NASDAQ:NVDA NVIDIA Corporation

ISIN: US67066G1040
Information TechnologySemiconductorsAI / Data-Center
NASDAQ · Santa Clara, CA · Semiconductors · mega-cap (~$4.92T) Analysis Status: On-Going
$203.28
+0.2% (day)
20 Jul 2026 · Signal v6

Changes Since Last Report — vs 6 Jul 2026 ($196.28)

Signals unchanged — HOLD across all three horizons. NVIDIA stays a HOLD, not a BUY: it remains in the Expensive band (clean P/E 35.8× = 1.52× the warranted 23.5×, above the 28× semis guardrail), firing the Gate-3 Valuation Ceiling ('great business, wrong price'). No Do-Not-Buy fires — the macro AI-concentration tail is armed but NOT triggering (breadth is broadening: equal-weight RSP beats cap-weight SPY on the 1-wk and 1-mo windows, above its 50-DMA — the skill's disarm signal), so its de-rating leg is a loud watch, not a live catalyst.

DISCLAIMER: This is a quantitative framework for educational purposes only. It is not financial advice. Always do your own research and consult a licensed financial advisor before making investment decisions.

NVIDIA Corporation

NVIDIA designs the accelerated-computing hardware and software that trains and runs the world's AI. Its core product is the data-center GPU (the Blackwell and now Rubin generations) plus the CUDA software stack, high-speed NVLink/Mellanox networking, and full 'AI factory' systems it sells to cloud providers, enterprises and governments. What sets it apart is a near-monopoly in AI training — roughly 80% of the accelerator market — protected by two decades of the CUDA developer ecosystem that most AI software is written against, and by a one-year product cadence rivals struggle to match. Beyond data centre it still sells GeForce gaming GPUs, professional visualisation and automotive chips. In short: the dominant 'picks-and-shovels' supplier of the AI build-out, earning extraordinary margins while that build-out lasts.

HorizonSignalComposite ScoreConfidenceKey Driver
Short-term (1–3 mo)HOLD5458%Weak-ish tape; capped at HOLD by the Valuation-Ceiling gate (Expensive band)
Medium-term (6–12 mo)HOLD5858%Elite business, but Expensive band fires Gate 3 -> HOLD ('great business, wrong price')
Long-term (3–5 yr)HOLD6660%Quality dominates long-run, but a full multiple caps entry until price/EPS converge
Next update: 2026-08-03 — default +14d (earnings 2026-08-26 beyond window)
Table of Contents
1Five-Pillar Scorecard2Hard Gates & Do-Not-Buy Status3Pillar Detail: Business Quality4Pillar Detail: Valuation Attractiveness5Pillar Detail: Underlying Drivers6Pillar Detail: Economic Alignment7Pillar Detail: Entry/Exit Timing8Economic Event Risk9Multi-Timeframe Technical Analysis10Price Chart (6-Month Daily)11Scenario Summary12Entry / Exit Rules13Position Sizing Context14Calibration Snapshot15Data Sources & Methodology
1

Five-Pillar Scorecard

Five independent scores — each 0–100 with its own confidence. The three fundamental pillars (Quality / Valuation / Timing) set the base BUY/HOLD/SELL via the Decision Matrix; the two context pillars (Underlying Drivers, Economic Alignment) then amplify a BUY to STRONG BUY or a SELL to STRONG SELL when both corroborate.

Business Quality

90
elite
conf 78%

Valuation Attractiveness

38
expensive
conf 72%

Entry/Exit Timing

53
neutral
conf 58%

Underlying Drivers

73
Tailwind
conf 70%

Economic Alignment

50
Neutral
conf 58%
2

Hard Gates & Do-Not-Buy Status

Binary safety checks — any TRIGGERED gate is a hard cap regardless of the scores above; CAUTION gates are sizing notes.
Financial Distress
Net cash. Debt/equity 0.07, interest coverage 544x, current ratio 3.44. No liquidity or leverage risk.
Earnings Event Risk
Next earnings 26 Aug 2026 - well beyond the 14-day window. No binary earnings blackout today.
Valuation Ceiling (Gate 3)
TRIGGERED. Clean TTM P/E 35.8x >= the 28x semiconductor guardrail floor, and 1.52x the rate-and-growth-warranted multiple (23.5x). Expensive band -> caps every horizon at HOLD ('great business, wrong price').
⚠️
Systemic Concentration Tail (macro)
CAUTION, not triggered. The 20 Jul Macro report's 'S&P 500 concentration / AI earnings-quality unwind' tail is ARMED but NOT triggering: its own breadth-tell shows equal-weight RSP beating cap-weight SPY on both the 1-week AND 1-month windows, with RSP above its 50-DMA - the skill's explicit disarm/falsification signal. Per the systemic-tail-inheritance rule the inherited de-rating leg is a loud WATCH (see 11), not a live catalyst, so Do-Not-Buy Trigger 2b does NOT fire (arm (a) also fails: clean P/E 35.8x = 1.52x warranted < 2.0x, and 1.28x the guardrail < 1.5x). It re-arms as a live risk only if breadth narrows or a dated hyperscaler capex cut / AI private-valuation markdown lands.
⚠️
Earnings-quality / Dilution (Gate 4)
CAUTION (not triggered). Q1 FY27 net income included $15.9B of unrealised mark-to-market gains on equity stakes (CoreWeave et al.) - ~14% of TTM net income, but ~23% of the latest quarter alone and trending up. Normalised out in 4; the clean number is what scored Valuation. SBC modest; share count falling on buybacks.
⚠️
Regulatory / Binary Event (Gate 5)
CAUTION. China export-control overhang (H20/Blackwell licensing). Material but not binary-existential - a note for sizing, not a HOLD cap on its own.
Severe Driver Collapse
Driver score 73 (Tailwind). AI-capex cycle intact; nowhere near the <=15 collapse threshold.
Net gate read: one hard gate is triggered - the Valuation Ceiling - which caps every horizon at HOLD. No Do-Not-Buy trigger fires: the macro AI-concentration systemic tail is armed but NOT triggering (breadth is broadening - equal-weight RSP leading cap-weight SPY on the 1-wk and 1-mo windows, above its 50-DMA - the skill's disarm signal), so the inherited de-rating leg is a loud watch (11), not a live catalyst. This is a great-business-wrong-price HOLD; the downside risk stays loud.
3

