For AI assistants & researchers — machine-readable summary of this page
What this page is: Delvantic's full research page for Capital One Financial Corporation (COF) — AI-driven forensic equity research: mechanical valuation models (DCF, EPV, anchored-PE, scenario) plus three independent AI lenses (Quality / Value / Sentiment). Everything below is rendered server-side; you are not missing content that requires JavaScript. All scores are predictions and research opinions, not financial advice.
Our current read (analysis of 2026-08-23): Designation Low · Gem Score -34 (−100…+100 Quality+Value blend) · Quality 5 · Value -66 · Sentiment 35 (timing only, not weighted) · Composite fair value $284.62 vs $220.00 at analysis
Page map (sections in order; each card carries a stable
reference-name attribute you can cite):
profile-header/price-overview— company profile, live quote, market capextended-analysis— the core: three AI lens reads with findings, scores, and the analyst memofuture-predictions— our forward price-band predictionsmarket-narrative/ai-findings/gpt-critique— narrative context, cross-model findings, and an adversarial critique of our own analysis (near the end of the document)- Members-only sections (render as login gates for anonymous readers):
price-history,income-trend,key-metrics,financials(statement tables),insider-trading. The analysis above is public; the raw data tables require a free account.
More for machine readers: site briefing at
/llms.txt ·
any ticker resolves at delvantic.com/stock/TICKER ·
raw inputs are public-company filings and market data (via licensed data feeds);
every model, score, lens read, and prediction on this page is Delvantic's own analysis.
Capital One Financial Corporation
COF NYSECapital One Financial Corporation is a diversified financial services holding company headquartered in McLean, Virginia. The company operates primarily through three business segments: credit card, consumer banking, and commercial banking. In its credit card segment, Capital One provides a broad range of branded and co-branded credit card products to consumers and small businesses, serving both prime and subprime customer segments. Its consumer banking operations offer checking and savings accounts, money market deposits, certificates of deposit, auto loans, and other retail lending products, delivered through digital platforms, physical branches, and café-style locations in selected U.S. markets. The commercial banking segment focuses on lending, treasury management, deposit services, and capital markets solutions for mid-sized and large corporate clients, as well as real estate and industrial borrowers. Capital One Financial Corporation serves customers across the United States, Canada, and the United Kingdom, with a strong emphasis on digital banking and data-driven risk management. Founded in 1988, it is recognized today as a major player in consumer credit and broader retail and commercial banking services.
Price Overview
Price History (1 Year)
Revenue & Net Income Trend
| Period | Revenue | Net Income | Net Margin | YoY/QoQ |
|---|
Key Metrics
EPS (Diluted): 4.03
Total Equity: $113.62B
Shares: 608,684,864
Total Debt: $1.09B
Cash: $62.09B
EBITDA: N/A
Total Debt: $1.09B
Cash: $62.09B
Revenue: $53.43B
Revenue: $53.43B
Revenue: $53.43B
Total Equity: $113.62B
Tax Rate: 8.5%
Equity: $113.62B
Total Debt: $1.09B
Cash: $62.09B
Current Liabilities: N/A
Long-Term Debt: $0.00
Total Debt: $1.09B
Total Equity: $113.62B
Shares: 608,684,864
Shares: 608,684,864
CapEx: -$1.58B
Shares: 608,684,864
Stock Price: $221.63
Net Income: $2.45B
Industry Benchmarks
Income Statement (Annual)
Last updated: Aug 5, 2026 9:35am (18d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Revenue | $30.4B | $34.3B | $36.8B | $39.1B | $53.4B |
| Cost of Revenue | — | — | — | — | — |
| Gross Profit | — | — | — | — | — |
| Operating Expenses | $10.3B | $12.4B | $13.3B | $14.0B | $18.4B |
| Operating Income | — | — | — | — | — |
| Net Income | $12.4B | $7.4B | $4.9B | $4.8B | $2.5B |
| EBITDA | — | — | — | — | — |
| EPS | $27.04 | $17.98 | $11.98 | $11.61 | $4.03 |
