On a recent episode of The Investing for Beginners Podcast, value investor Daniel Levy reframed a question that trips up nearly every retail investor: when is a high P/E actually expensive? His answer flips the math. A P/E of 50 equals an earnings yield of just 2%, which sounds punishing. But that yield is only one side of the equation. “If it’s growing earnings at 20% and you have a 2% earnings yield, your return is still pretty good,” Levy said. The trap is paying 50x for a 5% grower. There is no cushion for disappointment.
That single mental model explains why two of the most discussed stocks on the NASDAQ are. That includes NVIDIA (NASDAQ: NVDA | NVDA Price Prediction) and Tesla (NASDAQ: TSLA). Both of which can carry “expensive-looking” multiples and end up in radically different places on the risk spectrum.
NVIDIA: A High Multiple That Compresses Quickly
NVIDIA trades at a trailing P/E of 31x and a forward P/E of 23x, with a PEG ratio of 0.63. The headline multiple is well below the 50x threshold Levy uses as a stress test. And the growth side of the equation is doing the heavy lifting. In its most recent quarter, NVIDIA posted $1.87 in non-GAAP EPS against an estimate of $1.77, with revenue of $81.62 billion, up 85.2% year over year, and net income up 210.6%. EPS has climbed in a straight line from $0.81 in Q1 FY2025 to $1.87 in Q1 FY2026.
This is the case Levy describes as cushioned. A 3% earnings yield paired with triple-digit profit growth means the multiple compresses fast, even with no price movement. CEO Jensen Huang framed the runway on the call: “The buildout of AI factories, the largest infrastructure expansion in human history, is accelerating at extraordinary speed.” Investors who want to verify the underlying numbers can read the Q1 FY2027 filing on SEC.gov. Shares are up 41.7% over the past year to $211.45.
Tesla: The Multiple Levy Warns About
Tesla is the opposite picture. The stock trades at a trailing P/E of 369x and a forward P/E of 196x, with a PEG ratio of 5.6. Flipping the multiple yields an earnings yield well below 1%. Levy’s stress test asks whether growth justifies it. Quarterly earnings grew 8.3% year over year on 15.8% revenue growth, and TTM EPS sits at $1.10. Q1 FY2026 EPS of $0.41 is well off the $1.19 quarterly peak from Q4 2022.
That is the asymmetry Levy describes. The reported numbers do not match a 369x multiple unless investors are paying for terminal value: Cybercab, the Tesla Semi, FSD licensing, and Optimus. Polymarket assigns just a 3% probability to a California robotaxi launch by June 30 and a 16.5% probability that Optimus ships by year-end. Bulls can point to the bright spots: FSD active subscriptions reached 1.28 million, up 51% year over year, and auto gross margin expanded to 21.1%. Tesla shares are up 27.36% over the last year to $409.94, but down 9.63% year to date.
The Takeaway From Levy’s Framework
Levy’s rule is mechanical. Convert the P/E into an earnings yield, then ask honestly whether the company’s compounding rate gives you a margin of safety if growth disappoints. A 3% yield against 85% revenue growth, as in NVIDIA’s case, leaves room to be partly wrong. A sub-1% yield against high-single-digit earnings growth, as in Tesla’s case, requires faith in optionality measured in decades. The pairing of multiple and growth rate is the entire game.