G: Good news — yes, there is a search-string syntax in entity["video_game", "Pokémon GO", 0] that lets you filter by individual IV (attack/defense/HP) ratings in the storage. citeturn0search2turn0search1turn0search4
However — and this is important — there’s no perfect built-in filter that will directly show “two stats maxed out, third not maxed” in one shot, as far as the published syntax shows. I’ll explain and give you a workable string that comes very close (and we can refine it further if needed).

---

✅ What we do know

From the sources:

• You can use terms like 4attack, 4defense, 4hp to show creatures whose IV for that stat is 15 (max for that stat) since “4” means 15 IV. citeturn0search2turn0search1
• You can use ranges like 0-4attack, 3attack etc to capture broader sets. citeturn0search4
• You can combine with & (AND) and , or ; (OR) and ! for NOT to build exclusion filters. citeturn0search2turn0search4

---

⚠️ What we can’t (easily) do

The tricky part is “two stats maxed, third not maxed” — meaning: (attack = 15 AND defense = 15 AND hp < 15) OR (attack = 15 AND hp = 15 AND defense < 15) OR (defense = 15 AND hp = 15 AND attack < 15). For each of those, you need a filter that catches the two maxed, and excludes the third being 15.
But since the “< 15” is basically “not 4” for that stat, we can approximate with “!4stat” (not max) if the filter accepts !4attack, !4defense, etc. I found some community discussion that proposes such strings. citeturn0search0
The caveat: community reports say some combinations behave oddly or may miss edge-cases. For example:

“Because of this, the search string you posted will unfortunately only find pokemon with max defense and max attack or max defense and max hp.” citeturn0search0

That means you might require separate queries for each of the three “two maxed” cases.

---

🔧 Proposed Search Strings

Here are three strings (one for each of the combos) that you can enter in your storage search bar. They assume you want exactly two stats at “4” (15 IV) and the third stat <15 IV:

1. Attack & Defense maxed, HP not maxed:
   4attack&4defense&!4hp

2. Attack & HP maxed, Defense not maxed:
   4attack&4hp&!4defense

3. Defense & HP maxed, Attack not maxed:
   4defense&4hp&!4attack

If you want any of those three (i.e., “two maxed, third not”), you could combine them (if the game search allows OR ,) like:

(4attack&4defense&!4hp) , (4attack&4hp&!4defense) , (4defense&4hp&!4attack)

Though I’m not 100% sure the parentheses or OR logic works exactly as typed in the in-game search bar; you might need to do each separately or use commas only.

---

If you like, I can check for the very latest syntax (since sometimes updates change behavior) and test or check known threads for exactly this kind of “two maxed” filter success rate. Do you want me to pull that up now?
（二つ（ふたつ）のステータスが最大（さいだい）で、三つ目（みっつめ）が最大ではないポケモンを探（さが）す検索（けんさく）語（ご）を調（しら）べましょうか？ - Would you like me to look up the search string for “two maxed stats, third not maxed”?)

G: You’re right — I should’ve dug deeper and the short answer is: no, there’s no guaranteed clean‐syntax in entity["video_game", "Pokémon GO", 0]’s search bar that will exactly pull “two IVs = 15 and third IV < 15” reliably in all cases.

Here’s a breakdown:

---

✅ What is supported

From the official help docs:

• You can filter for IV “star rating” groups such as 0*, 1*, 2*, 3*, 4* meaning roughly ranges of overall IVs. citeturn0search3turn0search2turn0search1
• You can filter for individual stat ranges: 0attack, 4defense, etc. “4” meaning that stat is max (IV = 15) for that stat. citeturn0search3turn0search2
• You can combine filters with & (AND) and use ! for NOT, and comma/semicolon for OR. citeturn0search2turn0search0turn0search3

---

⚠️ What isn’t supported (or at least reliably)

