Which Commanders Should You Actually Invest In? Ranked by Role, Season, and What You Spend

What we're less sure of, and it matters most here. Individual commander kits, skills and availability are verified to the highest bar we have. The ranking — the ordering of one commander above another — is the weakest measured number in this whole project: it agrees with observed outcomes about two-thirds of the time and lands within one tier 58% of the time, against a 90% bar we set ourselves and have not met. Read the per-commander facts as fact. Read the order as our best current reading, not a settled result.

Who are the best commanders in Rise of Kingdoms right now?

For a kingdom in Season 3 or later, these hold the top grade, S+, in their troop type:

• Infantry: Sun Tzu (Prime), Bai Qi, Liu Che
• Cavalry: Arthur Pendragon (as a secondary, cavalry open field, Season 3+)
• Archer: Qin Shi Huang, Alp Arslan
• Engineering: Afonso de Albuquerque

Earlier seasons have different answers: see Who should you build in KvK 1 and KvK 2? below.

Every commander tier list you'll find online is wrong the same way: it answers a question nobody asked. "Who is the best commander?" has no answer. "Who is the best archer open-field commander I can actually obtain in my kingdom's current season, at what I spend?" has an exact one — and that's the question this guide answers, because it's the one you're actually living.

This page is not written from opinion. It is computed from a panel-verified roster of 141 commanders with season-of-arrival confirmed by three independent in-game sources, 413 pairing records each carrying the mechanism of why the pair works, and a set of damage models that score pairs mathematically — and decline to score what they can't model, which no tier list on the internet will ever admit to.

First, three definitions — because every word here is load-bearing

Invest means sculptures, stars, skill levels, and talent resets — spending that is effectively irreversible. Troops retrain; a maxed commander is forever. That's why a wrong answer here costs months, and why "just build the meta" is not advice.

A grade is not a job title. A commander can be superb in a rally and terrible leading one — the grade measures usefulness in the role; the lead recommendation comes only from pairing records that say who sits in the anchor seat. Should you build a commander to lead your rallies and garrisons? shows you the most expensive version of this mistake.

An era window is an expiry date. Some commanders are only good in a specific span of KvK seasons — before it they're unobtainable, after it the meta outgrows their kit. Our database enforces 26 ruled windows, and two of them were set by expert ground truth after we got them wrong ourselves: Alexander the Great is a KvK2–KvK3 commander and Pyrrhus is KvK1–KvK2 only. A tier list without windows is a list of ways to waste sculptures.

Which commanders can you get in your kingdom's season?

Your kingdom's season is the first filter — and half the internet lies about it. The roster you can actually buy from grows by season (find yours in Your Kingdom's Season Decides Everything — Look It Up, Then Use It):

Here's the part that should make you distrust every other list: when we cross-checked a widely-circulated commander database against the in-game, screenshot-verified roster, its season-availability flags were wrong on 40% of comparable records — almost all of them endgame legendaries marked as obtainable in KvK1. We quarantined that source from our own pipeline the day we measured it. If a guide tells a KvK1 player to save for Xiang Yu, the guide's data is broken; he is not on your wheel.

What are the best infantry, cavalry and archer commanders?

There is never one "best commander" — there is a best per troop type per role. These tables are the open-field answer for Season 3 and beyond. Grades below are our letter grades, S+ down to D, availability-filtered and era-checked. These are your investment priorities for field fighting. (Why rage and skill damage decide so much of it: How Damage Actually Works in Rise of Kingdoms.)

How to read the order within a grade: commanders sharing a grade are tied — S+ is S+. Where several sit at the same grade we list the one to build first, so the top row of each block is your opening investment rather than a photo-finish you have to adjudicate. Position inside a tier is a reading aid, not a ranking.

Best infantry commanders

Best cavalry commanders

Grades in this table are scoped to cavalry open field and the role named; an overall tier elsewhere (for example Arthur Pendragon’s A+ in our commander records) measures something different and doesn’t conflict.

Best archer commanders

Best leadership and engineering commanders

Leadership and Engineering, briefly: Honda Tadakatsu, Philip II, and Trajan hold S for mixed-troop leads, and Afonso de Albuquerque is the lone S+ in Engineering — his John Hunyadi pairing (ranged skill damage + rage generation) is the highest-graded open-field pair on file.

And here is the anti-tier-list proof, from our own table. On the raw roster scale, the highest-graded infantry open-field commander on the epic scale is Sun Tzu (Epic) — S across the board on that scale (epic and legendary grades are not comparable; the legendary S+ belongs to Sun Tzu Prime). And our model refuses to put him in the Season 3 table above, because his ruled window is Preparation through KvK2: the epic kit falls off a cliff in Season 2, and by Season 3 recommending him is a defect. A grade without a window isn't information; it's a trap wearing a medal.

