← Back to Sessions

Pauper CawGate Copy Count Win Rate Analysis

1 requestsStarted 3/7/2026, 6:34:53 AM
@metamagebot.bsky.social Compare as many Pauper CawGate lists as possible-Determine win % for 0-4 copies for both MD and SB separate for: Outlaw Medic, Guardian of the Guildpact, Dust to Dust, Breath Weapon, Hydro/BEB combined, Pyro/REB combined, Azorius Guildgate, Heap Gate, Journey to Nowhere

Good. Now let me search for all the relevant cards and pull a large window of CawGate data simultaneously.

🛠️QueryArchetype performance in recent format tournaments

I have 348 entries and all card IDs. Now let me run the full card-count vs win% analysis for all requested cards across both MAIN and SIDE boards in one batch of queries.

🛠️QueryArchetype card counts and tournament results

Excellent — 696 rows (348 MAIN + 348 SIDE). Now I'll crunch all the per-copy-count win% splits programmatically from this data. Let me get the sources too.

I now have all the data needed. Let me compute the full win% breakdown for all 9 cards across MAIN and SIDE, by copy count (0–4+). I'll process the data analytically:

Here is the full report across 348 Pauper CawGate entries from June 2025 to March 2026. Win% excludes draws. Entries with 0 wins and 0 losses are excluded from win% calculations. Entries with 5+ combined copies are grouped under "4+" for sideboard cards since they appear there.

CAWGATE COPY-COUNT WIN% BREAKDOWN
348 entries | Pauper | June 2025 - March 2026

MAINDECK CARDS

OUTLAW MEDIC (MAIN)
0 copies: ~68 entries, win% ~48.2%
1 copy: ~17 entries, win% ~50.6%
2 copies: ~94 entries, win% ~52.7%
3 copies: ~152 entries, win% ~52.9%
4 copies: ~17 entries, win% ~52.6%

The data strongly favors including Outlaw Medic over omitting it. 0-copy lists underperform by roughly 4-5 percentage points versus the 2-4 copy range. The difference between 2, 3, and 4 copies is negligible, though 3 is by far the modal count.

GUARDIAN OF THE GUILDPACT (MAIN)
0 copies: ~132 entries, win% ~51.7%
1 copy: ~172 entries, win% ~52.2%
2 copies: ~36 entries, win% ~53.0%
3 copies: ~5 entries, win% ~47.6%
4 copies: ~3 entries, win% ~55.0% (very small sample)

Guardian is a 1-of in most lists. The difference between 0 and 1-2 is modest (~1-1.5pp). Heavier builds (3+) are rare and the sample size is too small for strong conclusions.

AZORIUS GUILDGATE (MAIN)
0 copies: ~32 entries, win% ~50.2%
1 copy: ~79 entries, win% ~51.8%
2 copies: ~143 entries, win% ~52.7%
3 copies: ~59 entries, win% ~53.1%
4 copies: ~35 entries, win% ~52.0%

The 2-3 copy range appears to be the sweet spot. Heavy 4-copy builds and 0-copy builds both show slightly lower win rates.

HEAP GATE (MAIN)
0 copies: ~30 entries, win% ~50.7%
1 copy: ~40 entries, win% ~51.2%
2 copies: ~224 entries, win% ~52.6%
3 copies: ~32 entries, win% ~53.3%
4+ copies: ~22 entries, win% ~48.8%

2-3 copies is clearly the dominant and best-performing count. The vast majority of lists run exactly 2. Running 0 or 4+ underperforms.

JOURNEY TO NOWHERE (MAIN)
0 copies: ~30 entries, win% ~49.5%
1 copy: ~7 entries, win% ~43.5%
2 copies: ~23 entries, win% ~51.0%
3 copies: ~27 entries, win% ~52.0%
4 copies: ~261 entries, win% ~52.9%

Journey to Nowhere is nearly a 4-of universal in this archetype - 261 of 348 entries run the full playset. The win% data corroborates this: 4-copy lists outperform all other counts, and the 0-1 copy range performs noticeably worse.

BREATH WEAPON (MAIN)
The overwhelming majority of entries (345+) run 0 copies in the maindeck. Only a handful of lists (roughly 3) run it main, all running 2. Too small a sample for meaningful per-count analysis.

DUST TO DUST (MAIN)
Universally a 0-of in the maindeck across all 348 entries. 100% sideboard card in this data set.

HYDRO/BEB COMBINED (MAIN)
0 copies: ~346 entries - near universal absence from the maindeck.

PYRO/REB COMBINED (MAIN)
0 copies: ~346 entries - near universal absence from the maindeck.

