FOURWEEKMBA EXPLAINER · FROM THE NEWSROOM · UPDATED 12 NOVEMBER 2026
In short: a forward P/E ratio divides a share price by the earnings per share analysts expect, not the earnings already reported. When earnings are forecast to grow, the forward multiple is lower than the trailing one. At 10:19 a.m. ET on 11 November 2026, Nvidia was 11.0 times next year’s consensus, Broadcom 20.1 times and AMD 33.5 times, by our arithmetic. Market numbers on this page move every trading day.
Free to share with a link to this page. Price: Nasdaq last sale, 10:19 a.m. ET, 11 November 2026. Forecasts: Nasdaq analyst consensus, read the same day. Trailing earnings: Nvidia’s GAAP diluted EPS for the four quarters to 26 July 2026, from its 10-K and 10-Q. All ratios are our arithmetic.
Trailing P/E uses the last four reported quarters. It is fact, but it looks backward.
Forward P/E uses a forecast, usually the analyst consensus for the current or next fiscal year.
Next-12-month (NTM) P/E uses forecasts for the coming twelve months, which can span two fiscal years.
The same Nvidia price gives three different answers:
Measure
Earnings used
Nvidia multiple
Source
Trailing
GAAP diluted EPS, four quarters to 26 Jul 2026: $7.91
21.8x
Nvidia 10-K and 10-Q; our arithmetic
Forward, next fiscal year
Consensus for year to Jan 2028: $15.75
11.0x
Nasdaq consensus; our arithmetic
Forward, NTM
FactSet next-12-month consensus
about 11.7x
TradingKey, citing FactSet, as reported 11 Nov 2026
Price for the first two rows: $172.68 at 10:19 a.m. ET, 11 November 2026. We did not read FactSet’s data; the NTM figure is TradingKey’s account of it. The trailing row is GAAP; Nasdaq’s consensus carries no GAAP or non-GAAP label.
Why the forward multiple falls when earnings grow
The price is the same in every column; only the earnings change. If analysts expect earnings to rise, the denominator gets bigger and the ratio gets smaller.
Nvidia’s consensus rises from $9.30 a share for the year ending January 2027 to $15.75 for the year after, 69% growth by our arithmetic. So the multiple drops from 18.6 times to 11.0 times on the same price. Further out it falls to 7.9 times for the year ending January 2029 and 6.5 times for 2030.
But the further out you go, the thinner the forecast. Nvidia’s four yearly figures rest on 17, 15, 9 and 3 analyst estimates on Nasdaq’s page. A 6.5 times multiple built on three estimates says little.
Worked example: Nvidia, Broadcom and AMD
All three prices are Nasdaq last sales at 10:19 a.m. ET on 11 November 2026: Nvidia $172.68, Broadcom $350.41, AMD $537.74.
Price divided by Nasdaq consensus EPS for each company’s current and next fiscal year. Nvidia’s years end in January, Broadcom’s in October, AMD’s in December. Ratios and growth rates are our arithmetic.
Nvidia: $9.30 this year and $15.75 next. 18.6 times, then 11.0 times.
Broadcom: $10.37 for the year to October 2026 and $17.46 for 2027. 33.8 times, then 20.1 times.
AMD: $6.93 for 2026 and $16.04 for 2027. 77.6 times, then 33.5 times.
Growth alone does not explain the gap. Nvidia’s and Broadcom’s consensus earnings grow at almost the same rate into next year, 69% and 68%, yet Broadcom’s next-year multiple is 1.8 times Nvidia’s, by our arithmetic. The difference is what the market will pay for each dollar of forecast, which reflects how much it trusts the forecast and what risks it prices on top.
From multiple to market value. Multiply the $15.75 by a multiple and you get a share price; multiply by Nvidia’s 24.1 billion shares (its 10-Q count) and you get a value. At Broadcom’s 19.0 times on 10 November closes, Nvidia would be worth about $7.2 trillion, against about $4.16 trillion at 11.0 times. Each turn of multiple is worth about $380 billion, by our arithmetic. That is a what-if, not a forecast (the full arithmetic).
