Size concentrated liquidity range
Concentrated liquidity earns fees only while price is inside the chosen range; a narrower range raises the fee APR but goes out of range faster and enlarges the divergence loss at the edge.
Procedure
Price to tick: tick = log(price_quote_per_base) / log(1.0001), floored to a multiple of the pool's tick spacing (60 = 0.3% pools, 10 = 0.05%, 1 = 0.01%). Inverse: price = 1.0001^tick.
Convert bounds to sqrt-price X96: sqrtPriceX96 = sqrt(price) * 2^96, widening to the nearest usable tick at or outside the target so the range never inverts.
Capital efficiency versus full range = 1 / (1 - sqrt(Pa/Pb)). For ETH at 2,000 with range [1,800, 2,200]:
python3 -c "import math; Pa,Pb=1800,2200; print(1/(1-math.sqrt(Pa/Pb)))"
15.62...
So ~15.6x the fee-earning depth of a full-range position for the same capital.
Token split: amount0 = L*(1/sqrtP - 1/sqrtPb) (base, active below the range) and amount1 = L*(sqrtP - sqrtPa) (quote). Choose L so the two amounts equal your deposit.
In-range divergence loss is bounded by the range: exiting below Pa leaves the position 100% base, and the loss versus holding equals the range width. Recompute the loss at both edges before depositing.
Fee APR, not APY: fee_apr = volume_24h * fee_tier * your_share / your_capital. Out-of-range time earns zero, so multiply by the expected in-range fraction.
Pitfalls
- Forgetting the position converts to 100% of the underperforming asset at each edge — a tight range is a leveraged mean-reversion bet.
- Setting bounds on the wrong-side price: Uniswap prices token1/token0, and the base/quote order flips when token0 is the stable.
- Ignoring tick-spacing rounding: a range "set" at 1,999 rounds to 1,980 and widens your risk.
Verification
cast call $POOL "slot0()(uint160,int24,uint16,uint16,uint16,uint8,bool)" --rpc-url $RPC
Take return[0] as sqrtPriceX96, compute price = (sqrtX96/2^96)^2, and confirm the current tick lies inside [tickLower, tickUpper].
Report the range, capital efficiency, token split, and fee APR with the slot0 read.
1---2name: "0xklen-size-concentrated-liquidity-range"3description: "Use when providing Uniswap v3-style concentrated liquidity: choosing ticks, converting a price range to sqrt-price bounds, and computing capital efficiency and in-range divergence loss of the position."4metadata:5 source: https://skillmd.com/skills/jiayaoqijia/0xklen-size-concentrated-liquidity-range6---78# Size concentrated liquidity range910Concentrated liquidity earns fees only while price is inside the chosen range; a narrower range raises the fee APR but goes out of range faster and enlarges the divergence loss at the edge.1112## Procedure13141. Price to tick: `tick = log(price_quote_per_base) / log(1.0001)`, floored to a multiple of the pool's tick spacing (60 = 0.3% pools, 10 = 0.05%, 1 = 0.01%). Inverse: `price = 1.0001^tick`.15162. Convert bounds to sqrt-price X96: `sqrtPriceX96 = sqrt(price) * 2^96`, widening to the nearest usable tick at or outside the target so the range never inverts.17183. Capital efficiency versus full range = `1 / (1 - sqrt(Pa/Pb))`. For ETH at 2,000 with range [1,800, 2,200]:1920 python3 -c "import math; Pa,Pb=1800,2200; print(1/(1-math.sqrt(Pa/Pb)))"21 15.62...2223 So ~15.6x the fee-earning depth of a full-range position for the same capital.24254. Token split: `amount0 = L*(1/sqrtP - 1/sqrtPb)` (base, active below the range) and `amount1 = L*(sqrtP - sqrtPa)` (quote). Choose `L` so the two amounts equal your deposit.26275. In-range divergence loss is bounded by the range: exiting below `Pa` leaves the position 100% base, and the loss versus holding equals the range width. Recompute the loss at both edges before depositing.28296. Fee APR, not APY: `fee_apr = volume_24h * fee_tier * your_share / your_capital`. Out-of-range time earns zero, so multiply by the expected in-range fraction.3031## Pitfalls3233- Forgetting the position converts to 100% of the underperforming asset at each edge — a tight range is a leveraged mean-reversion bet.34- Setting bounds on the wrong-side price: Uniswap prices token1/token0, and the base/quote order flips when token0 is the stable.35- Ignoring tick-spacing rounding: a range "set" at 1,999 rounds to 1,980 and widens your risk.3637## Verification3839 cast call $POOL "slot0()(uint160,int24,uint16,uint16,uint16,uint8,bool)" --rpc-url $RPC4041Take return[0] as sqrtPriceX96, compute `price = (sqrtX96/2^96)^2`, and confirm the current tick lies inside [tickLower, tickUpper].4243Report the range, capital efficiency, token split, and fee APR with the slot0 read.