Pillar Detail: Business Quality

A deep dive into the Quality score: business economics, moat, ROIC and the industry benchmark.
Business Quality — Pillar Score
Elite — near-monopoly economics, pristine balance sheet; the one soft spot is a competitive moat eroding at the margin from hyperscaler custom silicon (though NVIDIA's own inference share actually rose to ~74% in Q1).
90
conf 78%

Lifecycle & sector: Information Technology / Semiconductors, classified high-growth (Q1 FY27 revenue +85% YoY, elite margins). Scored on semis + hyper-scaler-demand metrics, not mature-company P/E-of-earnings logic.

Sub-signalValuePeer / contextScoreRead
Revenue trajectoryQ1 FY27 $81.6B, +85% YoY; TTM ~$253BFastest-growing mega-cap; decelerating off a huge base90Top-line momentum still exceptional
ProfitabilityOp margin 64%; gross 74%; clean net margin ~55%Semis median op margin ~25%95Monopoly-grade margins
Cash generationFCF/sh $4.90; FCF margin ~47%Elite; FCF conversion >90%92Cash is real, not accrual
Balance sheetNet cash; D/E 0.07; int-cov 544×; current 3.44Fortress96Zero survivability risk
Industry benchmark (GM + demand)Gross margin 74%; demand red-hot>> 55% healthy line92Pricing power + demand both green
Industry Benchmark — Semis (Gross Margin + Demand): GM 74% vs the 55% "healthy" line; utilisation/backlog red-hot. Score 92. NVIDIA sits at the very top of the semiconductor quality distribution.

Competitive Moat Scorecard (avg 81)

Pricing Power

90
Blackwell/Rubin priced at will; buyers absorb it

Network Effects

85
CUDA developer ecosystem; most AI code targets it

Switching Costs

73
CUDA lock-in strong but eroding via ROCm + PyTorch/JAX abstraction (see below)

Cost Advantage

71
Scale + TSMC access, but rivals share the same fabs; hyperscaler in-house silicon undercuts on TCO

Intangibles

88
CUDA, patents, brand, one-year cadence
Competitive Environment — the moat above is measured today; here is who is attacking it and which way share is trending. The Switching-Cost (73) and Cost-Advantage (71) sub-scores are derived from this read, not asserted.
RivalThreat typeShare trajectoryMoat-erosion vector
Hyperscaler custom silicon — Google TPU, AWS Trainium, MSFT Maia, Meta MTIA (via Broadcom/Marvell)Vertical in-house substitutionFastest-growing threat: Meta begins in-house AI-chip deployment from Sept; custom-ASIC sales projected +45% (TrendForce)NVIDIA's largest customers designing out dependence in the price-sensitive inference tier — erodes switching cost and cost advantage
AMD (MI350X/MI400)Direct merchant rivalNot taking share — NVIDIA inference revenue exceeds AMD's combined AI revenuePrice/TCO in inference; ROCm maturing vs CUDA — real but not yet biting
Intel (Gaudi)Direct rivalLosing / negligibleMinimal near-term pressure
Net effect on the moat: the headline is nuanced — NVIDIA's AI-inference share actually rose to ~74% in Q1 (the merchant rivals are not taking share today), so the trajectory is best read as stable, an improvement on last month's 'losing'. But the structural erosion vector is real and building: the hyperscalers' own custom silicon (Meta live from Sept, Broadcom/Marvell ramping) is the fastest-growing substitute in the price-sensitive inference tier. So Switching Costs sit at 73 and Cost Advantage at 71; overall competitive-threat level elevated (structural, not yet showing in share). This feeds the §11 Bear trigger and the §12 thesis-invalidation condition.