Should you build a commander to lead your rallies and garrisons?

Leads are pairs, not star ratings.

The single most expensive mistake in commander investment is reading a role grade as a lead recommendation. The canonical example — one we got wrong ourselves before an expert corrected it, and which we now check every recommendation against: Yi Seong-Gye. His open-field and garrison grades are S. His rally-use grade is B+. And in every rally record he appears in, he is the secondary, never the lead — his AoE is a payload someone else delivers. Star him for garrison and field; never build a rally around him.

Which pair should lead your rallies, and which should hold your city, are pair questions, and the record answers them by matchup and by depth of evidence, not by a ranked table. That analysis is its own guide, Rally vs Garrison vs Open Field: Three Different Games — and the Data That Proves It. What it measured, in brief:

• Garrison: build Gorgo + Heraclius. It is the volume-proven standard: 1,280 governors and 20,932 measured garrison battles. If you already run Gorgo and want a second wall, Gorgo + Hector is the next best-evidenced, at 676 governors and a 93% garrison share. Rank walls by how many governors run them, not by a predicted score. (How a garrison actually fights: How Garrisons & Reinforcements Actually Work in Rise of Kingdoms.)
• Rally: no pair is best until you name the wall in front of you. Shapur I + Ashurbanipal trades at 165%, which is why 1,420 governors have run it across 219,069 recorded battles, while the hardest-hitting pairs trade worst. Scout the garrison first, and put your strongest account in the lead seat: only the rally leader's technology, VIP and equipment apply to the whole rallied army.

Who should you build in KvK 1 and KvK 2?

Preparation through KvK2 is a different game with different names. Early kingdoms are not a smaller version of the endgame — the correct names are almost completely different, and the spend-band splits are explicit in our records:

KvK1 — garrison: Charles Martel + Yi Seong-Gye is the ruled best-in-slot for mid/high spenders (Martel's shield and defense let YSG's AoE fire continuously); the F2P line is Sun Tzu (Epic) + Yi Seong-Gye, with Martel + Sun Tzu as the pure-tank fallback. Rally: the undisputed premier KvK1 cavalry rally is Minamoto no Yoshitsune + Cao Cao — a whale march (single-target nuke plus defense shred, speed, and heal) — with Charles Martel + Yi Seong-Gye as the infantry/archer hospital-melter for everyone else.

KvK2 — the arrivals change the board: Edward of Woodstock + Tomyris becomes the top archer rally (whale-band: single-target nuke + poison and buff-strip, but it's squishy and needs field control); Saladin + Mehmed II is the cavalry rally that prevents garrison recovery; Alexander the Great + Yi Seong-Gye is the staple infantry march nearly everyone runs — and the counter to the Edward rally is Saladin + Yi Seong-Gye in your garrison. The wider field: Alexander's window is KvK2–KvK3, so starting him at KvK2 is perfectly timed; Pyrrhus is KvK1–KvK2 only — a fine early investment that must be finished early, because his window closes when Season 3 opens.

One honest abstention, because it's the difference between us and a content farm: for open-field marching pairs in KvK1–2 our verified record set doesn't yet meet our publication bar, so we won't rank them here — the garrison and rally answers above are where early wars are actually decided anyway. (Your civilization picks your starting commander: Best Civilization in Rise of Kingdoms — And Why The Answer Changes Twice.)

When should you invest in a commander?

Timing is where investment advice earns its keep, and the era windows are the timing (what opens each season: KvK Explained: What Each Season Is, What Actually Triggers the Next One, and What To Prepare):

The general law: never start an investment you can't finish before its window closes — with one whale-band exception: if sculptures aren't your constraint, a window commander can be worth maxing for a single decisive KvK, because winning that KvK compounds into everything after it.

Which commanders should whales and mid spenders build?

What you spend changes the correct answer. Whale: you buy the meta the day it opens. Minamoto + Cao Cao at KvK1, Edward + Tomyris at KvK2, and at Season 3 the S+ rows above, your kingdom's dominant troop type first. Your risk isn't picking wrong — it's picking early into a closing window; check the table under When should you invest in a commander? before every wheel. (What that speed costs end to end: How Fast Can You Max Your Account — And What Does That Speed Cost?.)

Mid spend: Charles Martel + Yi Seong-Gye carries you through two full seasons in three different roles — the single most versatile investment of the early game. At Season 3, build one S+ open-field pair for your troop type to completion before touching a second; a finished S pair beats two half-built S+ pairs every time.