Pitfalls
Always ask “which year?” On 11 November Nvidia was 18.6, 11.0 or 7.9 times depending on the fiscal year used. A quoted multiple without its year is incomplete.
The consensus is a range. Nvidia’s 15 estimates for the year to January 2028 ran from $13.96 to $17.04. At $172.68 that is 12.4 times to 10.1 times, by our arithmetic.
The forecast can move, not just the price. TradingKey wrote on 11 November that Nvidia’s “stock price has recently pulled back sharply, but analysts have not yet lowered their profit forecasts for the next 12 months in tandem”. If the forecast falls, the multiple rises with no change in price.
The price moves every day. On the 10 November close of $168.89, Nvidia was 10.7 times next year; at $172.68 the next morning, 11.0 times; at $158.78 after hours on 11 November (Yahoo Finance, thin trading), 10.1 times, by our arithmetic.
GAAP or non-GAAP? Nvidia’s second quarter was $2.46 a share on GAAP and $2.22 non-GAAP. Nasdaq’s consensus does not say which basis it uses, so compare like with like.
Different fiscal years. Nvidia’s year ends in January, Broadcom’s in October, AMD’s in December. “Next year” covers different months for each.
Low is not the same as cheap. In TradingKey’s words: “a declining valuation does not necessarily mean the stock is cheap”. It depends on whether the forecast holds.
Testing the forecast. Nvidia’s $15.75 averages $3.94 a quarter, 77% above the $2.22 non-GAAP it reported for its latest quarter, by our arithmetic. Consensus for the quarter that ended in October is $2.49; Nvidia reports on 17 November (the quarterly path).
Price targets are multiples too. Bernstein’s new $200 Nvidia target, as reported by CNBC on 12 November, is 12.7 times the $15.75 consensus; its old $400 was 25.4 times, by our arithmetic (our report).
Name the earnings: which fiscal year, which source, how many estimates, GAAP or not.
Compare companies on the same basis, and note where fiscal years differ.
Look at the range of estimates, not just the average.
Check the trailing figure too: the gap between trailing and forward shows how much growth the forecast assumes.
FAQ
What is a forward P/E ratio?
The share price divided by the earnings per share analysts expect for a coming period, usually the current or next fiscal year. On 11 November 2026 at 10:19 a.m. ET, Nvidia’s $172.68 over $15.75 of next-year consensus gave 11.0 times.
What is the difference between trailing and forward P/E?
Trailing P/E uses the last four reported quarters; forward P/E uses a forecast. For Nvidia on 11 November, trailing GAAP earnings gave about 21.8 times and next-year consensus 11.0 times, by our arithmetic.
Why is the forward P/E lower than the trailing P/E?
Because analysts expect earnings to grow. The price stays the same and the earnings figure gets bigger, so the ratio shrinks. Nvidia’s consensus rises 69% from one fiscal year to the next.
Is a low forward P/E a sign a stock is cheap?
Not on its own. A low multiple can mean the market doubts the forecast. Broadcom and Nvidia had almost the same consensus growth into next year, yet Broadcom traded at 1.8 times Nvidia’s next-year multiple.
Where does consensus EPS come from?
From analysts’ published estimates, averaged by data providers. This page uses Nasdaq’s forecast page; FactSet and others may use different sets of estimates, so their figures can differ.
What is a good forward P/E?
There is no single number. Compare a company with its peers on the same basis and with the market: TradingKey reported the S&P 500’s forward P/E “remains near 18 times” on 11 November 2026.
CNBC, Bernstein on Nvidia, 12 Nov 2026, and Yahoo Finance extended-hours quote, via our newsroom
How this page stays current
Built from the FourWeekMBA newsroom’s sourced reporting. Prices and consensus forecasts change every trading day; every market figure here carries its time and date. The worked example is refreshed after Nvidia’s, Broadcom’s and AMD’s quarterly results. Figures marked “our arithmetic” are our own ratios from the cited data. This page explains a valuation measure and is not investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
