Who can actually lead a Pauper Commander (PDH) deck

5,737 cards can lead a Pauper Commander (PDH) deck, and only 765 are legendary creatures. The rule, the count by colour identity, what the pool costs, and how many lands real lists play.

Aymeric Lamontagne

Nov 10, 2026

Here is the honest headline: the commander is not the scarce part of Pauper Commander (PDH). In the card data Scryfall published on November 2, 2026, 5,737 distinct cards can sit in a PDH command zone. Only 765 of them are legendary creatures. Another 4,884, or 85% of the list, are creatures no other Commander format would let near the zone, like Fusion Elemental, Composite Golem and Bear's Companion.

The pool is wide, but it is not even. Almost three-quarters of it is one colour. No card has a four-colour identity, and no legal pairing gets there either. A Temur player gets exactly two names. The median eligible card, meanwhile, costs twenty cents, and the median public list we read plays 35 lands.

This piece covers the rule and who actually maintains it, what the list is made of, where the colours run out, what it costs, the edge cases that catch people out, and those land counts.

The rule, and where it actually lives

PDH is a fan format. It is run by a volunteer Rules Committee at PDH Home Base, not by Wizards of the Coast, and its rules page, read on November 4, 2026, is short:

• Commander: an uncommon creature, vehicle or spacecraft. It does not need to be legendary.
• The other 99: commons.
• Rarity: a card counts at a rarity if it has been printed at that rarity, judged on paper and Magic Online releases. Arena does not count.
• Life: 30 starting life and 16 commander damage.
• Banned: Mystic Remora, Rhystic Study, cards Wizards has removed from all constructed formats, Conspiracy cards and cards that refer to ante.

Everything else follows Commander: 100 cards, one copy of each, and a 99 inside the commander's colour identity.

Two things sit underneath that page, and they do most of the work. The first is the committee's November 2024 framework post, which narrowed what "printed" means. Core and expansion sets, Masters sets, draft-innovation sets, bonus sheets and precons can shift a card's rarity. Arena-only products, promos, Secret Lairs, Masterpieces, box toppers and local-language releases cannot.

The second is a Scryfall search, is:paupercommander, which the same post introduced. It is a maintained list, not a formula, and it is the list the tools read. It returned 5,737 cards when we queried it live on November 4.

That split explains a page that looks wrong. Since the framework change, the "Pauper Commander" legality on a card page describes the 99 and nothing else. Laboratory Maniac reads not legal there, and it can lead a deck. It is one of 5,299 eligible cards that read that way. The other 438 read legal because they also have a common printing: Blastoderm is a common in Nemesis and an uncommon only in Vintage Masters on Magic Online, so it can lead a deck or sit in one. For the record, 10,998 cards are legal in the 99.

What the list is made of

Eligible cards

5,737 — Distinct cards, not printings

Non-legendary creatures

4,884 — 85% of the list

Legendary creatures

765 — 13% of the list

Vehicles and spacecraft

73 — 63 vehicles, 10 spacecraft

Backgrounds

15 — Second commander only

Scryfall’s is:paupercommander list, split by the front face’s type line. Card data of November 2, 2026.

Rarity is the filter here, and legendary status is decoration. The usual question of "which legend do I want?" becomes "which uncommon body do I want to see every game?"

The legends are also newer than they feel. Of the 765, 680 first reached uncommon in 2020 or later, and 387 did so in 2025 or 2026. Marvel Super Heroes and its Commander decks supplied 147 of those on their own. The older legends are a thin layer: 25 of them date from 1994 and 1995, and only 28 more arrive between 1996 and 2017.

So the list moves, and it moves quickly. Our index already holds Star Trek and Star Trek Commander, dated November 13, 2026, with 78 uncommon creatures and spacecraft that are not yet on the list. Of those, 53 are legendary creatures. Read 5,737 as a count on one date, not a fixed number.

Where the colours run out

Four colours: none. A zero like that looks like a counting bug, so we checked. The index holds 27 four-colour cards, among them Atraxa, Praetors' Voice, Breya, Etherium Shaper and Yidris, Maelstrom Wielder. Every printing of every one is rare or mythic, so none of them qualifies.

Pairs do not rescue it. The list holds 30 cards with plain Partner, ten with Partner with, 20 that choose a Background and 15 Backgrounds, and all 75 are a single colour. That gives 740 legal pairings: 135 land on one colour, 605 on two, and none on three or more. In PDH a pair is another way to build a two-colour deck. It is not a way to go wider.