Which commanders should free-to-play and low spenders build?

F2P/low: the epics are your legendaries — Sun Tzu (Epic) is the best commander in the early game at any price, and he's free. Take the free-tier lines under Who should you build in KvK 1 and KvK 2?, accept the Season 2 rebuild as the cost of playing free, and bank your legendary sculptures for the Season 3 wave rather than trickling them into commanders whose windows are closing. And read the rest of this section before you spend a single head — the low-spend game has its own arithmetic, and it is not the whale answer with smaller numbers.

If you don't spend much, the question changes shape. Everything above ranks commanders by how well they fight. That is the right question for a whale and the wrong one for you, and it takes about a year of wasted sculptures to find that out the hard way.

Here's why. A whale's commanders are all near-max, so the differences between them are marginal — which is exactly why they need careful ranking. Your binding constraint isn't which pair is 8% better. It's whether you can get anything to a usable level at all.

The arithmetic that decides everything

Sculptures to get a commander to 5/5/1/1 (the denominations explained: Sculptures, Tomes & Pick-One Chests: The Denominations Everyone Misreads) — the bar where it does its job in a low-spend lineup:

Free commander sculptures arrive at roughly 1.35 a day from tavern chests. So a full year of free income is about 493 sculptures — 2.6 legendary field-minimums at absolute best.

⚠️ And "at best" is doing heavy lifting there, because free sculpture income is undirected. The chests hand you whoever they hand you, spread across your whole roster. You cannot aim it. So the real constraint isn't the rate — it's that the rate cannot be pointed at the commander you actually want.

That single fact reorganises low-spend play, and it produces four rules.

Rule 1 — A pair of epics arrives a season earlier than a pair of legendaries

Two legendaries at 5511 costs 380 heads. Two epics costs 280. That hundred-head gap is 74 days of perfectly-aimed free income — and far more once you account for the income scattering.

Epics are the low-spend meta not because they're stronger, but because they arrive. A finished epic pair fighting this season beats a legendary pair that's still half-built when the season ends.

Rule 2 — Concentrate on two or three names. Never spread.

Given a year buys under three field-minimums, an account that spreads sculptures across eight commanders arrives nowhere. Three commanders at the bar beat eight at halfway, every time, and it isn't close.

Rule 3 — Reach the field minimum and STOP until you're sure

This is the highest-value habit available to you, and it costs nothing to adopt.

There is an undo — the Rites of Renewal event refunds a legendary's materials at 100 gems per sculpture. Which means a mistake caught at 5511 costs 19,000 gems (~$76) to reverse, and one carried all the way to 5555 costs 69,000 gems (~$276).

Carrying a mistake to full multiplies the cost of being wrong by 3.6×. So take a candidate to the bar, actually field it, confirm it earns its place — then finish it.

⚠️ Two honest caveats: that reset is documented for legendary commanders only, and we have not confirmed whether epic investment can be unwound at all — so treat every head you put into an epic as permanent. And the event is progression-gated, so a newer kingdom may have no undo at any price.

Rule 4 — Aimable income is the scarcest thing you own

Since free sculptures can't be pointed, the channels that can be pointed are worth many times their raw count. Reserve them exclusively for the two or three names you've committed to:

• Daily Special Offer (Best Packs To Buy in Rise of Kingdoms — Ranked by the Numbers, by Game Stage, and by When To Buy Them) — 1 to 10 heads of a commander you select, around $2.99/day for the Collection. The only reliably directed channel below whale spend, and the single most efficient thing a low spender can buy.
• MGE rank rewards (MGE: How To Actually Get The Commander You Want) — sculptures are of the commander you chose. Free and aimed. But the featured pool rotates, so hoard through the wrong cycle and dump everything on the right one.
• The Wheel is directed but gem-priced at roughly 820 gems per featured head. That's a spend channel wearing a free costume; don't plan around it.
• The Medal Store — three directed rows, priced in Medals of the Conqueror, per sculpture head: Constance (Elite) at 400, permanent and unlimited · Aethelflaed (Legendary) at 1,500, capped 3/day · one featured Epic slot rotating weekly, unlimited inside its window (its price varies by rotation, so check the store; it was 1,000/head when we recorded it). The arithmetic is blunt: Aethelflaed to 5511 is 190 heads = 285,000 medals and at least 64 days at the cap, so she's a marathon you start early or skip; the featured Epic is 140 heads = 140,000 medals at that recorded price against a one-week window, so it only completes from a banked medal pile — hoard medals through the wrong rotations and dump on the right one, exactly like MGE. And we now know how fast medals arrive, which turns every bill above into days. Medals come from Expedition — a daily payout for every stage you hold at 3 stars, claimed at reset. It scales with how far you've cleared:

That makes Constance the best free commander purchase in the game for a low spender, and it isn't close. At Elite rarity she needs only 110 heads to 5511 — fewer than an Epic's 140 or a Legendary's 190 — and at 400 medals a head that is 44,000 medals with no daily cap, so income is the only limit. Push Expedition first: it is the cheapest thing you can do, it pays every day forever, and it is what makes every other row on this list reachable.

