Dragon Rider Army Guide 2026 (Hydra at TH18)

> Quick answer: Dragon Riders took an HP and damage cut at top level in the July 9 balance update, and they're still one of the two most-used armies at TH18. The version that works now is Hydra — Dragon Riders paired with Dragons or Electro Dragons rather than mass Riders alone. Bring Invisibility for the hero charge, Overgrowth for the defensive core, and expect Seeking Air Mines to be the thing that ends your attack.

Supercell was unusually direct about why the nerf happened. Dragon armies led by Dragon Duke and Dragon Riders were showing up in roughly 20% of TH18-versus-TH18 war attacks, and combined with Thrower and Healer armies, just two compositions accounted for about 40% of all attacks at the top.

That's a meta problem, not a balance accident. So the nerf was aimed at usage rate rather than at making the troop bad.

And it worked exactly that way — Riders are weaker and still completely viable. I've been running them all week and the attacks that fail now are the ones that were always sloppy.

Worth being precise here, because there's a lot of nonsense circulating with confident-sounding numbers attached.

Supercell deliberately did not publish stat tables for this update. Their words were that they're focusing on explaining the intent behind changes rather than turning the post into a stats breakdown. So anyone quoting exact hitpoint figures is reading them out of a datamine or making them up, and I'm not going to repeat numbers I can't source.

What the official post does say: the Dragon Rider lost hitpoints and damage at its top level only. Lower levels are untouched, which matters if you're not maxed — your Riders didn't change at all.

Dragon Duke's trap damage reduction was decreased, so he takes more from Seeking Air Mines than he used to. His rage ability was left alone.

On the other side, Electro Dragon got hitpoints and damage at its top two levels, and Meteor Golem plus Meteormites got a significant buff. That combination is what pushed the meta from mass Riders toward mixed air.

Supercell's official July balance post is the only source worth quoting on this. For the full patch rundown, we broke it down in the July 2026 balance changes post.

Mass Dragon Riders was always a bit of a blunt instrument. You threw fifteen of them at a side and let the healing carry it.

After the nerf that army loses just enough tempo to matter. Riders die slightly sooner, the wave stalls a fraction earlier, and against a base with a strong defensive core you run out of army before you run out of buildings.

Hydra fixes the tempo problem by mixing damage types. Dragon Riders push forward and soak, Dragons or Electro Dragons behind them clear buildings faster, and the two together clear a lane quicker than either does alone.

The Electro Dragon buff is what makes this work now rather than six months ago. Chain lightning that hits harder means the support half of the army actually contributes instead of trailing behind — I went through the unit properly in our Electro Dragon attack strategy guide.

My rough split at TH18 is Riders as the front line with a smaller Dragon or Electro core behind, adjusted by what the base's air defences look like.

Three spells decide whether a Hydra attack triples, and none of them are Rage.

Invisibility is the one to master. Used on a hero charge it lets the Royal Champion or Archer Queen walk past a defence cluster untouched, and the good players are carrying several. It makes everything except Walls and Siege Machines invisible and untargetable inside its radius, so it's protection and pathing control at once. If you're not confident with it yet, our Invisibility Spell guide covers the timing.

Overgrowth is the other one. Dropping it on the defensive core freezes those buildings out of the fight for over twenty seconds, which is usually enough for your air stack to clear a whole side before the core wakes up.

Freeze remains the panic button for Inferno Towers and the Town Hall Guardian.

Rage has quietly become the least valuable of the four in this army, because the problem isn't damage output, it's survival through the defensive core. Bring one, not three.

If your Hydra attack dies, it almost certainly died to Seeking Air Mines, and after the July change that's more true than it was.

Dragon Duke taking more trap damage means a well-placed cluster can delete your most expensive unit before he does anything useful. Six or seven mines grouped in one spot is genuinely capable of ending an attack in two seconds.

So the scouting habit that matters is counting. Look at the base before you drop anything and work out where the mines have to be based on what's covered and what's suspiciously empty. Air attacks fail from bad reconnaissance far more often than from bad micro.

The standard answer is a sacrificial unit — a lone Balloon or a single Dragon sent ahead to trigger the cluster. It costs you camp space and it saves the attack.

That's also why the anti-air base designs are so effective right now, and we mapped out how they're built in the TH18 anti-Dragon-Rider CWL bases post.

Hydra is a lane army. You pick a side, you commit, and you don't get to change your mind — which makes the thirty seconds of scouting worth more than any amount of deployment skill.

Start with air defences. Three or four clustered centrally is a different problem from four spread across the compass, and the spread version is generally harder for air because you can't freeze them all.

Then the Town Hall Guardian. It's the single biggest threat to a slow air push, and if it's positioned to cover your intended lane, either pick another lane or plan your Overgrowth around it.

Then Air Sweepers. Two facing your lane pushes your whole army back into the mines you were avoiding, and people ignore them constantly.

Last, look for the empty tiles. Gaps that don't make sense are traps, and at TH18 that means Seeking Air Mines. The base designer put them where an air army wants to fly.

If the base beats all four of those checks, take the other side. There's no shame in attacking the weaker half — our TH18 attack strategy guide covers when to switch.

Funnel first, always. Two or three Riders on each shoulder of the lane to clear outside buildings, so the main stack doesn't drift.

Then heroes with the funnel, not ahead of it. The Royal Champion charge with Invisibility support is what opens the core, and sending her early into an un-funnelled base is the most common way this attack falls apart.

Main stack down the middle of the cleared lane once the funnel holds. Riders in front, Dragons or Electro Dragons a beat behind so they don't overtake and eat the traps.

Overgrowth on the defensive core the moment the stack reaches it — not before. Dropping it early wastes half the duration on buildings that weren't shooting yet.

Freeze held in reserve for whatever wakes up: Inferno, Scattershot, the Guardian.

And if the attack is clearly going to time out rather than fail, remember 4x speed exists after two minutes. It won't save a bad attack but it will finish a slow one, which we covered in the 4x speed battle mode post.

Yes, and I'd say that even if the nerf had been twice as harsh.

The change hit top-level Riders only, so most players' Riders are exactly as strong as they were last month. If you're mid-progression, nothing about your army changed.

At max level you lost a little, and the troop is still one of the two most-used armies at the highest Town Hall in the game. Usage rate that high after a targeted nerf tells you everything about whether it's still good.

Where I'd redirect effort: Electro Dragon is now genuinely worth levelling because of the buff, and Meteor Golem went from fringe to a real answer against Guardian defences — our Meteor Golem guide covers what that buff unlocked.

Supercell also flagged that they kept this update deliberately targeted ahead of the next Crafting Season, which reads as a warning that a bigger rebalance is coming. Don't panic-swap your whole upgrade plan around one patch. The armies that were good in July will mostly still be good in August.