| EPS (Diluted) | $26.94 | $17.91 | $11.95 | $11.59 | $4.03 |
Balance Sheet (Annual)
Last updated: Aug 5, 2026 9:35am (18d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Cash & Equivalents | $22.1B | $31.3B | $43.8B | $43.7B | $62.1B |
| Total Current Assets | — | — | — | — | — |
| Total Assets | $432.4B | $455.2B | $478.5B | $490.1B | $669.0B |
| Current Liabilities | — | — | — | — | — |
| Long-Term Debt | — | — | — | — | — |
| Total Liabilities | $371.4B | $402.7B | $420.4B | $429.4B | $555.4B |
| Total Equity | $61.0B | $52.6B | $58.1B | $60.8B | $113.6B |
| Retained Earnings | $51.0B | $57.2B | $60.9B | $64.5B | $65.2B |
Cash Flow (Annual)
Last updated: Aug 7, 2026 12:34am (16d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Operating Cash Flow | $12.3B | $13.8B | $20.6B | $18.2B | $27.7B |
| Capital Expenditure | -$698.0M | -$934.0M | -$961.0M | -$1.2B | -$1.6B |
| Free Cash Flow | $11.6B | $12.9B | $19.6B | $17.0B | $26.1B |
| Acquisitions (net) | -$669.0M | -$1.2B | -$2.8B | $0 | $16.5B |
| Net Debt Issued / (Repaid) | — | — | — | — | — |
| Dividends Paid | -$1.1B | -$950.0M | -$931.0M | -$932.0M | -$1.5B |
| Stock Buybacks | -$7.6B | -$4.9B | -$718.0M | -$734.0M | -$4.1B |
| Net Change in Cash | -$18.7B | $9.2B | $12.5B | -$84.0M | $18.4B |
Growth Trends (YoY %)
Last updated: Aug 5, 2026 9:35am (18d ago)| Metric | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|
| Revenue Growth | +12.5% | +7.4% | +6.3% | +36.6% |
| Gross Profit Growth | — | — | — | — |
| Operating Income Growth | — | — | — | — |
| Net Income Growth | -40.6% | -33.6% | -2.8% | -48.4% |
| EBITDA Growth | — | — | — | — |
Dividend History (Last 20)
Last updated: Aug 5, 2026 9:35am (18d ago)| Date | Dividend | Declaration | Record | Payment |
|---|---|---|---|---|
| 2026-05-19 | $0.80 | — | — | — |
| 2026-02-19 | $0.80 | — | — | — |
| 2025-11-17 | $0.80 | — | — | — |
| 2025-08-15 | $0.60 | — | — | — |
| 2025-05-23 | $0.60 | — | — | — |
| 2025-02-18 | $0.60 | — | — | — |
| 2024-11-14 | $0.60 | — | — | — |
| 2024-08-12 | $0.60 | — | — | — |
| 2024-05-10 | $0.60 | — | — | — |
| 2024-02-09 | $0.60 | — | — | — |
| 2023-11-10 | $0.60 | — | — | — |
| 2023-08-04 | $0.60 | — | — | — |
| 2023-05-12 | $0.60 | — | — | — |
| 2023-02-03 | $0.60 | — | — | — |
| 2022-11-10 | $0.60 | — | — | — |
| 2022-08-05 | $0.60 | — | — | — |
| 2022-05-13 | $0.60 | — | — | — |
| 2022-02-11 | $0.60 | — | — | — |
| 2021-11-12 | $0.60 | — | — | — |
| 2021-08-06 | $0.60 | — | — | — |
Deep Analysis
Risk : Reward — if the last quarter repeats
Live · as of 2026-08-19 08:44Recovery pays +223%; another quarter like the worst recent one costs 39%. Ratio 5.6:1. Caveat: revenue-DCF fits financials poorly (reported revenue omits net interest) — treat the legs as rough.
| Case | Growth | Margin | Fair value | vs price ($220.00) |
|---|---|---|---|---|
| Bull — recovery | +53% | 17.3% | $709.73 | +223% |
| Base — stabilizes | +36% | 15.0% | $387.21 | +76% |
| Bear — keeps slipping | +18% | 12.8% | $197.61 | -10% |
| Stress — last quarter repeats | +31% | 5.5% | $133.20 | -39% |
Narrative Economics
market-narrative step).
Claude Reading
The raw numbers tell a story the models are partially misreading. The revenue step-function from ~$10B/quarter through Q1 2025 to $12.5B in Q2 and $15.2-15.6B thereafter isn't organic growth — that's the Discover close (completed May 2025), and it makes every YoY and CAGR figure in the momentum block nonsense. The 20.5% "revenue CAGR" is a merger artifact. So is the -$4.28B Q2 2025 loss, which is almost certainly CECL Day 2 provisioning on the acquired Discover book plus deal costs. Strip that out and Q3-Q1 2026 run-rate NI is ~$2.5B/quarter or ~$10B annualized, putting normalized EPS closer to $15-16 and forward P/E around 14x — not the 54x headline. The pre-flight AI and thesis eval both nod at this ($15 normalized EPS, 13-14x), but the valuation synthesis anchoring on $159 fair value and the market-forces "catastrophic credit losses" thesis appear to be pricing off trailing GAAP earnings that include the merger charge. That's a material error.
