TLDR: If your list is finalized, order exactly 100 deck cards. For an unsettled extra cards commander proxy order, add only options tied to a real decision: usually 3–5 flex cards for a few contested slots or 6–10 targeted test cards during active tuning. Keep alternate treatments, tokens, and other helper pieces separate from the active deck.
A Commander deck can become a minor logistics project surprisingly quickly. The solution is not to add a random stack of “maybe useful someday” cards. Start with one commander and 99 other cards, then give every optional card a specific job: test a planned swap, provide alternate art, support gameplay, or prepare for a confirmed upgrade.
The baseline is a complete 100-card Commander deck
Commander decks contain 100 cards: the commander plus 99 other cards. The format generally uses singleton construction, apart from basic lands and cards whose rules explicitly allow additional copies. Every card must also comply with the commander's color identity. Wizards provides the current fundamentals on its official Commander format page.
That makes 100 the clean default order quantity for deck cards. Commander Proxies does not require you to add optional cards, and a locked list does not become more complete at 105. Before ordering, verify the exact 100 using the audit in our guide to turning a list into a finished Commander proxy deck.
Duplicates of ordinary nonbasic cards can still be useful as alternatives, but they are not simultaneous additions to the singleton deck. If you order two different versions of a removal spell, label one as an option and choose which copy belongs in the active 100 before playing.
How many extra cards should a Commander proxy order include?
| Status of your list | Suggested extras | Best use |
|---|---|---|
| Final and tested | 0 | Order the exact commander plus 99 |
| A few slots remain contested | 3–5 flex cards | Compare direct alternatives for specific slots |
| The deck is being actively tuned | 6–10 test cards | Test a focused package or several identified swaps |
| You want different treatments | Only the versions you expect to use | Choose between readable alternate-art or frame options |
| A release may change the list | 0–3 confirmed candidates | Prepare only for cards with a clear intended replacement |
These ranges are planning guidance, not format rules or a vendor requirement. A focused group of four cards is often more useful than 20 unrelated possibilities because each option has an identified route into the deck. If you cannot name the card that would leave, the prospective addition probably belongs on the maybeboard for now.
Use flex cards for genuinely contested slots
Flex cards are best when the deck is essentially finished but two approaches remain plausible. A graveyard deck might be deciding between another protection spell, a repeatable recursion piece, and graveyard interaction. A token deck might compare a payoff against another board-protection effect. Three to five alternatives are usually enough to resolve those decisions without creating a second decklist in the box.
Choose flex cards by role, not merely by enthusiasm. Each one should answer a question such as:
- Does this deck need one more cheap interaction spell?
- Is the expensive top-end threat too slow for the intended pod?
- Would another land or low-cost ramp piece improve early turns?
- Does the deck need protection more often than another value engine?
- Is a flavorful card worth the performance tradeoff at this table?
Write the intended swap beside each option in your deck notes. “Card A replaces Card B if the curve feels heavy” is actionable. “Card A seems neat” is how a deck box acquires its own junk drawer.
Add test cards as a package, not a pile
Six to ten extras make sense when you are actively tuning several connected slots. The key word is connected. If you want to test a stronger artifact package, for example, include the enablers and payoffs being considered along with a list of what comes out. Testing one high-synergy card without the support it needs may tell you very little.
A useful test package might contain two pieces of interaction, two ramp alternatives, two synergy payoffs, and two possible cuts for an overly expensive top end. That gives you several controlled configurations while keeping the active deck recognizable. Change a small group, play under the same expectations, and record which version you used.
Power expectations matter here. The official Commander page presents Brackets and the Game Changers list as tools for discussing the kind of game players expect. If your test package adds fast mana, efficient tutors, compact combos, or similarly consequential cards, discuss the resulting change with the pod. Our guide to building a Commander Brackets proxy deck covers that pregame fit in more detail.
Alternate art is a presentation choice
Ordering two treatments of the same card can be worthwhile when you have not settled on a visual direction or when one version may be easier to read. Keep the alternate outside the active deck and swap versions between games rather than shuffling both into a singleton list.
Readability should beat novelty. Check the card name, mana cost, type line, power and toughness where applicable, and the rules information you will need during play. Highly stylized art can look excellent in a gallery view while becoming a tiny comprehension exam from across the table.
