Every common valuation model on Nvidia, side by side: what each says, which ones fit, and the range they give
On November 5, 2026, Nvidia fell 9.3% in one day. It closed at $186.64 on November 6, 22% below its 52-week closing high and 19% lower than a month earlier. Everyone is asking the same thing: is it undervalued now?
There is no single answer, because there is no single valuation model. So we ran every common model on Nvidia, side by side. For each one we show what it says, whether it fits a company like Nvidia, and why. The models that fit give a range, and the range tells you more than any one number.
Data as of the Nov 6, 2026 close. Statements from Nvidia's SEC filings (quarter to July 26, 2026); analysts' consensus read Nov 9, 2026. Education, not investment advice.
| Model | What it says | Fits Nvidia? | Why |
|---|---|---|---|
| DCF on free cash flow (after stock-based pay) | bear $43–49 · base $124–148 · bull $202–244 | ✅ best fit | It values the cash the business produces, not accounting profit. It is very sensitive to the discount rate and to how fast growth slows. |
| Reverse DCF | the price needs 25–28% a year of cash-flow growth for 5 years | ✅ | It shows what the price assumes, without forecasting. |
| Forward P/E | 20.0× this fiscal year, 11.7× next; 15–25× next year's earnings → $211–352 | ⚠️ partly | It is only as good as next year's +71%. In a chip cycle, forward multiples look lowest just before estimates fall. |
| Price-to-sales vs its history | $253 at its 10-year median | ⚠️ weak | It ignores margins: 65% today, about 25% in 2019. |
| Graham growth formula | $167 / $225 / $284 at 10% / 15% / 20% growth | ⚠️ | The answer is whatever growth rate you pick, so it adds little beyond the reverse DCF. |
| P/E vs its own 10-year median | $352–436 | ❌ | The median comes from a smaller, faster-growing, lower-margin company. Today's Nvidia is a different business. |
| PEG | 0.28, which implies about $660 at PEG = 1 | ❌ | It assumes today's growth rate lasts for years. Growth this fast doesn't. |
| Price-to-free-cash-flow vs its history | $322 | ❌ | That history (since 2022) is all boom years, so its median isn't a "normal" level. |
| Graham number | $41 | ❌ | It was built for asset-heavy, slow-growth firms. Nvidia's value isn't in its book. |
| Dividend discount model | $0.87 | ❌ | Nvidia returns cash through buybacks ($55.3B over the last year), not dividends. |
| Multiple (trailing 12 months) | 10-year median | Today | Where today sits in 10 years |
|---|---|---|---|
| P/E | 54.9× | 23.5× | 3rd percentile |
| Price-to-sales | 20.3× | 15.0× | 32nd percentile |
| Price-to-free-cash-flow (since 2022) | 61.5× | 35.7× | the lowest since 2022 |
On every multiple, Nvidia is near the low end of its own history. But that history mostly measured a smaller company in a boom, so "cheap against its past" is weak evidence on its own. That's why the cash-flow models carry more weight here.
One more detail matters. About $24B of profit in the first half of this fiscal year was non-operating income, mostly gains on stakes in private companies, not from selling chips. Without them, the P/E is closer to 27×. Free cash flow was 66% of net income over the last year.
| Growth | |
|---|---|
| The price needs (reverse DCF) | 25–28% a year for 5 years |
| Analysts, next fiscal year | +71% earnings |
| The last 5 years | earnings +82.5% a year, sales +66.9% a year |
Capital Copilot's DCF and Reverse DCF tabs show different numbers for Nvidia ($78.56 a share, and 43.8% a year). That's because they use different inputs: they go straight to long-term growth after year 5 instead of slowing down gradually, use free cash flow before stock-based pay, and use their own discount rate. This report slows growth to 3% over 5 more years and subtracts stock-based pay. Both are right for their assumptions; that's why we read a range.
The price asks for far less growth than analysts expect for the next 18 months. A 5-year average of 25–28% fits two strong years of cash-flow growth followed by flat or slower years:
| Value per share | From | |
|---|---|---|
| Low | $75–90 | a cycle path with one down year |
| Central | $125–200 | the DCF base to bull case, and analysts' path then +15% a year |
| High | $210–280 | 15–20× next year's expected earnings, if next year isn't a peak |
$186.64 sits in the upper half of the central band. Whether that's a good entry depends on one question: do you believe next year's earnings are a step up that lasts, or a peak?
| Nvidia | Taiwan Semiconductor | |
|---|---|---|
| Price (Nov 6 close) | $186.64 | $375.44 |
| P/E vs its own 10-year median (as Capital Copilot computes them) | 26.6× vs 50.2× (far below) | 24.2× vs 21.2× (above) |
| Expected earnings growth, next year | +71% | +34% |
| Return on invested capital | 77% | 47% |
| Price move, last month | −19.0% | −18.0% |
Both fell about the same amount in a month, which reads as a move across the chip sector, not something specific to one company. Taiwan Semiconductor trades above its own history and Nvidia well below its own. Taiwan Semiconductor gets paid whoever wins the chip design, while Nvidia's margins carry more of the cycle.
Prices: daily closes to Nov 6, 2026. Statements: SEC EDGAR (Nvidia, quarter to July 26, 2026). Consensus: Alpha Vantage, read Nov 9, 2026. Rates: 10-year Treasury 4.84% (Federal Reserve H.15 via FRED, Nov 5, 2026). Discount rate 12.95% (also shown at 11.5%). Capital Copilot figures computed Nov 8, 2026. Education, not investment advice: these are scenarios and ranges, not a recommendation to buy or sell.
Capital Copilot turns market research into audio you can listen to. Free plan, no time limit.
Free beginner course, lesson 3: what a share is, common and preferred stock, dividends, splits and buybacks
What Microsoft, Alphabet, Amazon, Meta, Oracle and CoreWeave pay to borrow, what that price says about risk, and why interest rates matter more than you think
What the filings show about cloud revenue, capital spending and backlog at Google, Amazon and Microsoft, and the signals to watch
What return on equity and return on capital really measure, and the two numbers that decide your return
Briefings, reports and audio TL;DRs — free, with no time limit.
Comments
Loading comments