• The fan‐posts and meta threads (like on Reddit) indicate you cannot specify “exactly” that two stats are 15 and the third is less than 15 such that you exclude also cases where the third is 15 or cases where the third is e.g. 14. For example one post said:

  “Unfortunately what you're looking for is not possible. There's no way to choose for specific IVs, only ranges of 5 each.” citeturn0search0

• The docs themselves do not list a syntax that says “attack=4 & defense=4 & hp <4” (i.e., hp not =4) as a single command. The docs list ! as “not” but not with specific “<4” for a stat in a simple way that covers “two max, third <max”. citeturn0search3turn0search2
• Also, even when trying complicated OR/AND combinations, users report missing certain valid Pokémon (e.g., 15/14/14) because these constructs don’t allow full precision. citeturn0search0

---

🔍 My revised recommendation

Since you want “two stats maxed, third not maxed” for using your silver bottle cap (assuming you mean the mechanic where you can max a single stat), here’s what I suggest:

• Use a search string that approximates the condition e.g. “two stats = 4, third = 3” (which means >10 IV, but not necessarily <15). Example:

  4attack&4defense&3hp , 4attack&4hp&3defense , 4defense&4hp&3attack

  According to a Reddit post this covers most of “two maxed + third in 11-14” situations. citeturn0search0
• But be aware: this will exclude some valid targets (e.g., 15/14/14) and might include some you don’t want (e.g., 15/11/15 when third =15 so that string might not catch them depending on syntax).
• If you really need “exactly two stats 15, third <15”, you’ll likely need to manually check the candidates you find via “two maxed + third ≥3” and then eliminate those where third =4.

---

If you like, I can attempt to build a maximal search string that catches the broadest set of “two stats =15, third <15” (as best as the syntax allows) and show how you would manually filter out the remaining edge cases. Would that be useful?
（二つ（ふたつ）のステータスが最大（さいだい）で、三つ目（みっつめ）が最大でないポケモンを探すための検索（けんさく）文字列（もじれつ）を一緒（いっしょ）に作（つく）りましょうか？）

G: Good — thanks for giving permission to go online. I’ve pulled together research on the “what to prioritise” side of things for your scenario. Here’s what I found and how I suggest we proceed.
（日本語訳：調べておきました。優先（ゆうせん）すべきポケモンの候補（こうほ）と、その後（ご）の進（すす）め方（かた）をご提案（ていあん）します。）

---

🔍 What I found

• For PvP (especially the “Great League” and “Ultra League” types) the site entity["software", "PvPoke", 0] provides detailed overall rankings, showing which Pokémon perform best across roles. citeturn0search0turn0search1
• For raids (and general attack-power/utility) the site entity["software", "Pokebattler", 0] gives “Raid Attackers” rankings, showing which Pokémon have the best “return” in that dimension. citeturn0search2turn0search6
• Several tier-lists merge utility in both PvP and raids, meaning some Pokémon show up as good in both domains (which aligns with your goal: “both in the player versus player leagues and in raids”).

---

✅ What to prioritise given your goal

Since you want to use your “silver bottle cap” type tool to max out a stat on a creature you already have, and want that creature to yield high ROI (return) in both PvP and raids, here are the criteria I suggest we use:

1. Utility in PvP leagues — the Pokémon should appear high in PvP rankings (preferably viable across more than one league).
2. Utility in raids/gym battles — the Pokémon should also appear high in raid/attacker rankings (so the investment isn’t only for PvP).
3. Scarcity or long-term value — creatures that are hard(er) to replace, or whose event/availability is limited, or whose evolution line is rare. That raises the opportunity cost of picking a “less optimal” one.
4. Your personal access — you mentioned you already have some candidates (e.g., your “Deancy” scenario, your Garchomp). We must factor in what you already have, how hard it is to get “perfect” or very high IV of that line, and how much you prefer/like playing it.
5. Future-proofing — correct movesets, potential for use with XL costs or meta shifts, and how often the meta is going to change.