Three colours: thin, and lopsided. There are 67 cards across ten identities, split sharply into two groups.

The mechanism is print history. Three-colour uncommons appear when a set is built around three colours. The shards have had Invasion block, Shards of Alara block and Streets of New Capenna. The wedges have had Khans of Tarkir and Tarkir: Dragonstorm, which between them supply 13 of the 14 wedge cards. No wedge has a legendary creature on the list at all.

Five colours: twelve, and mostly not what they look like. Colour identity counts mana symbols in the rules text, so only four of the twelve are printed in all five colours: Fusion Elemental, Garth One-Eye, and the two that declare it, Sphinx of the Guildpact and Transguild Courier. Six are printed in one colour, including Fleshformer, a three-mana black creature that opens a five-colour 99 with a single activated ability. Karolina Dean, Runaway does it as a four-mana red legend. The last two are colourless: Composite Golem and the Vehicle Marshals' Pathcruiser.

The same rule fattens the three-colour row. Of those 67, 25 are not printed in three colours. The committee's own rules use Dreadwing as the example, a black creature whose blue and red ability sets its identity. This is why a count by printed colours gives a different and less useful answer: the deck is built from identity, so the list has to be read that way.

Everywhere else: plenty. The thinnest mono-colour identity still has 780 options, and the thinnest colour pair has 98.

If a legend matters to you, it is worth reading the last column rather than the second. Blue has 780 cards, but 79 legends. Colourless has 326, but just six.

What the pool costs

On November 2, 2026, the median eligible commander cost $0.20 at its cheapest paper printing. The 90th percentile was $0.37. Of the 5,736 cards with a paper price, 5,449 were under $1, which is 95%. Buying one copy of each would have come to about $3,906, a number nobody needs but which does put the format in proportion.

Scarcity of colour does not show up in that price at all. The median three-colour card was $0.17 and the median five-colour card $0.15, against $0.19 for mono-colour. Legendary creatures were barely dearer than non-legendary ones, at $0.22 against $0.19. Prices here follow print runs and demand from other formats, and neither of those cares how crowded an identity is in PDH. A rare identity is a deckbuilding problem, not a price problem.

The expensive end has a stranger explanation. Just 13 cards cost $50 or more, and for ten of them the qualifying uncommon printing exists only on Magic Online. The rule classifies the card rather than the copy, so a paper player can lead with one, but the only paper copies are old rares.

So a format built on cheap cards has a $184.50 commander available on request. It is legal. It is also a strange way to spend the money a pauper format saves you.

The edge cases the data exposes

Most arguments about PDH legality land on one of these.

Arena does not count. There are 197 creatures, vehicles and spacecraft whose only uncommon printing is on Arena, and none is on the list. For 159 of them Arena is the only place they exist. The other 38 are the ones that catch people: Nightveil Specter and Loxodon Smiter are uncommons in Arena's Pioneer Masters and rares everywhere else, so neither can lead.

Magic Online does count. On the other side of the same rule, 87 cards qualify through a Magic Online printing alone, among them Hermit Druid and Zombie Master. Both are rares on paper.

Local-language releases do not count. Ball Lightning and Triskelion have uncommon printings in Renaissance, sold only in German, French and Italian. The framework post removed them by name.

Only the front face matters. The committee's rules say so, and they exclude land creatures. Dryad Arbor was an uncommon in Future Sight and still cannot lead. The 11 creature-and-land modal cards like Akoum Warrior are in. The 37 uncommons with a non-creature front and a creature back, such as Invasion of Zendikar, are out.

Uncommon planeswalkers cannot lead. The index holds 21 planeswalkers with a qualifying uncommon printing, and none is on the list. The committee allowed them for one month in January 2025. Its "colourless plus one" experiment was likewise for July 2026 only, so colourless commanders are back to zero colours.

A Background cannot lead alone. The 15 uncommon Backgrounds on the list count only as a second commander, beside a creature that says "choose a Background". Common Backgrounds can still go in the 99, as the committee's FAQ notes.

The Zeta Set is the live one. Secret Lair x MSCHF: The Zeta Set, released September 2, 2026, printed cards at rarities they had never held. For eight eligible cards it is the only uncommon printing that counts, so they are on the list because of it.