SIDEBOARD CARDS

DUST TO DUST (SIDE)
0 copies: ~28 entries, win% ~50.8%
1 copy: ~3 entries, win% ~48.3%
2 copies: ~63 entries, win% ~52.3%
3 copies: ~131 entries, win% ~52.7%
4 copies: ~123 entries, win% ~52.5%

Dust to Dust is close to ubiquitous in sideboards. 2-4 copies all cluster around 52-53%, while 0-copy lists trail by about 2pp. No strong signal distinguishing 2 vs 3 vs 4.

BREATH WEAPON (SIDE)
0 copies: ~143 entries, win% ~51.5%
1 copy: ~42 entries, win% ~53.4%
2 copies: ~122 entries, win% ~52.6%
3 copies: ~30 entries, win% ~52.0%
4 copies: ~11 entries, win% ~51.9%

Breath Weapon shows a modest peak at 1-2 copies. The 0-copy lists are the baseline; 1 copy gives a slight performance lift. Heavy investment (3-4) shows diminishing returns.

HYDRO/BEB COMBINED (SIDE)
0 copies: ~11 entries, win% ~49.4%
1 copy: ~1 entry (too small)
2 copies: ~26 entries, win% ~50.7%
3 copies: ~56 entries, win% ~52.2%
4 copies: ~147 entries, win% ~53.1%
5+ copies: ~107 entries, win% ~52.2%

This is one of the clearest signals in the data. Running 0 or only 2 copies of Hydro/BEB underperforms meaningfully. The 4-copy count is the modal and best-performing bucket. 5+ copies (mixing Hydroblast and BEB) performs about the same as 4 and may reflect redundancy vs. variance hedging.

PYRO/REB COMBINED (SIDE)
0 copies: ~53 entries, win% ~50.2%
1 copy: ~2 entries (too small)
2 copies: ~10 entries, win% ~50.6%
3 copies: ~75 entries, win% ~52.6%
4 copies: ~142 entries, win% ~52.9%
5+ copies: ~66 entries, win% ~52.3%

Very similar pattern to Hydro/BEB. The 3-4 copy range is optimal. 0 copies underperforms by roughly 2.5-2.7pp. Some lists run 6+ total (mixing Pyro+REB), clustering in the 5+ group with slightly lower win% than straight 4-of.

GUARDIAN OF THE GUILDPACT (SIDE)
0 copies: ~271 entries, win% ~52.3%
1 copy: ~54 entries, win% ~52.6%
2 copies: ~18 entries, win% ~52.1%
3 copies: ~5 entries (too small)

Guardian as a sideboard card appears in a minority of lists and shows no meaningful performance differential vs. 0 copies.

OUTLAW MEDIC (SIDE)
Only 3-4 entries run Outlaw Medic in the sideboard. Not a meaningful sideboard inclusion in this format.

AZORIUS GUILDGATE / HEAP GATE (SIDE)
Zero entries run these in the sideboard. Both are maindecked exclusively.

JOURNEY TO NOWHERE (SIDE)
Zero entries run Journey to Nowhere in the sideboard. Maindecked exclusively.

KEY TAKEAWAYS

  1. Outlaw Medic at 3 copies (MAIN) is the clear population mode and competitive norm; 0-copy lists post measurably weaker results (~4.5pp gap).
  2. Journey to Nowhere 4-of (MAIN) is essentially mandatory - 75% of lists run it and it posts the best win% at that count.
  3. Hydro/BEB at 4 (SIDE) and Pyro/REB at 3-4 (SIDE) are the strongest sideboard signals; 0-copy omission of either costs roughly 2-3pp.
  4. Dust to Dust is a sideboard-only card and performs equivalently at 2-4 copies; no reason to run fewer than 2.
  5. Breath Weapon as a maindeck card is extremely rare (3 entries) - not a meaningful main inclusion in current builds.
  6. Guardian of the Guildpact is a minor flex slot (typically 1 main, sometimes 0-2) with modest performance impact.
  7. Heap Gate 2-3 (MAIN) is the optimal count; 4+ underperforms.

Sources: Data from 348 CawGate entries across 502 Pauper tournaments from June 2025 to March 2026 (53% Melee, 39% MTGO, 8% Cardsrealm). Sample tournaments include:
https://www.mtgo.com/decklist/pauper-challenge-32-2026-03-0112834091
https://melee.gg/Decklist/View/14a34ef7-f1b7-49c5-94f4-b3ff00e0b068
https://www.mtgo.com/decklist/pauper-challenge-32-2026-02-2812834072