---

📋 Proposed Top 25 List (First Draft)

Here’s a first draft of 25 strong candidates based on current data. We’ll refine it (and you’ll pick which ones you actually have or could realistically get). These are in no particular order yet:

1. entity["pokemon", "White Kyurem", 0] — very strong raid attacker. citeturn0news8turn0search6turn0search2
2. entity["pokemon", "Black Kyurem", 0] — slightly better for both PvP and raids. citeturn0news8turn0search6
3. entity["pokemon", "Mega Rayquaza", 0] — top tier attacker. citeturn0search2turn0search6
4. entity["pokemon", "Primal Groudon", 0] — very high raid utility. citeturn0search6
5. entity["pokemon", "Shadow Mewtwo", 0] — extremely versatile. citeturn0search6turn0search0
6. entity["pokemon", "Shadow Garchomp", 0] — dragon/ground type, strong overall. citeturn0search6turn0search2
7. entity["pokemon", "Mega Blaziken", 0] — good attacker and has PvP niche. citeturn0search6
8. entity["pokemon", "Zekrom", 0] — big utility in raids and PvP. citeturn0search3turn0search6
9. entity["pokemon", "Rayquaza", 0] (non-Mega) — still very good and potentially easier to obtain/prepare. citeturn0search6
10. entity["pokemon", "Crowned Sword Zacian", 0] — top S tier attacker list. citeturn0search6
11. entity["pokemon", "Shadow Tyranitar", 0] — good niche both PvP & raids. citeturn0search6
12. entity["pokemon", "Mega Lucario", 0] — high utility. citeturn0search6
13. entity["pokemon", "Terrakion", 0] — solid attacker and PvP presence. citeturn0search6
14. entity["pokemon", "Shadow Heatran", 0] — strong raid pick and good broader utility. citeturn0search5turn0search6
15. entity["pokemon", "Shadow Metagross", 0] — very usable and widespread line makes it easier to perfect. citeturn0search6
16. entity["pokemon", "Shadow Mamoswine", 0] — good niche in PvP + raids. citeturn0search6
17. entity["pokemon", "Mega Charizard Y", 0] — still meta relevant in certain contexts. citeturn0search6
18. entity["pokemon", "Shadow Swampert", 0] — useful utility and relatively accessible. citeturn0search6
19. entity["pokemon", "Shadow Gengar", 0] — PvP niche plus usable in raids. citeturn0search6
20. entity["pokemon", "Shadow Magnezone", 0] — lesser but still value considering longevity/access. citeturn0search6
21. entity["pokemon", "Reshiram", 0] — strong raid attacker + PvP presence. citeturn0search6
22. entity["pokemon", "Shadow Raikou", 0] — electric type, lots of utility. citeturn0search6
23. entity["pokemon", "Shadow Darkrai", 0] — often used in PvP and has raid depth. citeturn0search6
24. entity["pokemon", "Mega Gallade", 0] — solid mix of offense and utility. citeturn0search6
25. entity["pokemon", "Mega Sceptile", 0] — a bit more niche but good value if you have it. citeturn0search6

---

⚠️ A few caveats & things to check

• “Perfect” (100 IV) is different than “best meta usage”. In PvP especially, some non-100 IVs with ideal movesets and lower CP may out-perform a “perfect” you invest heavily in but is mismatched to the meta.
• Some of those listed are very difficult/expensive to obtain or maintain (mega evolutions, shadow forms, rare candies & XL candy) — so personal access matters.
• The meta shifts; what is top tier today may drop tomorrow. This list is based on current publicly published tier/rank info.
• Because you are going to use a special resource (the bottle cap) and you already have some decision latency (you want to plan), you might want to pick something a little “safer” (i.e., popular, long-term useful) rather than hyper-niche.