ROIC & Capital Allocation

ROIC is top-of-peer-set (FMP ROE/ROA sub-scores both 5/5). Capital allocation disciplined — buybacks (share count falling ~1%/yr), a token dividend, reinvestment at very high returns. Management skin-in-the-game scored 72: founder-led (Jensen Huang), but insider sales run on 10b5-1 plans and SBC, while modest as a % of revenue, is real. Quality confidence 78% (−: high-growth names are inherently harder to score; the earnings-quality distortion required normalisation).

4

Pillar Detail: Valuation Attractiveness

Sector-appropriate multiples, FCF yield, reverse-DCF implied growth, embedded optionality, and the analyst-consensus cross-check.
Valuation Attractiveness — Pillar Score
Expensive on the anchor: clean P/E 35.8× breaches the 28× semiconductor floor and is 1.52× the warranted multiple. Forward metrics are more forgiving, but the disciplined read is Expensive → Gate 3 + the Trigger-2b Do-Not-Buy.
38
conf 72%
Earnings-quality decomposition (run first). Q1 FY27 reported net income of $58.3B included $15.9B of unrealised mark-to-market gains on equity stakes (CoreWeave and other AI-ecosystem holdings). Over the trailing twelve months, non-operating income is ~$27B pre-tax = ~14% of net income (but ~23% in the latest quarter alone, trending up). These gains are real GAAP income but non-operating and reversible. We therefore score valuation on the clean/operating number, not the reported one.
MetricReportedClean (operating)
TTM EPS (diluted)$6.53$5.67
TTM P/E (at $203.28)31.1×35.8×
PEG (vs ~24% growth)0.28 (uses hyper-growth — meaningless)~1.5
THE ANCHOR — Warranted-Multiple Valuation. A company is worth the present value of its cash flows: growth lifts the warranted multiple, the discount rate lowers it. Score = 35.8× (clean) ÷ 23.5× (warranted) = 1.52 → Expensive band (< 40). And the clean 35.8× is above the 28× semiconductor guardrail floor on its own — Expensive on the floor alone (even the reported 31.1× clears it), independent of the ratio. Double-confirmed. A Full/Expensive-band name is not eligible for STRONG-BUY amplification, and the floor breach fires the Gate 3 Valuation Ceilingcaps every horizon at HOLD ('great business, wrong price'). It does not fire Do-Not-Buy Trigger 2b: although the macro AI-concentration tail is armed, its own disarm signal (breadth broadening — RSP beating SPY on the 1-wk and 1-mo windows) is active, so there is no live triggering de-rating catalyst; and arm (a) fails (1.52× warranted < 2.0×; 1.28× the guardrail < 1.5×).

Implied-growth read (narrative colour): at $203 on clean TTM earnings the market is embedding ~24%/yr for 5 years; our disciplined estimate is ~18% — the price embeds more growth than we are willing to underwrite. On forward FY27 estimates the picture softens (fwd P/E low-20s×) — the growth-stock tension is real, and it is the honest counter-case to the Do-Not-Buy: if the aggressive forward growth is delivered, the name grows into the price. The framework will not pay up for the flawless path at a rich multiple — hence HOLD, not a fresh buy — while flagging the armed (not-yet-triggering) concentration tail as the loud downside watch.

Lens (weight)ReadScore
Warranted-multiple anchor (40%)35.8× vs 23.5× warranted = 1.52× → Expensive; also > 28× floor~28
Sector median (20%)Fwd P/E low-20s× roughly in line with mega-cap semis; trailing rich50
Own-history decile (15%)~6th decile of its own 5-yr range (not extreme vs itself)55
PEG (10%)Clean PEG ~1.5 (fair-ish for the growth)50
Analyst consensus (15%)Consensus $319.48 (+57%), median $300 (+48%), high $500, low $218 — strongly bullish; but a lagging herd at ~76% bullish (all 'maintains')85

FCF yield (universal anchor): ~2.4% (FCF/sh $4.90 ÷ $203) — the "3–5% expensive-needs-growth" zone, and below it. Confirms the anchor.

Embedded Optionality / Free Upside (tilt, not a re-rating). (1) NVIDIA's equity-investment portfolio (CoreWeave, Nebius, Nscale and other AI-cloud stakes) — a call option on the neocloud build-out, though currently inflating reported earnings rather than hidden; (2) Rubin / Rubin-Ultra and the networking + software (NIM/AI Enterprise) attach; (3) sovereign-AI + enterprise-inference TAM (the new Palantir government-AI partnership, 20 Jul) not fully in numbers. Net: the core AI-factory business justifies most of the $203; the optionality is why you keep watching, not why the stock is cheap — the core is already richly priced. Tilt applied: ~+4.

FMP ratings cross-check: Overall A- (4/5) — ROE 5/5, ROA 5/5 (elite quality) but P/E 2/5 and P/B 1/5 (poor). This independently confirms the split verdict: superb business, expensive price.

Valuation confidence 72% (+ hard analyst targets, grades, ratings snapshot all available; − the earnings-quality distortion and the reported-vs-clean gap widen the interval).

