin = 1.005`	Gain +0.5% attack speed per gold received. The higher your income, the stronger it gets; the greed build turns gold into combat power.
Library	`baseEffectChancePercentage = 50; chanceReductionStartSuccessCount = 3; reducedEffectChancePercentage = 20`	50% chance to copy an adjacent enchantment, dropping to 20% after 3 successes. A free copy machine for enchantment builds.
Trebuchet	`projectilesPerBuilding = 1`	+1 projectile per building. Building count becomes a multiplier; core of tower/building builds.
Doomstower	`damageMultiplierOnLifeLost = 9; MaxRayTargets = 99`	Damage x9 when you lose a life, up to 99 targets. A comeback piece: the fewer lives you have, the stronger it is.

B Average; needs support

Building	Client field	Why
Cauldron	`poisonStacksPerLevel = 3`	3 poison stacks per level. Only valuable with Venom / Poison Thorns; thin on its own.
Wall Maker	`healthBoostPerYear = 1.1`	Wall HP x1.1 per year (compound). Pure defense unless you play reflect/stall.
Orchard	`rootsPerLevel = 3; StunDuration = 9`	3 roots per level, 9s stun. Control material; only multiplicative with Weakspot.
Flametower	`strengthPerTroop = 2`	Damage scales with nearby troop count; a ground tower for swarm builds.
Mangler	`strengthPerTroop = 2`	Same as above: damage per nearby troop; comes in the Blood king starter.
Recycler	`damageBoostPerLevel = 0.02`	+2% damage per level with no adjacency restriction, but the same linear curve as Quarry.
Reactor	`RaysPerRainbowCard = 1`	Rays per rainbow card. Only Refraction / rainbow builds can feed it.
Repeater	`raysPerLevel = 3`	3 rays per level; pure count stacking with a low ceiling.
Trashmaker	`TrashCardsPerLevel = 1`	+1 trash card per level. Only an economy piece with Salvage (pit refunds).
Portal	`DefaultTravelYears = 2`	Teleports troops across years; a Time king utility piece with no direct combat stats.
Dragon's Den	`plotDestroyedBoost = 1.2; dragonSpeed = 12`	x1.2 boost after destroying a plot; only demolition builds can use it.

C Low marginal value or niche

Building	Client field	Why
Camp	`statBoost = 1.02`	+2% single-plot stats per year; same tier as adjacency buildings but narrower.
Forest	`hpBoost = 0.02`	+2% HP per year. HP is not damage; lower value than buildings on the same tier.
Spire	`raysPerLevel = 3; raySubdivisionCount = 30; rayAmplitude = 0.1`	Healing rays; only comes online with Spiremania / Revitalize.
Great Hall	`BaseStatInheritance = 0.5; SpawnRandomness = 2.77`	Inherits 50% of base stats with randomness; variance beats the expected value.

Ratings are editorial; the field values are extracted and now shown on each building page.

Special troops the generic score cannot rank

Troop	Cost	Real role	Client field
Mercenary	10	Unit count is set by gold, not the card face	`MTroopCount = 0; goldPerTroop = 9`
Lab Rat	2	The only troop without a level cap; can break the level-3 ceiling	`see card page`
Scapegoat	2	Economy piece: produces gold on death, not a combat unit	`see card page`
Wizard	3	Spell-build core: each spell card adds damage	`see card page`
Warlock	6	Tome-build front line; low card DPS is intentional	`see card page`
Soldier	2	Swarm component: more units, more damage	`see card page`
Hound	2	Scales with the number of digits in a value; field semantics not fully confirmed	`see card page`
Leech	2	Life steal: steals 9% of stats per level	`see card page`
Replicator	2	Refills 9 units every wave	`see card page`
Small Slime	2	Splits up to 2 levels; spreads across the board	`see card page`
Slime	3	Same as above: split-and-spread piece	`see card page`
Defender	3	Reflect wall: near-zero attack is normal	`see card page`
Ballista	11	Single penetrating shot with a very slow attack; useless without levels	`see card page`
Ogre	3	+5% all stats per empty adjacent plot; stronger alone	`see card page`
Dwarf	2	Stronger with more locked plots	`see card page`
Warper	11	Summons minions with no cap	`see card page`
Thief	5	Teleports into the back line; assassin core with guaranteed-crit policies	`see card page`
Trapper	2	Control: one trap and a 0.5s immobilize	`see card page`
Piercing Demon	6	Line penetration; positioning decides real damage	`see card page`
Paladin	6	Converts damage from HP ratio; trades tankiness for output	`see card page`
Boar	3	Rider inheritance: rider strength scales with HP	`see card page`
Raptor	5	Transfers 100% of stats to the rider, speed x4	`see card page`
ZMegaminion	2	Legacy/test card; not obtainable	`id = zmegaminion`
ZBear	6	Legacy/test card; not obtainable	`id = zbear`
Mercenary	10	Unit count is set by gold; the static formula does not apply	`MTroopCount = 0; goldPerTroop = 9`

Level formula and verification

Units	`Units × level`
HP per unit	`MaxHP × MHPBoost^(level−1)`
Attack per unit	`AttackStrength × MAttackStrengthBoost^(level−1)`
Attack interval	`AttackDelay ÷ MAttackDelayBoost^(level−1) (a shorter interval is faster; attack speed = 1 ÷ interval)`
Crit (run bonus)	`CriticalChance + 0.001 × run index n` 8/8 save samples
Crit (per level)	`Unverified (no sample with L≥2 and MCritChanceBoost≠1 in the whole card set)` unverified
DPS	`attack ÷ interval × (1 + crit chance × (crit multiplier − 1)) × units`

Community wiki per-level tables. Imp 13 → 26 → 39 and Archer 9 → 18 → 27 match "units × level".
Attack speed direction. Hits/second rises with level, so MAttackDelayBoost is a divisor, not a multiplier.
Crit is two parts: the run bonus matches 8/8 local save samples (CriticalChance + 0.001 × n); the per-level part (MCritChanceBoost) has no L≥2 sample and stays unverified, so it is not a radar axis.
Consistency. Applying the formula to all 41 troops with a Cost field produces no negative, infinite or MaxLevel conflicts.

Known uncertainty: (1) Cost exists only for troops, so building cost cannot be cross-checked; (2) the wiki pages are from 2026-07 and some values have changed — the local client wins; (3) crit is capped at 100% in this table even though the data does not express a cap; (4) per-level crit is unproven — there is no L≥2 sample, so the Crit column is labelled community; the run bonus is additive and reconciled 8/8, and the radar/growth charts still exclude crit.