The pop is not the exit: 13 of Forge's 20 recent VC-backed debuts sit below issue price

The number Forge's July update does not print is the one that matters to anyone who bought Figma on screen rather than in the book. From the first-day close of $115.50 down to $18.09 at the end of Q2 2026 is a fall of 84.3% ^[12]. The 45.2% decline Forge reports ^[4] is measured from the $33 issue price ^[3], which is a price almost nobody outside the allocation ever paid. The 250% first-day return ^[3] was not a return for the public market. It was the entry price.

Seven of twenty is a 35% hit rate, which leaves 13 names below issue ^[11]. And the seven is a mark as of June 30 ^[1]. SpaceX was one of them, up 26.6% at $170.86 on that date after pricing at $135 ^[5]. Two weeks later it closed at $136.08 ^[6], a 20.4% give-back ^[13] that left it $1.08, or 0.8%, above where it priced ^[14]. Forge says the company is reportedly due to report Q2 earnings in early August ^[7]. A bad print and the count of winners is six.

Forge's own framing is that this is a selective market rather than a closed one, and that the dispersion argues for diversified late-stage private exposure across companies, themes and vintages ^[17][9]. The first half is what the data says. The second half is where the report recommends the thing its own numbers complicate: Forge notes that SpaceX's Forge Price rose materially through the private period, with notable acceleration before the listing ^[16]. Acceleration into a listing is the point at which the private entry price stops being a discount to the public one. The report is right that entry valuation and portfolio construction remain critical ^[9], and that cuts at the pre-IPO round bought at the accelerated mark exactly as it cuts at the aftermarket buyer of the pop.

Scale is the part that travels. Forge puts Anthropic and OpenAI approaching $1 trillion, behind only SpaceX's $1.25 trillion private valuation ^[8]. Those marks are being set in a venue where the holder sets the reference price, and then tested in one where 65% of the last cohort broke issue ^[11]. Forge's stated lesson is that strong IPO demand does not necessarily translate into durable public performance ^[10]. The corollary it leaves for the reader is that a private mark carries even less information about the exit than a first-day close does, and the first-day close cost Figma's aftermarket buyers 84.3% ^[12].