PvP IVs Explained: Why a 0/15/15 Can Beat Your Hundo

Great League and Ultra League do not limit your Pokemon's level or stats. They limit one number: CP, at 1,500 and 2,500. So the question is never "which IVs are highest" but "which IVs pack the most stats into 1,500 CP" — and the CP formula does not charge the same price for the three stats.

CP = floor( Attack × √Defense × √Stamina × CPM² ÷ 10 )

Each stat is the species' base stat plus the IV (0 to 15), and CPM is a multiplier that grows with level. Look at where the square roots are. Attack goes in at full value; Defense and Stamina go in under a root. Raise Attack by 10% and CP rises by 10%. Raise Defense by 10% and CP rises by a little under 5%.

Under a cap, that makes every point of Attack cost about twice as much CP as a point of Defense or HP. A spread with low Attack leaves CP headroom, and the only way to use that headroom is to power up further — which raises all three stats at once. That is the whole trick. Zero Attack is not good in itself; it is good because it lets the same Pokemon be several levels higher when it touches the cap.

To compare two IV spreads of the same species you need a single number. The community standard is stat product: Attack × Defense × HP, taken at the highest level that still fits under the cap, with HP rounded down the way the game does it. There are 16 × 16 × 16 = 4,096 possible spreads per species, and "rank 1" just means the one with the highest stat product for that league.

Two limits are worth knowing before you trust it too far:

• It compares spreads within a species. It says nothing about whether the species is any good. That is what the Great League, Ultra League and Master League tier lists are for.
• It is a measure of total bulk and damage, not a battle simulation. It does not know your moves or your opponent. Treat it as "how much Pokemon did I fit under the cap", nothing more.

Take a perfect 15/15/15 Azumarill. Its stats are 127 / 167 / 240, and at level 40 (CPM 0.7903) the formula gives 127 × √167 × √240 × 0.7903² ÷ 10 = 1,588 CP. That is over the cap, so the hundo has to stop earlier: the highest level that fits is level 36, at 1,497 CP.

Now a 0/15/15. Its stats are 112 / 167 / 240. Less Attack means less CP at every level, so it keeps fitting under 1,500 for much longer: all the way to level 45.5, at 1,499 CP. Same species, same cap, nine and a half levels higher.

Attack, Defense and HP are the real in-battle stats at that level (base plus IV, times CPM). Read the first and last rows together: the 0/15/15 gives up 6% of its Attack and gets 6.6% more Defense and 12 more HP in return. Multiply the three and it comes out 6.7% ahead on stat product. Of the 4,096 possible Azumarill, 0/15/15 is first; the hundo is 2,550th, in the bottom half.

The levels in the table come from the game's full CPM table, half levels included, checking every level until the CP passes 1,500. For your own Pokemon, the IV checker turns the CP and the appraisal bars into exact IVs and the current level, and lists its CP, HP, Attack and Defense at the milestone levels.

Look at the scale of that table again. The gap between the best Azumarill that exists and a hundo is under 7%. The worst of all 4,096 spreads, a 15/0/2, still reaches 89.6% of the best. IVs are a finishing touch on a PvP Pokemon, not the foundation. The species, its moves and a second charged move decide far more than the last few percent of bulk.

And "close to the best" is a big group. For Azumarill, 179 spreads are within 2% of rank 1 and 558 are within 3%. You do not need the one perfect Pokemon; you need any of several hundred.

A keep rule you can apply on the catch screen

• Attack 0 to 5, Defense and HP both 10 or higher: keep it. For most Great and Ultra League species this lands within a few percent of the best, as the 5/15/15 and 2/14/15 rows show.
• Attack 6 to 9 with high Defense and HP: usable. Keep it until something better turns up, and do not feel bad about building it if the species is hard to find.
• Attack 10 or more: this is a raid or Master League Pokemon, not a Great League one. Do not transfer it for that reason alone — just do not spend Great League Stardust on it while a better spread is plausible.

The mistake in both directions is the same: treating rank as pass or fail. Transferring a 4/13/14 because it is "only rank 150" throws away a Pokemon that plays identically in almost every game. Spending a week hunting rank 1 of a common species costs you nothing but time; spending 100,000 Stardust to rebuild a Pokemon for 1% more bulk costs you a second charged move on something else.

Master League

No cap, so nothing is gained by holding Attack down. You want 15/15/15 at level 50, or level 51 with the Best Buddy boost. Every IV point is simply more stats.

Species that cannot reach the cap

Low Attack only helps if the cap is what stops you. If a species runs out of levels first, the best spread is the biggest one. Mantine is a clean example in Ultra League: even a 15/15/15 at level 50 is 2,383 CP, short of 2,500, so its best Ultra League IVs are the hundo.

Species that need XL Candy to get there

Some picks only approach the cap beyond level 40. A 15/15/15 Medicham is 1,431 CP at level 40 and a 15/15/15 Registeel is 2,447 CP, so without XL Candy the highest IVs you own are the best you can field, and with XL Candy the usual low-Attack logic returns. Going from level 40 to 50 costs 296 XL Candy and 250,000 Stardust in full, and XL Candy only starts dropping at trainer level 31. Budget for that before you commit to one of these; the level 40 to 50 guide covers whether it is worth it.

The small print

A handful of species are deliberately run with a little more Attack than rank 1 because the extra damage tips a specific matchup. That is a species-by-species call made by people who simulate those matchups, and it is a refinement for players already winning with a rank 50 Pokemon. It is not a reason to keep high-Attack spreads in general.

You cannot get a 0-Attack Pokemon from most of the places you get Pokemon. The game sets an IV floor per encounter type, and the floor applies to all three stats, Attack included.

Three practical consequences:

• Community Day is for volume, not for weather. Hundreds of ordinary wild catches of one species is the best low-Attack lottery the game offers. The weather-boosted ones are the worse tickets.
• Trading rerolls IVs. A raid legendary is stuck at 10 Attack or more when you catch it, but traded to a friend its IVs are rolled again from that friendship level's floor. That is the only route to a low-Attack legendary for Ultra League. It is a Special Trade, so plan which friend and which day; see lucky and special trades.
• A Lucky Pokemon is the wrong shape for a capped league. With a floor of 12 it always carries expensive Attack. Its half-price power-ups belong in Master League.

No floor is given here for GO Battle League reward encounters. Niantic widened their IV range in March 2025 and current sources contradict each other on the minimum.

IVs do not change when a Pokemon evolves, so the time to judge a PvP candidate is while it is still a cheap base form. Evolving and then discovering the CP jumped past 1,500 is the most common way Stardust and Candy get wasted.

1. Read the exact IVs. The appraisal bars are enough: put them and the CP into the IV checker to get the three values and the current level.
2. Check the species is worth it in the tier list for the league you are building for. A rank 1 spread of a Pokemon nobody uses is still a Pokemon nobody uses.
3. Check the evolved CP fits. If the evolution would already be over the cap at its current level, that individual is out for that league, whatever its IVs. High-level wild catches of strong species fail this test often; the lower-level catch is the keeper.
4. Deal with Shadow questions first. Purifying adds 2 to each IV, Attack included, and cannot be undone. Decide that with the Shadow guide before any Stardust goes in.
5. Then spend. Evolve, power up to just under the cap, and buy the second charged move. If Stardust is tight, the Stardust priorities guide puts this purchase in order with everything else competing for it.

Tag the keepers as you go so they survive the next storage purge; the clean-up routine has the search strings for finding PvP candidates by CP range.