5

Pillar Detail: Underlying Drivers

The dominant external force the stock is tethered to, scored 0–100. A context pillar: it does not change the base signal — it feeds amplification (tailwind ≥65 can lift BUY→STRONG BUY; headwind ≤35 can push SELL→STRONG SELL).
Primary Driver
AI / data-center accelerator capex cycle
73
Tailwind (>=65 - amplification-eligible, but base signal is HOLD so no amplification)

Primary driver: the AI / data-center accelerator capex supercycle — hyperscaler and sovereign spending on AI infrastructure. NVIDIA is the geared bet on the direction of that spend. (Not a commodity-price driver, so the Step-2b commodity-trend overlay does not apply.)

Horizon (weight)ReadScore
Historical (25%)Data-center revenue compounded from ~$47B to ~$253B TTM in two years — the strongest capex wave in tech history85
Current (50%)Blackwell/GB300 ramp sold out; Rubin on track; inference share ~74%; ~$1T demand visibility cited through 2027. But the 22–30 Jul hyperscaler prints are a binary — a single capex guide-down is an index-level event72
Forward (25%)Consensus capex still robust but decelerating; custom-silicon substitution + China export limits cap the upside66

Driver score 73 → Tailwind. This is ≥65, so it is eligible to amplify a BUY to STRONG BUY — but the base signal is HOLD, so no amplification occurs and the driver does not change the three fundamental pillar scores. Thesis-invalidation floor: a hyperscaler capex cut (not just deceleration) — that is the dial to watch, and it is the same event that would trigger the (currently armed-but-not-firing) systemic concentration tail in §11 — the downside watch; the market-wide read comes at the 22–30 Jul hyperscaler prints. Driver confidence 70%.

6

Pillar Detail: Economic Alignment

How the current economic climate sits relative to this stock, read from the latest Macro-Economic report. Classifies the macro pressure (Tailwind / Neutral / Headwind) — the second amplification input — and frames a long entry as Trend-Following or Contrarian with a 0–100 conviction.
Stance · Pressure
Neutral · Neutral
50
conviction

The 20 Jul MacroDriver regime is Stagflation-lite (energy-supply-shock driven) — a narrow, contested lead. It maps Information Technology (XLK) to Underperform short / Neutral medium / Outperform long (U / N / O): a near-term Headwind (mega-cap concentration + a 10-Y near 4.5%+ pressuring long-duration multiples + the 22–30 Jul earnings binary), Neutral medium, structural Tailwind long. Anchoring on the medium horizon, the economic pressure is Neutral, so no amplification is enabled either way (and the base signal is HOLD regardless). Stance Neutral, conviction 50 — a mild net near-term drag, informational. The same report carries an armed 'S&P 500 concentration / AI earnings-quality unwind' tail aimed at NVIDIA's cohort — but it is armed and NOT triggering: the macro report's own breadth-tell shows equal-weight RSP beating cap-weight SPY on both the 1-week and 1-month windows, with RSP above its 50-DMA — the skill's explicit disarm/falsification signal. So the inherited de-rating leg is a loud watch (§11 Bear), not a live catalyst, and it does not fire a Do-Not-Buy. This is a material change from the prior report's Trend-Following/Tailwind read (the 3 Jul macro had XLK O/O/O).

Source: sector-map (XLK — NVDA not in the macro watchlist set) · Macro report 2026-07-20

7

Pillar Detail: Entry/Exit Timing

The risk-reward framework, relative strength vs SPY and the sector ETF, the macro overlay, news-derived sentiment, and the catalyst cluster.
Entry/Exit Timing — Pillar Score
Neutral. Primary trend (monthly/weekly) still up, but the stock is -14% off its May peak, below a falling 50-DMA, and the tool reads confluence bearish. The daily MACD is tentatively turning up.
53
conf 58%

Risk-reward: $203 sits below the falling 50-DMA (~$210) but well above the rising 200-DMA (~$192) and the $189.8 late-June swing low. A logical stop below that support is ~$186 (~2.3 ATR; ATR $7.5 = 3.7%). Reward to the base target ($225) is ~+11%; to bull ($295) ~+45%; downside to the bear ($145) is -29%. Not the tightly-skewed setup you want to initiate on — and the Do-Not-Buy overrides it anyway.

Relative strength: NVDA is -14% from the $236.54 May high; over the last 3 months it is roughly +2% vs SPY +6% (a ~4-point lag) as capital rotates within tech (Apple +12%/mo, NVDA -4%/mo per the tape). Vs the SMH semis ETF it lagged over 3 months (memory names led) but outperformed the falling SMH over the last month. 52-week range position ~54% (mid). Net: a modest relative laggard vs SPY.

Macro overlay (semis = medium sensitivity, 15%): the 20 Jul regime maps XLK to short-Underperform — a near-term headwind (rate + concentration drag); Fed on hold into a contested Stagflation-lite regime. Neutral-to-mildly-negative near term.

Sentiment: analyst grades are a wall of maintains (0 upgrades / 0 downgrades in 30 days) — no fresh conviction. News tone is genuinely mixed: positive product/partnership flow (Palantir government-AI deal, Omniverse toolkit, ~74% inference share, $1T demand visibility cited) against a loud valuation/rotation bear thread (Apple overtaking NVDA as most valuable, SOXX -19% off its June high, Meta in-house chips from Sept, crash-risk pieces). Blended sentiment ~50.

