Pricing Under Volatility: The Trough-Lock Edge

Every quote is a position on the next twelve months of rates. Most desks take that position on experience and last quarter's numbers. I measured the ground under it: air and road freight rates across six EU markets, 2003 to 2024, every finding test-backed.

The headline: air runs about 1.6x road's volatility (coefficient of variation 0.155 against 0.095), and the gap is not a pandemic artifact - it holds at 1.35x with 2020-2022 removed and 1.41x pre-COVID. In a typical year air moves inside a 6.7% band, road 2.8%. Two modes, two pricing problems: validity windows, indexation clauses and surcharge logic sized for one do not fit the other.

Seasonality is tradable. Air climbs about 3.2% into a hard Q2 peak. Across 85 backtested market-years, trough-season locks (Q4 to Q1) beat peak-season locks in 75% of years, average edge 1.8% - Germany 8 of 9. That edge is yours to spend either way: price sharper at the trough to take the award, or hold margin at the peak with the index behind you. Out of sample, the Q2 climb repeated in 10 of 12 market-years in 2024 and 2025.

Since July 2026 the bands run live. A streaming extension of the study scores every rate movement as it arrives and fires a repricing alert when a lane breaks its band. Replayed over 2019-2022, the 2020 air shock triggers exactly where the study's charts show it; replayed over the quiet 2015-2018 window, the alarm stays close to silent. A live requoting signal, not a quarterly PDF.

Know where the rule breaks. In Poland road, not air, is the volatile mode. In France the timing rule fails (44% hit rate on the freight-only series). Per-lane validation comes first; no price hangs on an unvalidated pattern.

Why this exists: twelve-plus years of pricing and tendering on the seller side, plus buyer-side tender platform configuration for 68+ enterprise shippers. I price against the evaluation the buyer actually runs.

The full study with every test sits in a self-built five-tab monitor, live below.