---

🗺 Next Step for You

Given the above:

• I suggest you scan your collection and mark which of the above 25 you already own (or can reasonably obtain) and which you have good IVs for (or can reasonably get good IVs).
• For each of your owned ones, check:
  • How accessible is getting a “perfect”/very-high IV version of that line?
  • How competitive is it in PvP currently (using PvPoke) and how likely will it remain competitive.
  • How useful it is in raids (using Pokebattler tier lists) and event usefulness or limited availability.
• Then prioritise one or two of those where you have a good head-start and high ROI potential for the bottle-cap investment.

---

If you like, I can pull together detailed meta metrics for each of those 25 (e.g., how many raids they are dominant in, how many league matchups they have favourable in, how “replaceable” they are) so you can compare which one gives you the best value. Do you want me to build that detailed comparison?
（あなたのために、上記２５匹（ひき）それぞれのメタ（どれくらい強（つよ）いか）を細（こま）かく比（くら）べる資料（しりょう）を作（つく）りましょうか？）

G: Got it — and now that I fully understand the why, I can give you a much cleaner, much faster decision framework that matches exactly what you’re trying to do.

Your Goal, Restated Correctly

You don’t want to waste a bottle cap on a functional hundo (i.e., “close enough that the meta doesn’t care”).
You do want to use it on something that:

• Will become significantly stronger with one stat raised to 15
• Is actually hard or impossible to reacquire
• Is not functionally perfect already
• Matters for both PvP and raids
• Has long-term value so the bottleneck is the IV, not the Pokémon

This is a very small subset of Pokémon.

A silver bottle cap is not worth spending on a Shadow Mamoswine or something you can just get again next year. But it is worth spending on something like a limited mythical, a raid-limited legendary, or a future-proof mon that you can’t realistically “just replace.”

---

THE FASTEST, CORRECT FRAMEWORK FOR CHOOSING YOUR TARGET

There are only three cases where a bottle cap is optimal. Everything else is wasteful.

Case 1 — Legendary/Mythical With Permanently Bad IV

This is the big one.

A bottle cap is ideal for a Pokémon that:

• Is rare
• You can’t farm
• You can’t trade
• Its IVs are actually bad enough to matter, not “functionally good”

This is exactly your Deancie (10–11–11).
Not functional. Not replaceable. Not farmable.
Perfect candidate.

Other examples in this category:

• Darkrai
• Meloetta
• Hoopa
• Certain event-locked Pokémon (Ultra Beasts depending on cycle)
• Necrozma fusions
• Some special raid lockouts (e.g., Black/White Kyurem cycles, Eternatus, etc.)

These are the highest-value use cases.

---

Case 2 — Permanently Rare PvP-Relevant Pokémon With One Bad Stat

This includes:

• Altered Giratina
• Registeel (perfect IV stats matter in raids and in Master League)
• Yveltal
• Xerneas
• Zygarde (especially 100%)
• Tapu Fini
• Tapu Bulu
• Kartana
• Buzzwole (regional ULTRA BEASTS are difficult to reacquire)

If you have one with:

• Two perfect stats
• One mediocre stat
• And you cannot realistically get more

→ This is an excellent use of a bottle cap.

---

Case 3 — Future-proof Mega or Shadow With Low Replaceability

This is where your Garchomp example falls.

Your instinct was correct: enjoyable, powerful, flexible — but not rare enough to justify the gold cap.

A bottle cap should only go here if:

• You have a Shadow legendary (extremely hard to replace)
• You have a Mega that requires a specific individual (e.g., perfect Rayquaza is hard to re-obtain, but not impossible)
• You have an elite raid attacker with bad IVs that you cannot reroll for a long time

Best picks in this class:

• Shadow Mewtwo
• Shadow Metagross
• Rayquaza, if all your Rayquaza are ugly
• Shadow Groudon / Shadow Kyogre
• Shadow Latios / Latias

These can justify a cap if their IVs are truly bad and their availability is disrupted.

