How to Use U.GG Data to Optimize Your Item Builds and Win Rate

U.GG processes millions of ranked game records from Riot's public API every patch, aggregating item purchase data, win rates, and pick rates for every champion across all roles. The raw data includes games from every rank tier, but the displayed statistics are filtered by default to show Platinum and above — a threshold where champion mastery and decision-making are sufficiently developed that the build data reflects intentional choices rather than random item purchases by players who are still learning the game.

Patch filtering is one of the most important features U.GG offers. The site defaults to the current patch, which means you are seeing build data only from games played on the most recent version of the game. After a significant patch with item or champion changes, this filter keeps you from running builds that were optimal two weeks ago but are now inferior because of number adjustments. Always check that the patch displayed in the top right of the U.GG champion page matches the current live patch before trusting the data.

The rank filter dramatically changes which builds appear. Challenger-filtered data on a champion like Aphelios looks completely different from Platinum-filtered data because high-ranked players have mastered matchup-specific adjustments that lower-ranked players have not yet developed. For most players climbing below Diamond, the Platinum-through-Diamond filter provides the most actionable data — it reflects builds used by players who are winning consistently but not making hyper-specialized adjustments that require hundreds of games of matchup experience to execute correctly.

U.GG's item win-rate charts show the statistical win rate of a champion when they purchase a specific item, broken down by item slot (first item, second item, and so on). A common misreading of this data is treating high win-rate items as universally strong. In reality, some items appear to have high win rates partly because they are purchased by players who are already winning — you tend to buy your third and fourth items when you are ahead and the game goes long. Items with inflated late-game win rates are not necessarily the cause of the win; they are correlated with games that lasted long enough to buy them.

The most reliable signal in U.GG's item charts is the first-item win rate for items with at least 1,000 game samples. This data reflects a choice made early in the game when builds diverge most meaningfully, and the sample size is large enough to reduce statistical noise. When Trinity Force shows a 54% first-item win rate on Irelia with 50,000 games in the sample, that is a robust signal. When an obscure item shows a 58% win rate but only 200 games, the sample is too small to trust — that is likely noise or a niche strategy that does not generalize.

The item frequency column — which shows what percentage of players bought each item — is equally important to the win-rate column. An item with 70% win rate but only 2% pick rate is a niche situational pick that few players use and that probably requires specific game states to be effective. An item with 55% win rate and 85% pick rate is the consensus optimal choice and the safest default. When learning a new champion, prioritize the high-frequency high-win-rate items over statistical outliers, even if the outlier looks impressive in isolation.

Starter items on U.GG are sorted by the combination of starting item and first back purchase, giving you a build path rather than just a single item. This is meaningful because some mythic items have multiple viable component paths. For example, Trinity Force can be built through Phage first (for the early health and damage) or Caulfield's Warhammer first (for the ability haste). U.GG's data shows which starting path wins more games on each specific champion, allowing you to make an informed choice rather than defaulting to the path you are most familiar with.

Long Sword versus Doran's Blade is one of the most common starter dilemmas for AD carries and bruisers. U.GG data consistently shows that Doran's Blade wins more games for most ADCs because the sustain it provides — 8% omnivamp on hit — lets you stay in lane through poke-heavy exchanges that would otherwise force you to back early. Long Sword starter paths are typically better for AD assassins who intend to complete an early Serrated Dirk and do not need the lane sustain that Doran's Blade provides.

Corrupting Potion, Doran's Ring, and Doran's Blade represent three distinct mid-lane starting philosophies. U.GG data shows Corrupting Potion winning in lanes where the enemy can out-trade you early and you need the health and mana restoration to survive past level six. Doran's Ring wins in lanes where you have an early kill threat and want to maximize your damage output. Doran's Blade on mages like Sylas reflects a specific playstyle built around auto-attack trades. Always check the starter win rates specific to your champion rather than assuming the default is optimal.

Mythic items define your champion's primary combat identity and provide a passive bonus to all your legendary items. Trinity Force provides bonus health, AD, and attack speed to each legendary item you own — scaling well with bruisers who stack multiple combat items. Kraken Slayer provides true damage on every third hit through its mythic passive and grants bonus AD to legendaries, making it the premier anti-tank ADC mythic. Understanding what your mythic scales with tells you which legendary items to prioritize in your full build.

Luden's Tempest versus Shadowflame is a common mythic choice debate for burst mages. U.GG data shows Luden's winning more often on poke champions like Ezreal mid or Xerath who want the movement speed bonus and consistent poke damage, while Shadowflame outperforms on burst assassin mages like Syndra or Veigar who need high single-target damage against targets with shields. The raw AP values are similar, but the passive conditions favor different playstyles entirely.

Galeforce versus Immortal Shieldbow represents the ADC mythic split between aggressive and defensive options. Galeforce's active dash is a mobility and assassination tool used on ADCs who position aggressively — Jhin, Miss Fortune, and certain Caitlyn matchups. Immortal Shieldbow provides a life-saving shield when you drop below 30% health, making it the standard safety pick for carries who expect to be dove or who are playing from behind. U.GG data often shows their win rates within 1-2% of each other, meaning the choice is highly matchup-dependent rather than a universal hierarchy.

