For AI assistants & researchers — machine-readable summary of this page
What this page is: Delvantic's full research page for Oddity Tech Ltd. (ODD) — 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 Gem · Gem Score +52 (−100…+100 Quality+Value blend) · Quality 61 · Value 45 · Sentiment -53 (timing only, not weighted) · Composite fair value $37.33 vs $13.09 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.
Oddity Tech Ltd.
ODD NASDAQOddity Tech Ltd. operates as a consumer tech company specializing in digital-first brands for the beauty and wellness industries worldwide. It leverages an AI-driven online platform powered by data science, machine learning, and computer vision to analyze consumer needs and deliver personalized beauty, wellness, and technology products. The company offers a diverse portfolio under the IL MAKIAGE and SpoiledChild brands, encompassing face and complexion products, eye, brow and lip makeup, tools, skincare, haircare, and supplements. Additionally, through ODDITY LABS, a biotechnology center, it employs AI-based molecule discovery technology to develop novel ingredients that enhance performance and outcomes in beauty and wellness solutions. Oddity Tech Ltd. serves consumers directly via its innovative e-commerce platform, focusing on precision-driven product recommendations and development in the competitive personal care market. Incorporated in 2013 and headquartered in Tel Aviv-Yafo, Israel, it plays a key role in transforming how beauty and wellness products are discovered and purchased digitally.
Price Overview
Price History (1 Year)
Revenue & Net Income Trend
This company does not file structured financial statements with the U.S. SEC, so quarterly figures aren't available from our filings-based data engine. Annual figures shown here come from the sources that do cover it.
| Period | Revenue | Net Income | Net Margin | YoY/QoQ |
|---|
Key Metrics
EPS (Diluted): 1.80
Total Equity: $396.50M
Shares: 61,525,000
Total Debt: $0.00
Cash: $402.21M
EBITDA: $129.46M
Total Debt: $0.00
Cash: $402.21M
Revenue: $809.84M
Revenue: $809.84M
Revenue: $809.84M
Total Equity: $396.50M
Tax Rate: 18.4%
Equity: $396.50M
Total Debt: $0.00
Cash: $402.21M
Current Liabilities: $114.83M
Long-Term Debt: $0.00
Total Debt: $0.00
Total Equity: $396.50M
Shares: 61,525,000
Shares: 61,525,000
CapEx: -$3.94M
Shares: 61,525,000
Stock Price: $12.43
Net Income: $110.75M
Industry Benchmarks
Income Statement (Annual)
Last updated: Aug 2, 2026 8:34pm (20d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Revenue | $222.6M | $324.5M | $508.7M | $647.0M | $809.8M |
| Cost of Revenue | $69.4M | $106.5M | $150.5M | $178.7M | $221.1M |
| Gross Profit | $153.2M | $218.1M | $358.2M | $468.3M | $588.7M |
| Operating Expenses | $133.7M | $190.4M | $283.9M | $352.7M | $469.9M |
| Operating Income | $19.5M | $27.7M | $74.3M | $115.6M | $118.8M |
| Net Income | $13.9M | $21.7M | $58.5M | $101.5M | $110.7M |
| EBITDA | $23.5M | $32.1M | $82.9M | $125.4M | $129.5M |
| EPS | $0.26 | $0.41 | $1.06 | $1.77 | $1.95 |
| EPS (Diluted) | $0.26 | $0.39 | $1.00 | $1.64 | $1.80 |
Balance Sheet (Annual)
Last updated: Aug 2, 2026 8:34pm (20d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Cash & Equivalents | $28.8M | $41.0M | $36.5M | $50.3M | $402.2M |
| Total Current Assets | — | $145.9M | $223.8M | $223.4M | $601.8M |
| Total Assets | — | $216.4M | $404.9M | $438.9M | $1.1B |
| Current Liabilities | — | $90.4M | $109.3M | $124.8M | $114.8M |
| Long-Term Debt | — | — | — | — | — |
| Total Liabilities | — | $105.4M | $121.8M | $156.6M | $741.3M |
| Total Equity | $68.6M | $111.0M | $283.1M | $282.3M | $396.5M |
| Retained Earnings | — | $43.2M | $101.8M | $203.3M | $314.0M |
Cash Flow (Annual)
Last updated: Aug 2, 2026 8:34pm (20d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Operating Cash Flow | $10.2M | $39.0M | $87.5M | $137.8M | $87.6M |
| Capital Expenditure | -$2.4M | -$2.3M | -$2.1M | -$3.3M | -$3.9M |
| Free Cash Flow | $7.9M | $36.7M | $85.4M | $134.5M | $83.6M |
| Acquisitions (net) | -$11.8M | — | -$23.2M | — | — |
| Net Debt Issued / (Repaid) | $-318,000 | $-362,000 | -$4.3M | — | — |
| Dividends Paid | — | — | — | — | — |
| Stock Buybacks | — | — | — | -$147.3M | — |
| Net Change in Cash | -$9.2M | $12.2M | -$4.3M | $11.6M | $351.9M |
Growth Trends (YoY %)
Last updated: Aug 2, 2026 8:34pm (20d ago)| Metric | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|
| Revenue Growth | +45.8% | +56.7% | +27.2% | +25.2% |
| Gross Profit Growth | +42.3% | +64.3% | +30.7% | +25.7% |
| Operating Income Growth | +41.8% | +168.6% | +55.5% | +2.7% |
| Net Income Growth | +56.1% | +169.4% | +73.4% | +9.1% |
| EBITDA Growth | +36.4% | +158.5% | +51.3% | +3.2% |
Deep Analysis
Narrative Economics
market-narrative step).