The balance sheet reinforces this. $113.6B of equity against a $136B market cap is 1.18x P/B — for a bank that just absorbed Discover's payment network (an actual moat, not a story) and is generating $26B FCF, this is not a stretched multiple. Compare to JPM at ~2.3x, BAC ~1.3x. If Capital One earns even a 10% ROE on $114B equity, that's $11.4B in earnings, or 12x forward. The market-forces model calling COF a "deteriorating subprime lender" reads like a stale 2008-era template — COF's subprime card mix has been declining for a decade, and Discover's prime revolver book actually improves the credit mix. The bear case that "unemployment at 4.1% is the bottom" is legitimate cyclical risk, but it's a macro call, not a company-specific indictment.
Where I'd push back on the bulls: the insider activity is uniformly selling — nine sales versus one award grant over three months. Small share counts, but zero open-market buying at $221 tells you insiders don't see obvious upside from here. The 2.2% ROE and 0.37% ROA on TTM basis are genuinely ugly and won't normalize instantly — CECL provisioning on Discover will bleed into 2026 prints, and merger integration typically takes 24-36 months to hit synergy targets. The $2-3B Discover synergy number in the thesis eval is management's number; historical bank M&A delivers ~60-70% of promised synergies. Also, the dividend payout ratio of 61.8% on depressed earnings is misleading — the actual dividend cost is manageable, but the ratio optically constrains buyback capacity while integration completes. And the "$62B cash" line is deceptive for a bank; that's regulatory liquidity, not deployable capital.
Net: I dissent from the synthesis "overvalued at $147 fair value." That number is built on merger-distorted trailing earnings and treats a one-time CECL charge as run-rate. The narrative layer's 60/40 story-to-fundamentals split also overstates the story component — the Discover network is a hard asset with real economics, not a vibe. My read: on normalized ~$15 EPS and a 12-14x multiple appropriate for a diversified card+bank at mid-cycle credit, fair value is $180-210. Current $221 is modestly full but not egregious, and the setup into 2026-2027 (synergy realization + credit normalization off the CECL trough) is asymmetric to the upside if unemployment stays sub-4.5%. The real risk isn't the one the market-forces model flagged (catastrophic credit losses) — it's that Discover integration slips, synergies underdeliver, and the stock dead-money's for 18 months while earning its way into the multiple. I'd wait for a pullback to the $190s rather than chase, but I would not short this and I would not call $147 fair value defensible.
GPT Reading
What jumps out is that the stock is not expensive on banking metrics if you believe 2025 was a one-off earnings trough, but it is very expensive if you take the reported earnings at face value. At $221.63, the market cap is $135.97B against year-end equity of $113.62B, so investors are paying only 1.18x book for a franchise that historically should earn materially more than the 2.2% ROE shown for 2025. That is the key tension. Revenue exploded from $39.11B in 2024 to $53.43B in 2025, then stayed elevated with quarterly revenue running $15.23B-$15.58B over the last three quarters, versus just $10.00B in 2025's first quarter and roughly $10B run-rate in late 2024. Yet that surge in revenue did not translate into annual profit: net income fell from $4.75B to $2.45B, and there was a stunning -$4.28B loss in 2025-06. The raw data says this is not a clean growth story at all; it is an earnings-recovery story masked by top-line expansion.
The more important question is whether the recovery is already visible. I think yes, partially. After the June 2025 blowup, net income rebounded to $3.19B in September, then held at $2.13B and $2.17B in the next two quarters. That means the most recent two-quarter annualized earnings power is roughly $8.6B, and if you include the stronger September quarter, the latest three-quarter run-rate is over $10B. On that basis, the headline 54.6x P/E is close to useless. The stock is trading closer to 13x-16x an earnings power range of about $14B-$10B, which is far less demanding and roughly consistent with a large card/bank hybrid in mid-cycle conditions. Pair that with $27.72B of operating cash flow and $26.14B of free cash flow in 2025, plus $62.09B of cash against only $1.09B of stated debt, and the balance-sheet/cash-flow picture looks much sturdier than the income statement alone implies. I do not buy the “high-growth profitable” classification, but I also think the “catastrophic credit losses by 2027” framing is too dramatic relative to the evidence actually here. The data shows a cyclical hit, not franchise impairment.