Catalysts: the late-July macro cluster (FOMC 29 Jul, GDP + Core PCE 30 Jul) plus the market-wide 22-30 Jul hyperscaler prints; NVDA own earnings not until 26 Aug. Some clustering of index-level events. Catalyst score ~52. Timing confidence 58% (-: below a falling 50-DMA; confluence bearish; RS soft vs SPY).

8

Economic Event Risk

High-impact macro releases in the next 14 days that could swing this stock, plus the last 7 days of surprises.

Upcoming events (next 30 days)

DateEventImpactForecastPreviousRelevant?Why
27 JulDurable Goods MoM (Jun)High0.3%−4.5%⚠ LowCapex demand read (indirect)
29 JulFOMC Rate Decision (Warsh)HighHold 3.75%3.75%✅ YesRate path → long-duration tech multiples; the regime event
30 JulQ2 GDP (Advance)High~1.1–2.0%2.1%⚠ MediumGrowth read; risk appetite for high-beta tech
30 JulCore PCE MoM (Jun)High0.3%0.3%✅ YesInflation → rate path → growth-multiple sensitivity

Recent surprises (last 7 days)

DateEventActualForecastSurpriseImpact
14 JulCore CPI YoY (Jun)2.6%2.8%−7.1% (below)Soft — eased the multiple headwind (last clean print before gasoline re-accelerates)
14 JulCPI MoM (Jun)−0.4%−0.1%belowSoft headline (energy leg); disinflation now reversing
17 JulMichigan Sentiment (Jul)54.451.0+6.7% (above)Consumer firmer than feared — keeps Reacceleration alive
17 JulHousing Starts (Jun)1.427M1.31M+8.9% (above)Firm housing argues against near-term demand destruction

The window is dominated by a late-July macro cluster: the 29 Jul FOMC (hold expected, no explicit Sep-cut green light into the energy shock), Q2 GDP + Core PCE on 30 Jul. Soft June CPI (14 Jul) eased the rate-driven multiple headwind, but the macro author flags it as the last clean disinflation print before gasoline re-accelerates on the Iran/Hormuz oil spike. Semis are medium macro-sensitivity, so no 3-day WAIT-override fires. For NVDA the far bigger event is off this calendar: the 22–30 Jul hyperscaler earnings — the trigger window for the armed AI-concentration tail.

9

Multi-Timeframe Technical Analysis

Trend, RSI and breakout status across monthly / weekly / daily / hourly / 15-minute, with a confluence verdict.
TimeframeTrendDirectionRSIMACDKey S/RBreakoutVol
MonthlyUptrend ↑Bullish67.0+, hist rollingS: 164 · R: 236.5Res-breakout0.4x
WeeklyUptrend ↑Bullish53.3+, flatteningS: 164 · R: 197.6 / 214Res-breakout0.1x
DailyWeakening →Neutral48.2+, turning upS: 189.8 / 194.7 · R: 214Res-breakout1.0x
HourlyWeakening ↓Bearish44.2−, flatS: 198 · R: 207Support-breakdown
15-minDowntrend ↓Bearish47.9+, basingS: 202 · R: 206
Confluence: Mixed — primary trend up, tactical trend down (tool: bearish) · MTF Score 55

The higher timeframes (monthly, weekly) remain in uptrends and price ($203) is above the rising 200-DMA ($192) — the secular bull structure is intact. But the stock is −14% off its $236.54 May peak and sits below a falling 50-DMA (~$210); the daily MACD histogram has just turned up (a tentative positive) while the intraday timeframes are still heavy — hence the tool's 'bearish' confluence read. Net: a pullback / consolidation within a larger uptrend. The reachable early entry is a reclaim of the $210 50-DMA on volume, or a tested higher-low bounce off $189.8–192 — neither confirmed. Key level: $189.8–192 (swing low + 200-DMA).

10

Price Chart (6-Month Daily)

A 6-month daily close line with SMA50 and key support/resistance — the visual companion to the MTF table.

NVDA 6-month daily. Peaked $236.54 (13 May), now $203.28 — a −14% pullback that has broken below the falling 50-DMA (~$210) but is holding well above the rising 200-DMA (~$192) / $189.8 swing low. Monthly/weekly uptrend intact; daily consolidating.

11

Scenario Summary

Bull / Base / Bear 12-month price paths with triggers and probability weights.

Bull $295 (12m, 25%)

AI-capex supercycle re-accelerates; Blackwell→Rubin ramp with no share loss; the multiple holds ~35–40× on rising clean EPS. Breadth stays broad and the concentration tail is disarmed. Reaches near the analyst median ~$295–300 (+45%). Trigger: hyperscaler capex guides up at the 22–30 Jul prints; custom-silicon share plateaus.

Base $225 (12m, 55%)

Growth delivers but the multiple compresses toward the warranted ~24–28× as the cycle matures and EPS grows into the price. ~$225 (+11%). Trigger: steady execution, modest digestion, no capex cut, breadth neither broadens decisively nor breaks. This is the probability-weighted centre of gravity.

