Why the Traditional KDA Formula Rewards the Wrong Kind of Excellence

The standard KDA ratio is deceptively simple: add kills and assists, then divide by deaths. On paper, this formula seems fair because it already includes assists. In practice, it treats every kill and every assist as equal units, even though the final blow is often the least demanding part of a coordinated takedown. A support may land a crucial stun, exhaust the enemy assassin, block a dash with a body, and sacrifice half a health bar to set up a kill. The carry then takes the gold, the kill feed, and the broadcast spotlight. The support receives one assist, which is mathematically useful but culturally invisible. This is where the KDA debate intensifies. The formula does not merely describe performance; it shapes narratives. Casters praise high kill counts, fans clip solo kills, and MVP voters often lean toward the player at the top of the damage chart rather than the player who enabled every objective. A 10/2/4 mid laner may post a 7.0 KDA, while a 1/3/18 support may post a 6.3 KDA. One looks like a star, the other like a role player, even if the support controlled vision, peeled for the carry, and started every winning fight. Deaths also carry an outsized psychological weight. Because deaths sit in the denominator, a single late-game death can erase a flawless early game, encouraging players to protect their ratio instead of making the aggressive play that wins the match. In some games, assists are only credited if the player dealt damage or applied crowd control within a short window, which means a support’s earlier setup can be erased by a timer. The problem is not that assists are ignored. The problem is that the scoreboard rewards the wrong kind of excellence.

The Support Player’s Case: How Assists Shape Every Winning Play

Supports, tanks, and initiators live in the assist column, but their most important work often happens before any assist is registered. Vision control, cooldown tracking, wave management, zoning, and peel do not always produce a number. A well-timed Lulu Polymorph or Thresh Flay can save a carry without appearing in KDA at all. When an assist is recorded, it usually means the player participated in a kill through damage, crowd control, or a debuff within a certain window. Yet the most valuable support play may be the engage that forces three enemies to burn flash, allowing a teammate to clean up later. By raw KDA logic, that sacrifice can look like a death; by teamfight logic, it is the reason the fight was won. Support mains argue that assists should count more because the current system still treats them as consolation prizes. In a five-player kill, the assist is shared, diluted, and easily forgotten. In a solo kill, the kill is celebrated as individual brilliance even when a support set it up. In Dota 2, assist gold helps supports scale; in Overwatch, assist-like eliminations blur the line further. A weighted assist system would change incentives. It would reward peeling for carries, warding before objectives, and initiating at the right moment. It would also reduce kill stealing, because taking a kill from a scaling carry may improve a personal ratio while weakening the team. If assists counted more, the support who goes 0/4/24 would no longer be described as having a poor score. That player would be recognized as the engine of the team’s success. The case for assists is not about giving everyone a trophy. It is about acknowledging that in team games, the player who creates the opportunity is often as valuable as the player who finishes it.

Role-Aware Metrics: A Fairer Way to Measure Teamfight Impact

A fairer system does not need to abolish kills or deaths. It needs to become role-aware. In League of Legends, Dota 2, Overwatch, and Valorant, the value of a kill depends on who gets it, when it happens, and what objective follows. An assassin’s kill on the enemy carry before a Baron fight may be worth more than a support’s kill on a fleeing tank after the fight is already won. Similarly, an assist from a healer who kept the initiator alive may be worth more than an assist from a player who landed one stray bullet. A weighted KDA could assign coefficients by role, champion class, or match context. Supports might receive 1.2 or 1.3 points per assist, while hard carries receive 1.0. Deaths could be weighted by gold swing, objective pressure, and time of game. But complexity brings risks. If players cannot understand the formula, they will distrust it. If the formula can be gamed, they will exploit it. A support could farm meaningless assists by hovering near fights without providing real value. A carry could avoid necessary deaths and lose the game while protecting a weighted ratio. Developers would need transparent breakdowns: assist quality, kill participation, vision score, objective damage, crowd-control duration, and save metrics. A better model might show traditional KDA alongside weighted assist contribution and objective participation percentage. The goal is not to crown supports as the only true stars. The goal is to stop pretending that a final hit is always the most important event in a teamfight. Role-aware metrics would not end the debate, but they would move it away from a single number and toward a richer picture of contribution. That picture is harder to summarize, but it is much harder to fake.

What a Weighted Assist System Could Change for Pros, Ranked, and Fans

If assists counted more, the effects would ripple through professional contracts, ranked ladders, and fan culture. Scouts already use advanced data, but public perception still leans on KDA. Free agents with strong kill participation but modest kill counts are often undervalued. A weighted assist model could give supports and tanks stronger negotiating positions, because their contributions would appear in a language owners and fans already understand. In ranked play, players might stop flaming supports for a “bad KDA” and start recognizing high assist participation as a sign of competence. Riot could adjust post-game honors, MVP voting, and even matchmaking ratings to reflect assisted plays. Valve could do the same in Dota 2, where support players often control the tempo through smoke ganks, saves, and vision. A fairer metric could also make role queue healthier, since fewer players would feel punished for choosing utility roles. But there are dangers. A system that overweights assists could encourage passive players to farm assists without creating advantages. It could also spark endless arguments about whether a 0/4/20 support truly outperformed a 12/5/6 carry. The answer is rarely absolute. A 12/5/6 carry may have supplied the damage; a 0/4/20 support may have supplied the vision, peel, and engage. The healthiest outcome is not a single new formula but a richer scoreboard. KDA should remain a quick reference, not a final judgment. If assists count more, they should count more because team games are won by chains of action, not by one player’s last hit. The debate intensifies because the stakes are real: recognition, money, and the meaning of skill itself. A scoreboard that better rewards assists would not make kills meaningless. It would make teamwork visible.