That said, I still land on fairly valued to slightly rich because the market is already giving Capital One substantial credit for normalization. A 1.18x price-to-book multiple is not distressed pricing for a lender that just posted 4.6% net margin and 0.37% ROA for the year. The dividend payout ratio of 61.8% also looks high against depressed earnings, which suggests management is signaling confidence in forward earnings, but it leaves less room for error if the recovery stalls. And while recent quarterly profitability has stabilized, the margins in the last two quarters, 13.7% and 14.3%, are better than the 2024 trough but still nowhere near the 2021-2022 earnings power implied by annual net income of $12.39B and $7.36B. In other words, the stock does not look optically expensive versus normalized earnings, but it also is not offering a bargain multiple on proven current returns. At $221, you are paying for a comeback that is underway but not yet fully demonstrated.
The best argument against my caution is simple: if the June 2025 loss was tied to a singular event, then the market is still underestimating earnings power. The quarterly sequence after that loss is compelling, especially with revenue stepping up from a roughly $40B annual base to a $60B+ run-rate. If Capital One can convert even a mid-teens net margin on the current revenue base, annual earnings would move dramatically higher than the $2.45B posted in 2025. In that case, 1.18x book would be cheap, not fair, and the stock could deserve a move well above $250. The thesis evaluation's idea that the market is anchoring to depressed earnings is not crazy. Also, the low debt-to-equity and strong free cash flow provide more resilience than many consumer lenders get credit for, and if the Discover-related strategic upside is real, there is legitimate embedded optionality the plain numbers here do not capture.
What would change my mind is straightforward. If the next two quarters keep revenue above $15B and deliver net income above $2.5B each without another reserve-style shock, then the recovery case hardens and I would move bullish, especially if annualized ROE pushes back toward 8%-10%. Conversely, if revenue stays high but margins slip back toward the 10%-12% area, or if another quarter produces sub-$1.5B earnings despite this bigger revenue base, then the topline strength is low quality and the stock is too expensive above 1.1x book. For now, the market is not irrational, but it is also not leaving much valuation cushion.
Grok Reading
The raw numbers describe a company mid-transformation, not a slow-motion credit collapse. Revenue stepped from a $39B run-rate in 2024 to $53.4B in 2025, with the quarterly print jumping permanently from ~$10B to the mid-$15B range after mid-2025—clearly the Discover combination landing in the numbers. The $4.28B loss in Q2 2025 is the ugly accounting and provision spike that accompanies that deal; strip it out and the subsequent three quarters delivered $3.19B, $2.13B and $2.17B of net income, a ~$8.5–9B annualized run-rate. That is still well below the $12B peak of 2021, but it is triple the depressed $2.45B full-year 2025 print that produces the optically absurd 54.6x P/E. Free cash flow of $26.1B against a $136B enterprise, a fortress balance sheet ($62B cash, $1.1B debt, 0.01x D/E), and a 1.18x P/B on $114B of equity are the more relevant anchors. At current run-rate earnings the forward multiple compresses to roughly 15–16x—ordinary for a scaled consumer lender—while any path back even to a mid-single-digit ROE (let alone the historical 10–15%) makes the present book-value multiple look cheap rather than rich.
The market-forces claim that Capital One is a “deteriorating subprime lender” heading for negative earnings by 2027 sits in open contradiction with $27.7B of operating cash flow and three consecutive post-deal quarters of double-digit net margins. The 20.5% revenue CAGR and 36% recent y/y growth are real; the –29% earnings CAGR is almost entirely the arithmetic of one merger quarter and the industry-wide credit normalization that every card issuer has already booked. Insider sales are present but small and mechanical; they do not override the cash-generation picture.
The strongest counter-argument is straightforward and quantitative. Reported 2025 ROE is 2.2%, net margin 4.6%, and the five-year earnings path has marched steadily downhill from $12.4B to $2.45B. Unemployment at 4.1% has little room to fall and ample room to rise; every historical COF drawdown has been led by charge-off spikes that models systematically under-estimate at the turn. The valuation synthesis fair-value range of $147–159 implies the stock is already 33% too expensive, and the 61.8% payout ratio on trough earnings leaves little cushion if credit costs re-accelerate. A smart skeptic would also note that Discover synergies of $2–3B are still ungained and regulatory or integration failure remains a binary risk. I weigh these factors as real but already largely reflected in the 1.18x book multiple and the collapsed trailing P/E; they justify caution on position size, not a structural short.
Two concrete data points would flip the view: (1) two more quarters in which net charge-off rates continue rising and quarterly NI falls back below $1.5B, proving 2025 was not peak stress; or (2) explicit guidance that Discover cost and revenue synergies will undershoot $1.5B annually, removing the principal upside lever.