If visual consistency is the real goal, make the decision at the deck level. Matching basics, related frame treatments, and a deliberate token package generally produce a more coherent result than ordering scattered alternates. See these cohesive Commander deck themes for ways to develop a unified treatment without sacrificing table function.
Tokens and helper pieces are separate from the 100
Tokens, emblems, counters, reminder cards, and double-faced-card aids solve gameplay problems, but they are not flex slots in the deck. Commander Proxies’ finished-deck guidance likewise distinguishes the active 100 from helper pieces and recommends checking inclusions and readability before ordering.
Audit the list for every object the deck routinely creates. Prioritize frequently used tokens, especially when multiple creatures have different stats, colors, abilities, or creature types. A generic marker may be fine for a token that appears once every ten games; it is less helpful when the board contains six distinct objects and everyone is trying to remember which one flies.
Keep helper pieces in a separate sleeve color, divider section, or token box. The physical distinction prevents a token or optional swap from being accidentally shuffled into the library and turning the next deck count into an archaeological dig.
Be cautious with cards from upcoming releases
Do not order speculative upgrades merely because a preview looks vaguely compatible with your commander. Card details, release timing, legality, and the surrounding list can change before you are ready to use it. Wait unless the candidate is confirmed and you can identify both its purpose and the card it would replace.
For a serious candidate, use a simple three-part test:
- Confirm the final card information and that its color identity fits the deck.
- Name the exact slot or package it would replace.
- Decide whether it improves the deck you want to play rather than merely raising its theoretical ceiling.
Adding one to three confirmed candidates can be sensible when an order must be placed before you have time to test. Adding every on-theme preview usually produces expensive clutter. A digital maybeboard is the better home for unresolved speculation.
Run this audit before placing the order
- Confirm one commander plus exactly 99 other active deck cards.
- Check color identity and singleton compliance, including any card that explicitly permits multiple copies.
- Remove accidental maybeboard entries and duplicate imports.
- Verify basic-land quantities rather than assuming an export handled them correctly.
- Label every flex or test card with its intended replacement.
- Separate alternate treatments from active deck copies.
- Count required tokens, emblems, counters, and double-faced-card aids separately.
- Choose treatments with readable names, mana costs, types, and game text.
- Discuss meaningful power changes with the intended playgroup.
- Save a dated copy of the final decklist and optional-card list.
Decklist imports deserve particular attention because maybeboards, categories, commanders, and tokens may be represented differently by different tools. If your list begins on Moxfield, follow the checks in our guide to ordering a full Commander proxy deck from a Moxfield list before treating the exported quantity as final.
Keep the active deck and extras easy to identify
Use a divider, separate compartment, contrasting outer sleeve, or labeled packet for optional cards. Maintain one authoritative decklist showing the current 100 and a second short list for alternatives. When you make a swap, update both records immediately.
This organization also makes table communication simpler. You can describe the deck being played now rather than every possible configuration sitting beside it. The five spicy alternatives in the deck box do not define the current deck until you put them into the active 100.
When you are ready to turn the final list and its deliberately chosen options into physical game pieces, you can use a service that prints MTG proxy and playtest cards. Keep any unofficial cards clearly understood as playtest pieces, and confirm that your group or venue accepts them before playing.
Casual permission does not equal sanctioned-event permission
Proxy expectations vary among playgroups and stores, so ask rather than assuming. Explain that the cards are unofficial, describe the deck’s intended power, and use versions that other players can read without repeatedly borrowing your phone.
Sanctioned Magic events have a different boundary. The Magic Tournament Rules prohibit non-genuine cards except for limited circumstances involving judge-issued proxies under the rules. Check the current Magic Tournament Rules and ask the event organizer if event eligibility matters. Optional extras do not change that policy.
Order extras only when they answer a specific question
The best default is simple: order the exact 100 when the list is finished. Add 3–5 flex cards when a few slots remain contested, or 6–10 targeted test cards when you are tuning a defined package. Treat alternate art and helper pieces as separate categories, and avoid speculative upcoming cards without identified swaps.
Before ordering, write a one-line purpose beside every extra. If each card has a credible path into the deck or a clear job beside it, the package is useful. If not, leave it on the digital maybeboard until the next revision. Your deck box already has enough responsibilities.