@pkmn/spreads

Library for parsing, displaying, and converting Pokémon stats, spreads, and ranges.

Installation

$ npm install @pkmn/spreads

Alternatively, as detailed below, if you are using @pkmn/spreads in the browser and want a convenient way to get started, simply depend on a transpiled and minified version via unpkg:

<script src="https://unpkg.com/@pkmn/spreads"></script>

Usage

This library provides utilities for dealing with a Pokémon's Spread (Nature/EVs/IVs) and the computed Stats that result from combining a spread with a Pokémon's base stats data. Additionally, @pkmn/spreads contains logic for dealing with a Range of Stats (StatsRange) or Spread (SpreadRange). Typically, an opponent Pokémon's exact stats are unknown, but the range of possible values it could be can be tracked and gradually winnowed down. It is generally recommend that programs work with Stats and StatsRange, but humans tend to understand Spread and SpreadRange better. which makes the latter pair more suitable for display purposes. Stats must also be converted into a Spread in order to fill in a PokemonSet.

import {Stats, StatsRange, Spreads} from '@pkmn/spreads';

const gen = 3;
const base = Stats.fromString('60/65/60/130/75/110');

const range = StatsRange.fromBase(base, gen);
const spread = Spread.fromString('EVs: 172/0-/0/148/0/188+');
const stats = spread.toStats(base, gen);

range.min = Spread.fromString('EVs: 100/0/0/60/0/40\nIVs: 10 Def').toStats(base, gen);
range.max = stats;

console.log(range.toSpreadRange(base, gen).toString());

The above example code will display the following:

Serious-Timid Nature
EVs: 100-172 HP / 60-148 SpA / 40-188 Spe
IVs: >9 Def

@pkmn/gmd and @pkmn/predictor make extensive use of this package and can provide helpful examples for how these classes are intended to be used.

Limitations

The algorithm used to reverse Stats into a Spread (which is extensively documented via comments in the source code) attempts to efficiently find the least constrained, optimal spread that still produces the same stats (minimizing use of EVs and keeping a neutral Nature where possible). Logic elsewhere within EPOké wants the flexibility to then further optimize the reversed Spread if possible. However, there are limitations with what the algorithm is able to do, and the Spread returned by it is not guaranteed to be legal, just that it computes to the correct Stats. IV and EV limits are observed, but the following is considered out-of-scope:

• The IVs of the Spread returned are not guaranteed to produce the same Hidden Power type or base power (the algorithm prefers to deduct from EVs as opposed to IVs, meaning its more likely that a Hidden Power set will be noticeable from the EVs and not the IVs).
• The algorithm is not aware of a Pokémon's gender, so the Attack DVs of the Spread returned in Gen 2 are not guaranteed to be accurate.
• The algorithm does make a best-effort attempt to ensure the HP DV is correct in RBY and GSC, but considers it too computationally expensive to attempt to guarantee the HP DV will always be the expected value based on the other DVs.
• The returned Spread will often be sub-maximal (< 508 EVs) when in reality it is more likely that EVs were maxed out and IVs were sub-maximal instead. This is mostly a display issue, other than the impact on Hidden Power as outlined above.
• The algorithm pretends that the 655 Stat 'Glitch' overflow scenario does not exist. This is only relevant in non-standard formats.

Browser

The recommended way of using @pkmn/spreads in a web browser is to configure your bundler (Webpack, Rollup, Parcel, etc) to minimize it and package it with the rest of your application. If you do not use a bundler, a minified index.umd.js bundle is included in the package. You simply need to depend on ./node_modules/@pkmn/spreads/build/index.umd.js in a script tag (which is what the unpkg shortcut above is doing), after which spreads will be accessible as a global:

<script src="./node_modules/@pkmn/spreads/build/index.umd.js"></script>

License

This package is distributed under the terms of the MIT License.