Big-3 Panel — where each AI stands
Advanced Analysis Forensic deep-dive · separate lenses
Capital One shows the profile of a well-capitalized consumer credit franchise: revenue expanded from $30.4B in 2021 to $53.4B in 2025 (with the 2025 step-up reflecting the Discover deal), operating cash flow is robust with OCF/NI at 4.44x, and reported FCF of $26.1B alongside $62B of liquid cash means near-term survival is not a question for this business. Accruals at -2.2% of assets and the strong OCF/NI ratio suggest reported earnings are backed by cash - the Altman Z of 0.36 flagged as distress is a false positive since Z-scores are not meaningful for banks, whose balance sheets are structurally leveraged by design. The soft spot is per-share value creation. Net income has fallen from $12.39B in 2021 to $2.45B in 2025 while diluted shares jumped from ~410M to 608.7M (roughly 49% dilution in one year, tied to the Discover stock-issued acquisition). Even normalizing for merger noise, that is a very large equity print, and EPS has collapsed across the window. Insider activity is exclusively selling (26 sales, 0 buys), though sizes are routine comp-driven dispositions rather than a red flag. Overall this reads as a competent, cash-rich, systemically important credit franchise in the middle of digesting a transformational acquisition. Integration risk, credit-cycle exposure, and the dilution math keep it from a higher tier, but the underlying business machinery looks intact.
Verify before trusting this (6)
- Discover acquisition synergy targets and integration milestones in the 10-K
- Credit card net charge-off and delinquency trends vs. reserves
- CET1 ratio and regulatory capital position post-merger
- Whether the 608.7M share count is the run-rate or includes further planned issuance
- Provision for credit losses trajectory and coverage ratios
- Segment breakdown of the $53.4B revenue - how much is legacy COF vs. Discover contribution
The price anchor is $220 against a composite FV of $159.09 and a signal-adjusted FV of $146.84, implying the market is paying a 38-50% premium to modeled deserved value. Even granting a Solid quality grade and the strategic logic of the Discover integration, that premium is hard to justify when share count is up ~50% year-over-year and net income per share has been trending down. The anchored-PE method corroborates the composite ($159), so the FV inputs look internally consistent rather than one runaway model - this is not a case of a busted DCF distorting the read.
Verify before trusting this (5)
- Discover integration cost synergies and revenue synergy timing in next 10-Q/transcript
- Reported vs adjusted EPS bridge and one-time acquisition charges
- Net charge-off trajectory and reserve build in card segment
- Deposit beta and NIM guidance
- Buyback pace once integration capital is freed
The tape is mildly risk-on (regime +46, VIX 15.2) and COF's beta of 1.02 means it participates cleanly in that lift without being whipped around. More importantly, the active narrative on this name has quietly shifted from 'subprime card lender' to 'diversified digitally-native bank with underwriting edge' - a steady-compounder story of moderate intensity and durability that Wall Street is actively validating (recent 36.6% run vs 20.5% long-term CAGR, coverage framing the stock as still below fair value despite a 102% three-year run). That is a textbook re-rating narrative in motion. Analyst/media tone is constructive: the headline flow is skewed toward 'looks cheap on intrinsic value' rather than 'credit cycle turning.' The Trump-account-closure story is noise - it is a compliance/AML story, not a franchise or regulatory-risk story, and the market has ignored it. Offsetting the tailwind: macro backdrop of a 10y at 4.63% and market PE 27.7 is a low-grade headwind for all equities, and the bear case (credit normalization, NIM compression) is a real overhang that limits how far sentiment can stretch. Net: the pressure is positive but not euphoric - a tailwind, not a mania.
Verify before trusting this (4)
- Next credit card delinquency and net charge-off prints - the single input that could crack the compounder narrative
- Whether analyst target revisions continue to trend up after the 102% run or start flattening (a sign the re-rating is done)
- Any escalation of the Trump-accounts story into a broader political/regulatory narrative
- 10y yield direction - a move above 4.75-5% would materially pressure the whole bank tape
This lens hasn't been run for this ticker yet.
When we made this prediction on Aug 7, 2026, COF was $220.00. We expect it to be $210.00 by Feb 2027, and we consider it great value under $165.00. This is an early model (v0.6.0) — the direction is more reliable than the exact price. Made Aug 7, 2026.
Blue is our prediction, starting the day we made it. Grey is a slower route to the same place — the same destination, taking longer. Black is the actual price, so you can see how we are doing.