Fortress Merge Units Tier List

Units are the bodies that leave barracks, ranches, archer camps, and mage towers. They are not the buildings. A S-tier mage tower can still spawn a useless mage if you parked it on the lane. Rank the unit, then place the building with the placement guide.

Steam’s feature list names four families: knights, horsemen, archers, mages. Mobile patches add skins and extra spawners; the jobs stay the same: melee stall, fast flank, back-line shots, area spells.

S tier — mages (when they live)

Mages print area damage onto clumps. They are why a wall is worth a tile: the pack stops, the fireball lands, the wave ends. They are fragile. If your mage walks to the entrance they are B-tier at best. Keep the tower in the back, take radius or cooldown skills, and do not merge every tower on sight — two casters cover more ground. See when to merge.

A tier — archers and Teslas’ friends

Archers are the safe ranged unit. They do not steal the mage’s identity; they add steady shots while mages wind up. An archer camp in the back row is the default “I need a spawner and the shop denied mages” pick.

Knights (barracks) sit in A when you need bodies on the choke. They are not your damage plan on a boss that slams the front, but they keep slimes off the Tesla. Leave extra barracks unmerged so the population cap stays high.

B tier — horsemen

Horsemen leave the ranch fast and look like a win. They also scatter, die in the open, and eat skills that would have been better as building HP. Take ranches when you already have a wall and a turret. Skip them as your first camp on a leaky map. A 10% horseman-damage skill is a trap if you field two riders and four mages.

C tier — extras and event minions

Seasonal heroes can summon extra minions (Undertaker-style packs on mobile). Those bodies are C unless the event says otherwise: nice, not a missing wall. Token Bank or Training Camp spawners from later patches live here until you measure their cap against a barracks.

Heroes themselves are not on this list. Rank them on heroes. A hero skill is a button; a unit is a faucet.

Skills that rewrite the list

The draft can add 15% attack radius, 10% horseman damage, or free blockers. Radius on mages can bump them from “good” to “the run.” Horseman damage on a mage board does nothing. Building cooldown is a unit buff in disguise: every camp spits faster. Take globals first — gold and upgrades.

Population caps matter more than attack numbers on the tooltip. Two buildings with cap 2 can beat one building with cap 3 and a long cooldown. That is the merge hesitation in unit language.

Boss-wave delta

On wave 10, knights go up as extra stall, horsemen go down if they dive the slam, mages stay S if they are still in the back. If the boss crushes the front row, do not rebuild a ranch; rebuild wood and Teslas. The boss guide is the conversion chart.

Mixing families

A wall plus mages plus one archer camp is a complete plan. A wall plus knights plus horsemen plus archers plus mages is a zoo. Skills will not buff all of them. When the draft offers a family buff, you want that family to be most of the board.

If you already committed to mages, the next camp is another mage or an archer, not a ranch “for variety.” Variety is how wave 10 eats a scattered front. The boss guide is ruthless about this; this list is the reason.

Knights stay in the kit as stall units, not as a damage identity. If they are your only spawners, you are playing a B-tier plan on purpose — that is legal on early maps, not on challenges.

Related pages

• Buildings — the shops that spawn these units.
• Placement — front melee, back mages.
• Gear — hero stats that should not steal the budget from camps.
• Guides hub — the full how-to chain.