Bear $145 (12m, 20%)

The dual downside. (i) Idiosyncratic: hyperscaler custom silicon (Meta live from Sept) + AMD take visible inference share and gross margin compresses. (ii) Systemic WATCH — the 20 Jul macro AI-concentration tail is currently armed but NOT triggering (breadth is broadening, RSP leading SPY — the disarm signal). IF it fires — breadth narrows, or a dated hyperscaler capex cut / AI private-valuation markdown lands — a cohort-level multiple de-rating takes the AI mega-caps from ~30–40× toward ~18–22× (NVIDIA's own CoreWeave-type gains reverse). Combined path ~$145 (−29%). Falsification / stays-disarmed: breadth keeps broadening and capex holds.

Probability-weighted 12-mo value ≈ $226 (0.25·$295 + 0.55·$225 + 0.20·$145) — ~+11% above the $203 price on an analyst-informed basis, essentially the base case. But the anchor-disciplined value (warranted multiple on clean earnings) is ~$195–205, at the current price, and the payoff is asymmetric: a shallow base-case gain against a −29% bear if the (currently armed-but-not-triggering) concentration tail fires. A multiple already at the anchor with no near-term edge is why the signal is HOLD, not a fresh buy — you are not paid to add here.

12

Entry / Exit Rules

Three independent entry paths (Fundamental · Technical · Catalyst) and three exit triggers (Stop-Loss · Thesis · Profit-Target). Any one entry path is a valid entry — the more that agree, the larger the position the conviction ladder suggests. Exits are graded by severity, not count.

How to read this — the Conviction Ladder

The three entry groups are alternative paths to a buy, not a checklist. A group counts only when all its sub-conditions hold. How many groups are satisfied sets the suggested size — it does not gate whether you may enter: 1 group = Half-Size (a valid starter/scale-in), 2 = Full-Size, 3 = Over-Size (highest conviction); 0 = Wait (no path open yet). A strong overall signal can still read Wait here when the stock is well above its entry zones — that flags "good business, no entry edge right now," not a contradiction. Exits are graded by severity of what is live, not by a count: a hard stop is an Exit on its own.
Entry conviction: Wait0 of 3 groups met — no entry path open

Fundamental — not MET

No valuation margin at $203 — clean P/E 35.8× is in the Expensive band; fair value ≈ current price / below.
⛔ Price $203 < disciplined fair value (~$200) — no margin of safety
✅ No earnings within 7 days (26 Aug)
✅ Underlying-Driver score ≥ 50 (73)

Technical — not MET

Below a falling 50-DMA; preferred entry is a $210 reclaim OR a confirmed support bounce — neither present.
⛔ Daily close > 50-DMA ($210) on >1.5× volume
⛔ OR a tested bounce off $189.8–192 support with a higher low + MACD turning up
✅ RSI 35–65 (48, daily)

Catalyst — not MET

No confirming event — earnings are 5+ weeks out.
· Post-earnings move > +5% with guidance raised (next print 26 Aug)
⛔ Volume > 2× 20-day average on an up-move

Forecast: 0 of 3 groups met → Conviction: WAIT. There is no entry edge at $203, and the Valuation-Ceiling gate caps the signal at HOLD regardless. Fundamental turns MET only on a pullback into the ~$186–195 zone (or EPS growing into the multiple at the 26 Aug print); Technical turns MET on a confirmed higher-low bounce off $189.8–192 (moderate — tape is above it) or a $210 50-DMA reclaim on volume (low near-term — the 50-DMA is falling ~3.5% above spot); Catalyst is earnings-dependent (26 Aug). Confidence: Moderate — the realistic path to a buy setup is a de-rating toward $186–195 or the 26 Aug EPS step-up.

Exit action: Holdno exit trigger is live — hold the position

Stop-Loss — not LIVE

⛔ Two daily closes below $186 (under the $189.8 swing low / 200-DMA)

Thesis Invalidation — not LIVE

⛔ A hyperscaler capex CUT (not just deceleration) — would also TRIGGER the (currently armed-but-not-firing) macro concentration tail
⛔ Hyperscaler custom silicon + AMD take material inference share and gross margin compresses through the low-60s%
⛔ CUDA moat visibly breaks as customers migrate to ROCm / abstraction layers

Profit-Target — not LIVE

⛔ Price into $295–300 (bull / analyst median) with RSI > 70 and no clean-EPS catch-up

Forecast: For an existing holder this is a Hold — no exit rule is live: the stop ($186) is ~8% below and unlikely absent a broad tech de-rating, and the earliest real risk is the 22–30 Jul hyperscaler prints (an index-level capex tell) and the 26 Aug NVDA print. HOLD caps fresh buying; it does not force a quality holder out.

Imagine you act at the current price of $203.28 · as of 20 Jul 2026

What if you bought now?