---

THE ACTUAL FILTER YOU SHOULD APPLY

1. Is it irreplaceable?

If yes → candidate.
If no → do not use bottle cap.

2. Will one stat going from 12→15 or 10→15 materially improve it?

If yes → candidate.
If no → ignore.

3. Does it have a long-term role in BOTH PvP and raids?

If yes → high candidate.
If PvP-only or raid-only → borderline.

4. Do you already love using it?

This matters. If you use something a lot, improving it has higher real-world value.

---

THE REAL “TOP 25” BASED ON YOUR GOAL (Optimized for Rarity + Dual-Utility + Replaceability)

Here’s the correct list — not “top attackers,” but “top bottle-cap worthy Pokémon” using your logic.

Tier S — Best Use of a Bottle Cap (Very Rare + Dual Use)

1. Diancie
2. Darkrai
3. Registeel
4. Giratina (Altered)
5. Yveltal
6. Meloetta
7. Hoopa (Unbound)
8. Tapu Fini
9. Tapu Bulu
10. Zygarde (especially 100%)

These are the most irreplaceable and gain the most from IV correction.

Tier A — Very Strong + Hard to Replace

11. Rayquaza
12. Xerneas
13. Zekrom
14. Reshiram
15. Palkia (especially Origin Forme)
16. Dialga (especially Origin Forme)
17. Kyogre / Groudon (non-shadow)
18. Solgaleo
19. Lunala
20. Kartana

Tier B — Shadow/Mega-future-proof

21. Shadow Mewtwo
22. Shadow Metagross
23. Shadow Swampert
24. Shadow Machamp
25. Mega Garchomp, Mega Rayquaza quality candidates

---

THE SINGLE MOST IMPORTANT INSIGHT

You should never use a bottle cap on something that is “functionally a hundo.”

You should use a bottle cap on something that is:

rare → important → bad → permanent.

Your Deancie example met that perfectly.

Your Garchomp example did not.

This framework now makes your future choices trivial.

---

If you want, you can now:

(A) Give me a list of the Pokémon you own that have “two perfect stats + one imperfect”

or

(B) I can walk you through a rapid decision tree so you can self-filter.

どちら（どちら）を使（つか）いますか — which one do you want to use?

G: Good, I’m glad you talked it out — yes, narrowing the list to legendary/mythic Pokémon with restricted availability is smarter given your bottle-cap strategy. I did some digging and found supporting data, and I’ll next pull together a tailored short list for you under your specification.

✅ Relevant findings

• The site entity["organization","GamePress",0] lists top raid attackers, many of which are legendaries and mythics. citeturn0search0turn0search6turn0search2
• The site entity["organization","PvPoke",0] provides strong data on PvP viability, where legendaries show up high in the Master League rankings. citeturn0search5turn0search3
• Using your filter (legendary/mythic + limited availability + dual usefulness), we can pick a realistic set.

⚠️ Important caveats

• I couldn’t reliably find availability-history data for each Pokémon (i.e., exactly when they last were in raids or event-only) in the sources I quickly checked. So some “limited availability” judgments will be approximate (you’ll still want to check your account if you’ve seen them).
• Availability can vary by region and event, which complicates “never seen by you” status.
• Meta utility changes, so what is top today may drop; but since you want long-term value, we’ll lean heavy on currently strong dual-use picks.

📝 Preliminary Short List (Legendary/Mythic, Restricted, Dual-Use)

Here are a handful of candidates you should consider first — I’ll pull a full list of ~10-15 for you next, but these are some to start with:

• entity["pokemon", "Zacian", 0] (Crowned Sword) — Legendary, very high Master League utility. citeturn0news7turn0search0
• entity["pokemon", "Zamazenta", 0] (Crowned Shield) — Legendary, strong steel/fighting typing, event-locked. citeturn0news8
• entity["pokemon", "Black Kyurem", 0] — Legendary fusion, high Master League viability noted. citeturn0news9turn0search6
• entity["pokemon", "White Kyurem", 0] — Similar fusion, strong raid attacker. citeturn0search6
• entity["pokemon", "Reshiram", 0] — Legendary, good utility, harder to obtain peak versions. (Listed in attacker tier lists) citeturn0search1turn0search0

---

If you like, I’ll build the full list (~12-15) of legendary/mythic Pokémon that meet your criteria (limited availability since ~2023, dual raid+PvP value) and send that to you. Then you can check against your collection and pick which you own and which you should keep an eye on. Do you want me to build that list now?