• Hornet Queen — Rare or mythic everywhere else. A seven-mana commander.
• Garth One-Eye — Mythic everywhere else, and one of only two five-colour legends on the list.
• Raffine, Scheming Seer — Mythic everywhere else. One of three Esper legends.
• Obeka, Brute Chronologist — Rare or mythic everywhere else. One of two Grixis legends.
• Ruxa, Patient Professor — Rare everywhere else.
• Lord Skitter, Sewer King — Rare everywhere else.
• The Jolly Balloon Man — Rare everywhere else.
• Clown Car — Rare everywhere else, and a Vehicle.

Read literally, the framework post should exclude them, because it says Secret Lairs cannot shift rarity. Scryfall's list includes them anyway, and cPDH Guide reports that the committee decided to admit the whole set, commons and uncommons. We could not find that decision on PDH Home Base itself: neither its rules page nor its announcements mentioned the set when we read them on November 4. Wizards made a similar reversal for 60-card Pauper on September 4, so the direction is plausible. If you are taking one of the eight to an event, ask the organiser first.

Two smaller oddities round this off. Four cards qualify only through Magic Online's Legendary Cube Prize Pack, a product the framework post does not name: Bastion Protector, Nest Invader, Sakura-Tribe Elder and Viscera Seer. And Mardu Hateblade is on the list even though the only printings Scryfall shows for it are commons. We cannot explain that one. It may be a data slip, and a $0.05 card is an odd place to plant a flag, but it is the kind of thing a maintained list can carry.

How many lands real lists play

Neither the rules page nor the committee's rules document gives a land count, so we counted real lists. On November 4, 2026 we read 473 public Pauper Commander decks on Archidekt: the most viewed and the most recently updated. We kept the 431 that are legal on today's data, meaning exactly 100 cards, commanders on the list and every other card legal in the 99. A land here is a card that is a land on its front face, so a spell with a land on its back does not count.

The median list plays 35 lands, and 36 is the most common single count. The middle half of lists sits between 34 and 37.

The curve explains a little of the spread, and only a little. The median list's non-land cards average 2.45 mana. Lists in the cheapest third average 34.5 lands, and lists in the most expensive third average 35.9. That is a real slope but a shallow one, with a correlation of just 0.19.

Colours change which lands a list plays more than how many. Mono-colour and two-colour lists both have a median of 35 lands, while three-colour and five-colour lists sit at 36, on small samples of 16 and 11. The basics tell the story better: the median is 30 basics in a mono-colour list, 25 in a two-colour list and 21 in a three-colour list.

Three caveats. First, these are public lists, not results: the count says what people build, not what wins. Second, the frames differ slightly. The most viewed lists have a median of 35 and the most recently updated have 36. Third, lands are only part of a mana base, and this count ignores ramp entirely. The extremes are real lists too, from 16 lands to one Karolina Dean deck with 97, which we choose to read as a dare.

For what it is worth, the Agent's starting target for PDH is 36 to 38 lands. On this sample that is the upper part of normal rather than the middle: 45% of lists fall in that range and 50% play fewer. Like every role count, it is a prompt for review.

The same sample answers one more question. Of the 431 lists, 288 lead with at least one legendary creature. That is 67% of lists, drawn from the 13% of the pool that is legendary. Players have 4,884 other creatures available and still mostly reach for the card that reads like a commander.

How Manamancer applies all this

Our builder does not run its own version of the rule. It imports Scryfall's list by card when the index refreshes, and the commander slot of a Pauper Commander deck reads that and nothing else. The 99 reads Scryfall's legality field separately and is checked against the commander's colour identity. Arena-only cards stay out of the format's search. A pair is accepted as Partner, Partner with, or a choose-a-Background creature with its Background, and a lone Background is refused.

The consequence is worth saying plainly. Our answer is Scryfall's answer, oddities included: the Zeta Set eight and Mardu Hateblade are all accepted in our builder too. When the list changes, ours follows at the next index refresh. For a finished deck, the format legality guide explains what the builder checks.

There is one gap. The playtest starts every deck on 40 life, whatever the format, and it counts commander damage without watching for 16. When you goldfish a PDH list there, set your life to 30 yourself and keep an eye on that number.

Build a Pauper Commander (PDH) deck — The free deck builder limits the commander slot to this list and checks the 99 against the format. It opens once you sign in.

The short version: if you are choosing a commander, the supply will not stop you unless you want a wedge, and nothing will get you four colours. Everything else is a matter of which of several thousand uncommons you would like to see every game.

May your commander cost less than the sleeve it sits in, and may nobody at the table ask you for a fourth colour.