You'd be risking ~8% to the $186 stop (bear −29% to $145) to gain base +11% ($225) / bull +45% ($295) — while the armed-but-not-triggering concentration tail sits as the loud downside watch.
  • Risking: $186 stop (−8.5%); bear $145 (−29%) on a concentration-unwind + custom-silicon share-loss combo; you'd be paying above any margin of safety (clean P/E 35.8× = 1.52× warranted) and below a falling 50-DMA, with the 22–30 Jul hyperscaler prints + 29 Jul FOMC as immediate path risk.
  • Gaining: base $225 (+11%) · bull $295 (+45%); ~2.4% FCF yield while you wait; and the equity-portfolio + Rubin optionality you own for free. But the probability-weighted value is ~$226 (+11%) but the payoff is asymmetric to a −29% bear if the armed-but-not-triggering concentration tail fires.
Read: the framework says HOLD, not a fresh buy — you are paying up (Expensive band) for no near-term edge. Waiting for $186–195 or an EPS catch-up (26 Aug) materially improves the deal.

What if you sold now?

You'd be giving up base +11% / bull +45% upside to protect against the −29% bear.
  • Giving up: base $225 (+11%) and bull $295 (+45%), the compounding of an elite franchise, and the portfolio optionality; you'd be selling roughly at fair value (~$200), not far below it.
  • Protecting: capital if the armed concentration tail fires and the bear ($145) plays out. But for an existing holder no exit rule is triggered right now — stop not hit, no thesis break, no profit-take (RSI 48, not >70).
Read: no mechanical reason to sell — stop not hit, no thesis break, no profit-take (RSI 48). This is a hold zone; trim into $295+ strength.
13

Position Sizing Context

Illustrative portfolio math (not advice) translating conviction into an allocation given risk-per-share and volatility.

Position sizing not computed — no risk budget or portfolio role was specified for this watchlist name. Illustrative context only: the signal is HOLD (capped by the Valuation-Ceiling gate) and the §12 Conviction Ladder reads WAIT (0/3 entry paths met), so the sizing guidance is to wait for an entry path (a de-rating toward $186–195 or an EPS catch-up) rather than assign a %. Beta is 2.21 — a given dollar position carries ~2.2× the market's daily swing (a 5% weight behaves like ~11% in risk terms); daily ATR ~3.7%. An existing holder is not forced out here.

14

Calibration Snapshot

Machine-readable snapshot of every score, level and signal, saved alongside the HTML so the next run can compute deltas.
{
  "ticker": "NVDA",
  "date": "2026-07-20",
  "version": "v6",
  "exchange": "NASDAQ",
  "ticker_display": "NASDAQ:NVDA",
  "exchange_ticker": "NASDAQ:NVDA",
  "isin": "US67066G1040",
  "api_ticker": "NVDA",
  "company": "NVIDIA Corporation",
  "sector": "Technology",
  "sub_industry": "Semiconductors",
  "section": "Technology / Nasdaq",
  "finder_ticker": "NVDA",
  "finder_exchange": "\ud83c\uddfa\ud83c\uddf8 NASDAQ",
  "analysis_status": "on-going",
  "lifecycle_stage": "high-growth",
  "beta": 2.211,
  "shares_outstanding": 24286000000,
  "user_horizon": null,
  "user_allocation_pct": null,
  "portfolio_role": null,
  "price_at_rating": 203.28,
  "signal_short": "HOLD",
  "signal_medium": "HOLD",
  "signal_long": "HOLD",
  "primary_signal": "HOLD",
  "composite_short": 54,
  "composite_medium": 58,
  "composite_long": 66,
  "quality_score": 90,
  "valuation_score": 38,
  "timing_score": 53,
  "driver_score": 73,
  "overall_confidence": 58,
  "quality_detail": {
    "industry_benchmark_name": "Gross Margin + Demand (Semis)",
    "industry_benchmark_value": "GM 74% / demand red-hot",
    "industry_benchmark_score": 92,
    "moat_score": 81,
    "roic_percentile_vs_peers": 98,
    "capital_allocation": 85,
    "management_skin_in_game": 72
  },
  "valuation_detail": {
    "fcf_yield": 2.4,
    "reported_pe": 31.1,
    "clean_pe": 35.8,
    "warranted_multiple": 23.5,
    "actual_multiple": 35.8,
    "warranted_ratio": 1.52,
    "discount_rate_r": 11.0,
    "risk_free_10y": 4.5,
    "g_near": 18,
    "g_term": 3,
    "val_multiple_basis": "clean TTM P/E",
    "val_band": "expensive",
    "implied_growth_rate": 24.0,
    "consensus_growth_rate": 24.0,
    "historical_valuation_decile": 6
  },
  "timing_detail": {
    "mtf_confluence": 55,
    "risk_reward_score": 52,
    "relative_strength_vs_spy": -3.8,
    "relative_strength_vs_sector": 0.0,
    "catalyst_clustering_score": 52,
    "dynamic_macro_weight": 0.15
  },
  "warranted_multiple": 23.5,
  "actual_multiple": 35.8,
  "warranted_ratio": 1.52,
  "discount_rate_r": 11.0,
  "risk_free_10y": 4.5,
  "g_near": 18,
  "g_term": 3,
  "val_multiple_basis": "clean TTM P/E",
  "val_band": "expensive",
  "nonop_pct_of_net_income": 14.3,
  "clean_pe": 35.8,
  "clean_peg": 1.5,
  "driver_name": "AI / data-center accelerator capex cycle",
  "driver_label": "Tailwind",
  "driver_amplification_eligible": true,
  "economic_alignment_stance": "Neutral",
  "economic_alignment_conviction": 50,
  "economic_alignment_pressure": "Neutral",
  "economic_alignment_source": "sector-map (XLK)",
  "macro_report_date": "2026-07-20",
  "competitive_share_trajectory": "stable",
  "competitive_threat_level": "elevated",
  "moat_score": 81,
  "fcf_yield": 2.4,
  "analyst_consensus_target": 319.48,
  "analyst_target_high": 500,
  "analyst_target_low": 218,
  "analyst_target_median": 300,
  "analyst_target_upside_pct": 57.2,
  "analyst_grades_consensus": "Buy",
  "analyst_bullish_pct": 76,
  "analyst_coverage_count": 79,
  "recent_upgrades_30d": 0,
  "recent_downgrades_30d": 0,
  "fmp_rating": "A-",
  "fmp_overall_score": 4,
  "fair_value_est": 200.0,
  "stop_loss": 186.0,
  "target_price": 225.0,
  "scenario_base_target": 225,
  "scenario_bull_target": 295,
  "target_bull": 295,
  "target_bear": 145,
  "scenario_bear_target": 145,
  "entry_groups_met": 0,
  "entry_conviction": "Wait",
  "entry_criteria_met": 0,
  "entry_criteria_total": 3,
  "exit_groups_live": 0,
  "exit_action": "Hold",
  "exit_criteria_met": 0,
  "exit_criteria_total": 3,
  "hard_gate_state": "caution",
  "gates_triggered": [
    "Valuation Ceiling (Gate 3)"
  ],
  "gates_caution": [
    "Earnings-quality (non-op gains)",
    "Regulatory / China export controls",
    "Systemic AI-concentration tail (armed, not triggering)"
  ],
  "do_not_buy_triggers": [],
  "next_update_date": "2026-08-03",
  "next_update_basis": "default +14d (earnings 2026-08-26 beyond window)",
  "currency": "USD",
  "time": "1200"
}