Situational items are where U.GG data requires the most careful interpretation. The site shows which items are purchased frequently across all games, but situational items by definition are not purchased in all games — they appear in specific enemy team compositions. Mortal Reminder's win rate looks modest in aggregate because it is only purchased against healing-heavy compositions, but its win rate in those specific games is dramatically higher. U.GG's "situational items" tab separates these context-dependent purchases from core build items.

Grievous Wounds items — Mortal Reminder for AD carries, Chemtech Putrifier for AP supports, Chempunk Chainsword for AD bruisers — should appear in your build any time the enemy team has two or more significant healing sources. Soraka, Yuumi, Aatrox, and Sylas all represent major healing threats. Without Grievous Wounds reducing their healing by 40%, fights often go to the enemy team purely because of the sustain differential. U.GG data confirms this: in games where the enemy has heavy healing and you purchase Mortal Reminder as your third item, win rates climb significantly.

Armor penetration items like Lord Dominik's Regards require specific conditions to outperform other legendary items. U.GG shows Lord Dominik's win rate peak in games against at least two tanks, which is where the bonus damage against high-health targets reaches its maximum value. Against squishy compositions, Black Cleaver or Infinity Edge often outperform it. Check the enemy composition before your fourth and fifth item purchase to confirm whether the armor pen stack is actually needed or whether raw damage items would yield higher effective damage output.

The optimal U.GG build represents the statistically strongest choice in aggregate, but individual game states create specific conditions where deviations are correct. The most common deviation is building a defensive item earlier than the optimal build recommends when you are behind and need to survive. If you are 0-3 as Jinx and the enemy Zed is 4-1, building Immortal Shieldbow as your second item rather than Kraken Slayer is correct even if U.GG shows Kraken Slayer has a higher win rate — you need the shield to stay alive long enough to reach your power spike.

Ability haste stacking is another legitimate deviation from standard builds. On champions like Zed, Talon, or Katarina who deal damage through repeated ability usage, building two or three ability haste items can outperform the standard mythic-plus-AD build in games that go to 35 minutes or beyond. U.GG data is patch-averaged, and specific compositions where you need to reset abilities more frequently may favor higher CDR even when the aggregate data does not reflect it.

The most important rule for build deviation is that you should be able to articulate why you are deviating. "I want to try something different" is not a valid reason. "Their team has three tanks and I need Lord Dominik's as my third item instead of fourth to deal relevant damage in the mid-game teamfights" is a valid reason. Build improvisation without a clear rationale is almost always worse than following the aggregate-optimal path, especially below Diamond where the sample size justifying a deviation is unlikely to match what pro players and high-Elo theorycrafters have already tested.

U.GG's skill order section is one of the most underused features on the site. It shows which ability players max first, second, and third across the entire filtered sample, along with the win rate of each maxing pattern. For most champions, the max-first ability is obvious — Jinx maxes Q for attack speed, Zed maxes Q for poke damage. But on flex champions like Kennen, Karma, or Twisted Fate, where multiple abilities have reasonable max-first arguments, the win-rate data cuts through the theorycrafting debate with empirical evidence.

Level one ability selection is also tracked on U.GG. This matters most for jungle champions, where the starting ability determines your clear speed and early game vulnerability. Lee Sin players who start E versus Q have meaningfully different early game power profiles, and U.GG data shows which start produces higher win rates in aggregate. For laners, the level one choice is often between a damage ability and a mobility or defensive ability, and the correct answer depends on whether you are the aggressive or passive side of the matchup.

Skill order deviations from the aggregate are sometimes correct based on matchup. Taking one rank of W early on Caitlyn to activate the headshot interaction, or taking an early point in Vayne's Tumble instead of Silver Bolts in a matchup where kiting ability matters more than sustained DPS, are examples of matchup-specific adjustments. However, these deviations should be based on a specific mechanical interaction you have already practiced, not a vague intuition. If you are not certain why you are deviating from the U.GG skill order, default to the statistically optimal path.

Professional and high-Challenger players often run builds that diverge from U.GG's aggregate data because they are optimizing for conditions that do not exist at lower ranks — coordinated team compositions, specific strategic setups, or champion pool dynamics that require itemization flexibility. Cross-referencing U.GG aggregate data with Probuilds.gg or U.GG's Challenger-filtered builds helps you identify when high-level players consistently deviate from the aggregate, which is a signal that the aggregate may be suboptimal under specific conditions.

A common pattern is that pro players and Challengers sometimes build more ability haste than the aggregate recommends on mid laners, reflecting a preference for rotation speed over raw damage output. They can execute on the value of lower cooldowns better than lower-ranked players, making the haste more valuable for them. When you see a consistent deviation in Challenger-filtered data, ask yourself whether you have the mechanics and decision-making to actually extract that value before copying the build wholesale.

The safest and most productive use of U.GG for players below Diamond is to establish a default build from the aggregate data and then make two or three game-specific adjustments per game based on enemy composition. Master the core build for your primary champion to the point where you do not have to think about it during the game — that mental bandwidth should go toward in-game decision-making. Reserve build research for off-sessions between games, not during champion select when you are already managing bans, team composition, and rune selection simultaneously.