Champion scaling in LoL: how base stats grow per level

Every champion starts at level 1 with base stats and gains “growth” per level — but not linearly. Riot’s formula makes each level give slightly more stat than the last, which explains why some champions feel weak early and monstrous at level 18.

Updated:

Riot’s growth formula

The vast majority of a champion’s base stats (health, mana, attack damage, armor, magic resist, health and mana regen) use one and the same formula to scale with level:

stat(level) = base + growth × (level − 1) × (0.7025 + 0.0175 × (level − 1))

At level 1 the second term is 0, so you have exactly the base. From there, each level adds growth multiplied by a factor that grows with level. That factor (0.7025 + 0.0175 × (level − 1)) starts at 0.7025 and rises to 0.7025 + 0.0175 × 17 = 1 at level 18.

In other words: by the time you hit level 18, you’ve received the equivalent of 17 full “growth” values accumulated, but distributed in an increasing way — the first levels give less than the nominal growth and the last ones give more. Don’t memorize the formula: the site’s scaling tool graphs it for you with each champion’s real values.

stat(nivel) = base + growth × (nivel − 1) × ( 0.7025 + 0.0175 × (nivel − 1) )
The factor is ~0.72 at level 2 and exactly 1.0 at level 18. The scaling Lab plots it.
A stat per level (example)
600Level1995Level61478Level112300Level18

Example: 600 base + 100 growth, with Riot’s real formula. The per-level gain grows (the curve accelerates).

Why the curve accelerates at the end

The trick is in the factor (0.7025 + 0.0175 × (level − 1)). The per-level increment isn’t constant: each level-up gives you growth × (0.685 + 0.035 × (level − 1)). Going from level 1 to 2 gives about 0.685 × growth; going from 17 to 18 gives almost double, 1.245 × growth.

That’s why the curve is a gentle upward parabola, not a straight line. The total stat grows in an accelerating way: the last levels are worth quite a bit more than the first ones in base stats. This rewards closing out long games with high-growth champions.

A handy mental average: the factor starts at 0.7025 and ends at 1, averaging about 0.85. So stacking 17 levels gives roughly growth × 17 × 0.85 ≈ 14.5 × growth of extra stat at 18 if you eyeball it, though the exact figure is the formula above.

Since all linear stats use the same curve, a champion with high armor growth, for instance, doesn’t “take off” suddenly: they gain defensive advantage gradually and increasingly, level by level.

The exception: attack speed

Attack speed does not follow the parabolic formula. It scales linearly off base attack speed, by a growth percentage per level: AS(level) = baseAS × (1 + growthPercent × (level − 1)).

That’s why attack speed growth is expressed as a percentage and applied as an additive bonus, just like the attack speed from items or runes. A champion with 2.5% AS growth gains that 2.5% of their base AS each level, at a constant rate.

This difference matters when comparing auto-attackers: two champions with the same base AS but different growth percentages diverge linearly, not in an accelerating way. And since item AS adds to this base as a bonus, you should look at total AS at the relevant level, not just the base.

In the site’s scaling tool, attack speed appears with its own linear curve, separated from the other stats, so you don’t confuse it with the parabolic growth.

Comparing early vs late with no items

To judge who wins a duel “raw” (no items), it’s not enough to look at level-1 base stats or level-18 ones: you have to look at the exact level of the fight. An early-game champion usually has high bases and modest growth, so they dominate the first levels but lose relative strength as others scale.

A late-game champion usually has low bases and high growth. The parabolic curve favors them: since the last levels give more stat per the formula, their early disadvantage flips around level 16-18, especially if their kit also scales with items.

Practical trick: compute both champions’ stats at the same level and compare them. For example, two champions with the same base health but growth of 90 vs 110, at level 11 (factor ≈ 0.8775) differ by (110 − 90) × 10 × 0.8775 ≈ 176 health — a real gap that decides who wins the trade.

The site’s scaling tool lets you overlay two champions and see, level by level, where their health, damage or resistance curves cross. Combine it with the champion pages, the tier list and the power spikes guide to understand why a champion “rises” or “falls” in different phases of the game. Where each level comes from is in levels and experience.

Scaling: early-game vs late-game champion
Early (high base, low growth)Late (low base, high growth)
Level1Level6Level11Level18crossover · Level11

Attack damage per level for two archetypes, with Riot’s real formula. The early champion hits harder until the crossover level; after that the late champion overtakes it. Knowing where that crossover is tells you when to force and when to wait.

FAQ

Do stats rise by the same amount each level?

No. Because of Riot’s formula, the per-level increment grows with level: going from 1 to 2 gives less stat than going from 17 to 18 (almost double). That’s why the curve accelerates and the last levels are worth more in base stats. Attack speed is the exception: it rises linearly.

Why does attack speed scale differently?

Because it uses a separate linear formula: AS(level) = baseAS × (1 + growthPercent × (level − 1)). The growth is a percentage applied as an additive bonus, just like AS from items and runes, instead of following the parabolic curve of the other stats.

How do I compare an early-game champion with a late-game one, no items?

Compute both champions’ base stats at the same level with the growth formula and see who wins at that point. The early-game one usually has high bases and dominates early; the late-game one, with high growth, catches up and overtakes near 16-18 thanks to the curve. The site’s scaling tool graphs it for you.