Note what the cap does to the other two. Aethelflaed's 3/day limit means she consumes only 4,500 medals a day no matter how rich you are — her 64 days is a cap floor, not an income problem, and no amount of Expedition clears shortens it. The featured Epic is the reverse: uncapped but windowed, needing ~17 days of even a maxed income against a 7-day slot, which is exactly why it only ever completes from a hoard.

⚠️ Denominator: about 19% of the full-clear figure is a banked pattern for stages 71–79 rather than a direct reading. The shape holds; we print 8,000+ rather than a precise number until the stages-71–79 readout lands.

And never spend universal legendary sculptures or golden heads on a gathering commander. Those have their own cheap directed source. Universal heads are combat currency, and at your budget there is no way to get them back.

What we can and can't tell you here

We can compute which pairs are reachable on a given sculptural budget and score them at the skill level you'll actually run them at — that's real, and it's a different answer from the whale table above.

But we're labelling the confidence honestly: the pair models are validated against battle records at maximum kit. Scaling them down to 5511 is principled kit math rather than a measurement, because battle reports don't expose the investment level of the marches in them. So treat the low-spend ordering as well-reasoned rather than proven — stronger than a tier list, weaker than the whale ranking above it.

⚠️ And there's a second gap, which is about getting the commander rather than raising one. Everything above prices what it costs to raise a commander — the sculpture bill is exact and we stand behind it. Where we're weaker is telling you where a specific commander comes from. Of our 141 commander records, 82 carry an acquisition channel that's vague or absent — a "legendary pool" entry that doesn't tell you which chest, event, or shop actually puts heads in your bag. Only 59 name a channel you could act on.

That gap lands hardest exactly here, on this section, and for a reason worth saying out loud: a whale buys past any channel, so this never bites them. You're the reader it bites. When we tell you to concentrate on two or three names, the unanswered half is whether those names are even reachable through the channels your account has.

What's solid meanwhile: the channels in Rule 4 above are verified, the sculpture bill is exact, and the broad classes are known — most commanders come through heads and universal heads via events and chests, a few are locked behind Expedition, and some route through the Medal Shop. What we can't yet do is tell you reliably which of those applies to a named commander. We're building that model rather than guessing it, and until it's done, verify a commander's source in-game before you commit a budget to chasing them.

The doctrine matters more than the ordering anyway. Which two or three names you commit to, and how far you carry them before you're sure, will decide your account far more than whether you picked the third-best pair instead of the first.

Which commander mistakes cost the most?

The traps — each one from a verified record.

1. The availability lie. 40% of season flags in a popular dataset were wrong. If a list says an endgame legendary is on your KvK1 wheel, the list is broken, not the game.
2. The grade-as-job-title trap. YSG rally-use B+, rally records: always secondary. A star rating is not a seat assignment.
3. The windowless tier list. Sun Tzu (Epic) is S+ on paper and a defect to recommend at Season 3. Alexander and Pyrrhus each cost real players real sculptures learned the hard way.
4. The early Prime. No Prime before KvK3. Ever. We never recommend one, because it kept appearing in the wild.
5. The rarity twin. There are two commanders named Pelagius, and their pages carry the identical title. Their actives are the discriminator: the Epic opens with Charge (single target, DF 300 + a 2-second burn), the Legendary with Reconquista (a fan hitting up to 3 troops). Verify which one you're staring at before a single sculpture moves.

What this guide can't know

It can't see your roster. It doesn't know that you're KvK2 with Alexander at 4 skills, 137 universal sculptures banked, an alliance that runs archer rallies, and an MGE for a commander whose window closes for you in one season. That's where the advisor comes in: you tell it about your account; screenshot reading is live on paid plans. Once you've told it your commanders, your season and your sculptures, the answer isn't "Bai Qi is S+," it's "finish Alexander's expertise before your KvK3 starts in 5 weeks, skip this MGE, here are the two marches that change."

Free: Your first five reads are full ones — the whole answer, chart and all. After that, your questions get the short read: the key fact, straight, and the right next question to ask. Companion $19/month or $190/year · Emperor $69/month or $690/year · two months free on annual