Working prototype · public freight indices + Mattel's own disclosures · ocean-freight margin early warning

Freight Bridge: what a shipping shock does to the holiday build, before the print.

Mattel builds inventory through Q3 to feed the holiday quarter, and locks ocean-freight rates months ahead of knowing the bill. Freight Bridge takes a public container-rate shock and reads it straight onto next year's landed cost and gross margin, in operator terms: a 1-99 severity score, a colour band, and the basis points at risk.

A Mattel supply leader put the 2021 shock on the record as "just a shipping tax." Fair, so let's compute the tax. Pick a scenario: Freight Bridge moves a freight index against Mattel's disclosed sourcing and cost structure, separates the temporary freight tax from the structural margin question, and reports one number you can take into an order-pacing meeting. Every anchor is cited; the monthly curve between anchors is synthetic and labelled as such.

01 · Run a freight scenario

One click, one verdict.

Each preset moves the ocean-freight index Mattel pays on its Asia-to-US holiday inventory. The verdict is computed live against Mattel's disclosed China-sourcing share and the disclosed weight of ocean freight + resin in its cost base. Sliders are optional: open them only if you want to tune the assumptions.

...

Mattel · holiday landed cost

...

...

...

...

...

Operator moves at this band

    Margin bridge: base year to scenario year

    ...

    The freight tax, decomposed

    Where the basis points come from, and which part is temporary

    Freight index, calibrated to cited anchors

    Drewry WCI composite, $/40ft · ▲ = scenario shock

    Tune the assumptions optional · sliders

    Presets set these for you. Move a slider and the verdict, bridge, and decomposition all recompute. Defaults are Mattel's disclosed values where one exists; the elasticity and freight-cost share are labelled assumptions.

    02 · The brief

    How the number is built, and where it's honest about not knowing.

    The model, the calibration anchors, and the honest limits click to expand

    What Freight Bridge computes

    It is a landed-cost bridge, not a forecast model. A freight-index shock raises the per-unit ocean-freight cost on the Asia-sourced share of goods; that lands as a cost-of-goods increase; price and mix recover part of it within the year; the residual hits gross margin in basis points. The math is deliberately legible: you could redo it on a napkin, and the "Tune" panel lets you.

    InputDefaultWhere it comes from
    Freight-index shockper scenarioDrewry WCI composite anchors: 2019 avg $1,420/40ft, Sept-2021 peak $10,377 (≈ +630% vs 2019), 2026 ≈ $3,433. The "2021 again" preset uses the cited peak; others scale off it.
    Asia-sourced share50%Mattel disclosed China sourcing ≈ 50% in 2024, targeted under 40% (2025) and under 15% (2026); the slider lets you walk the rebalance.
    Ocean freight share of landed cost7%Assumption. Mattel disclosed ocean freight and resin together at ≈ 15-20% of cost base in Q3 2021; ocean alone is not separately disclosed, so 7% is a labelled mid-point, not a quote.
    Price/mix recovery35%Assumption. Pricing elasticity is not public; shown as a labelled lever. Production replaces it with Mattel's realised price/mix.

    Bands: green / CLEAR (severity 1-33, <~80bp at risk) · amber / WATCH (34-66) · red / ACT (67-99, a 2021-scale hit). The score is a monotone map of the basis points at risk; the band thresholds are fixed, the inputs are yours.

    Why freight, and why now in the calendar

    Mattel's own 10-K says it builds inventory through the first three quarters for the holiday peak, and that it contracts ocean-freight capacity and rates well in advance. So the freight bill on the holiday build is committed months before the margin shows up, which is exactly the window where a cheap public-index read buys operator lead time. This sits one layer over an inventory-glut radar: inventory tells you how much you over-ordered; freight tells you what the boat cost to bring it, and the two compound in the same quarter.

    Honest limits

    • The monthly freight curve is synthetic. Only the anchor points are cited (2019 avg, Sept-2021 peak, Aug-2023 −72% YoY, Jun-2026 level). The smooth path between them is an interpolation tuned to those anchors and is drawn dashed and labelled "synthetic"; it is not a claimed monthly series. A production build pulls the real weekly Drewry/FBX print.
    • Ocean-freight share is an assumption, not a disclosure. Mattel discloses freight + resin together. The 7% default is a labelled mid-point; the verdict moves with it, on purpose, so the buyer can see how much the answer depends on a number only Mattel knows.
    • No tariff layer here. This is the freight tax only. Tariffs and the sourcing-rebalance cost premium are a separate bridge; folding them in without Mattel's internal numbers would be guessing.
    • Elasticity is a lever, not a finding. The price/mix recovery slider is honest about being unknown; it is the first thing a real engagement would replace with realised data.
    • This is not affiliated with Mattel. Every Mattel figure is from a public filing, earnings call, or press report cited below; nothing here uses non-public data.

    03 · Sources & method

    Every anchor, every disclosure.