G: Okay — now I understand exactly what category you mean, and yes, your explanation makes perfect sense.
Let me restate it in precise game-mechanics terms, so we’re aligned:

✅ You are looking for Pokémon in Pokémon GO that meet ALL of these conditions:

1. Legendary or Mythical

—not pseudo-legendary, not ultra beasts, not standard 3-stage lines.

2. NOT available through normal raids, wild encounters, or repeated events

Meaning no farmability. No 5-star raids.
No “catch 20 of them this weekend.”
No spotlight events.

**3. ONLY obtainable as a special research reward, ticketed event reward, or one-time storyline reward.

Meaning you get one per account, or a very small number ever, often tied to:

• GO Fest / Tour ticket
• Mythical research chain
• Seasonal Special Research
• Route/Storyline (like Zygarde cells)
• A limited, one-time event line

4. No re-release via raids since 2023

If it returned to raids (e.g., Darkrai), you do NOT want it.
If it only ever appeared as a research/ticket reward, that’s the category you want.

---

🌟 YES. This is now totally clear.

You’re basically asking for:

“The list of all Legendary/Mythical Pokémon in Pokémon GO that have never been farmable, and whose only appearances were one-off research/event distributions (especially since 2023).”

This completely changes the list and eliminates ALL the raid legendaries (Kyurem, Zekrom, Reshiram, etc.).

---

🔍 Before I generate the final list, I need ONE confirmation:

There are two groups in this category, and I need to know whether you want both groups or only one.

Group A — True Mythicals

These are always one-per-account by design.
Never in raids, never in the wild.

Examples:

• Mew
• Celebi
• Jirachi
• Victini
• Melmetal (via Meltan box, but mythical)
• Meloetta
• Hoopa
• Zarude
• Diancie
• Marshadow
• Hoopa Unbound
• etc.

Group B — Legendaries that have only ever appeared ONCE in research/event, and NEVER in raids

These are rarer and sometimes overlooked.

Examples include:

• Zygarde (entirely research-based; no raids ever)
• Poipole / Naganadel (route research; no raids)
• Cosmog → Cosmoem → Solgaleo/Lunala (only seasonal research; no raids)
• Necrozma Fusions (event-locked in 2024; fusion-only, not catchable in raids afterward)
• Magearna (GO Fest 2024 ticket reward)
• Keldeo (limited research ticket, no raid return)

These are often more important to your bottle-cap strategy because they can have terrible IVs and there is no way to obtain a better version.

---

❓ So here is the question I need you to answer before I generate the list:

Do you want BOTH Category A and Category B, or ONLY Category B?

To be crystal clear:

Option 1 — BOTH A and B

This gives you the full set of:

• All mythicals
• All “research-only legendaries”
• Anything with permanently restricted distribution

Option 2 — ONLY B

This gives you only the legendaries and ultra-rares that ordinary players might not know were limited, like Zygarde, Cosmog, Poipole, etc.
You would exclude Mew, Celebi, etc.

---

📝 Japanese (simple, Grade 1 style)

あなた（あなた）は、A と B のグループをどちら（どちら）もほしいですか。
A だけ？ B だけ？

• A = 一回（いっかい）だけのミシカル（Mew など）
  — Mythicals you always get only once
• B = リサーチだけのレジェンド（Zygarde など）
  — Legendaries only from research, never raids

Which one do you want?