Signal HOLD / HOLD / HOLD (unchanged vs 6 Jul at $196.28). It stays HOLD — not BUY — because the name remains Expensive (clean P/E 35.8× = 1.52× warranted 23.5×, above the 28× guardrail), firing the Gate-3 Valuation Ceiling ('great business, wrong price'). No Do-Not-Buy trigger fires: the macro AI-concentration tail is armed but NOT triggering (breadth broadening — RSP leading SPY on the 1-wk and 1-mo windows — the skill's disarm signal), and arm (a) fails (1.52× warranted; 1.28× guardrail). Price +3.6% since the last report; Valuation eased 39→38; Timing 54→53; Driver 74→73; Economic Alignment Trend-Following/Tailwind/58 → Neutral/Neutral/50 (XLK flipped to short-Underperform). Competitive trajectory improved 'losing'→'stable' (inference share ~74%). Next update 3 Aug (default +14d).

15

Data Sources & Methodology

Audit trail of every data source: fully available (✓), fallback (⚠), or failed (✗), plus provenance-based confidence haircuts.
Data Source Status
get_stock_snapshot / get_company_profile Price $203.28, mktcap $4.92T, beta 2.211, ISIN US67066G1040
get_income_statement (8q) Earnings-quality decomposition — $16.4B total other income (~$15.9B non-op equity gains) in Q1 FY27
get_financial_ratios Margins, FCF, leverage; TTM P/E 31.0×, P/FCF 41×
get_multi_timeframe_analysis 5 timeframes incl. intraday; confluence bearish
get_price_target_consensus / _summary Consensus $319.48; median $300; 27 analysts/qtr
get_grades_consensus / get_stock_grades Buy consensus (76% bullish); 0 up / 0 down in 30d (all maintains)
get_ratings_snapshot FMP A- (4/5); P/E 2, P/B 1
get_analyst_estimates Forward FY28–FY31 EPS/revenue
get_economic_calendar Soft June CPI 14 Jul; FOMC 29 Jul; GDP+PCE 30 Jul
get_earnings_calendar Next earnings 26 Aug 2026 (est EPS $2.09, rev $91.8B)
get_polygon_news (12) Meta in-house chip from Sept; NVDA inference share ~74%; Palantir gov-AI deal; 'struggling in 2026'
Macro-Economic state (20 Jul) Stagflation-lite; XLK U/N/O; 10-Y ~4.5%+; AI-concentration tail ARMED
Relative strength (SPY, SMH) NVDA 3-mo +2% vs SPY +6% (lag); vs SMH mixed (led 1-mo as memory rolled)
Impact on scores: All primary sources returned cleanly — overall confidence (58%) is set by the min pillar confidence (Timing 58% / Valuation 72% / Quality 78%), not by data gaps. The one provenance caveat: reported earnings are distorted by non-operating gains, handled by the §4 clean-earnings normalisation (which is why Valuation confidence is 72% not higher).
DISCLAIMER: This is a quantitative framework for educational purposes only. It is not financial advice. Always do your own research and consult a licensed financial advisor before making investment decisions.