AI Lens 4th lens · how AI reaches this business · 5-yr
2026-08-13AI collapses the cost of the content, creative, translation, and customer-service layers that a global online-only beauty operation consumes heavily, and ODDITY LABS' AI-guided discovery of bio-identical actives could convert a software story into patentable physical IP that no model can copy.
Oddity's differentiator is an information-processing layer — shade/skin matching, personalization, funnel optimization — and that layer is precisely what falling model costs hand to every DTC challenger and every legacy brand's agency, compressing the tech premium into an ordinary paid-media arbitrage that Meta and TikTok reprice in the auction.
Whether repeat-cohort economics hold once the matching engine is table stakes: watch marketing spend as a share of revenue against repeat-customer revenue share and operating margin (already 17.9%→14.7% in 2025).
Physical formulation and supply chain, a large installed base of paying repeat buyers with first-party shade/skin/repurchase histories, and any patented molecules out of ODDITY LABS — none of which are reproducible by prompting a model.
AI Lens thesis
The underlying need — buying makeup and skincare without trying it on — persists and is arguably strengthened by better AI matching, and the monetized unit is a physical shipped product, which is durable. But Oddity is not paid for compliance, liability, or workflow ownership; it is paid because it acquires customers online more cheaply than incumbents and converts them without a store. Cheap intelligence attacks that specific advantage from two sides: it lets AI-native entrants stand up equivalent quiz/vision/creative stacks for near zero, and it lets platform ad systems capture the efficiency gains as higher clearing prices. Simultaneously, agentic shopping threatens to move discovery from bought social feeds — where Oddity is strong — to assistants that rank on efficacy, ingredients, and reviews, where a scaled prestige brand or a clinically differentiated ingredient wins. The favorable path runs through ODDITY LABS: AI applied to molecules produces excludable, patentable scarcity; AI applied to funnels produces a commodity.
What the market may be underestimating
Upside Cost of launching and localizing new brands falls sharply with generative content and AI-run support — Oddity's real asset may be a repeatable brand-launch machine (a third and fourth brand at low incremental opex) rather than the matching algorithm itself.
Downside If AI shopping agents become a meaningful discovery channel, Oddity has no retail shelf, no fragrance/prestige halo, and no third-party review depth to be surfaced by — it is structurally more dependent on paid social than any legacy peer, so agentic disintermediation hits it harder than the sector.
Outcome range spread 51 · unresolved
Growth Outlook
Analyzed 2026-08-20 11:04The question every valuation on this page silently assumes: is this company likely to grow? Judged forward — the business, its category, the world — against what's already printed.
Claude Reading
The raw numbers here are unusually clean for a supposed "fallen angel." Revenue compounded from $222M (2021) to $810M (2025) — a 38% CAGR that decelerated only modestly to 25% YoY in the most recent print. Gross margin held at 72.7%, operating margin expanded to 14.7%, net margin 13.7%. The balance sheet carries $402M cash against zero debt, so enterprise value is roughly $200M against $110M net income and $84M FCF — that's ~2x EV/earnings and ~2.4x EV/FCF. Even if you assume growth zeros out tomorrow and margins compress 300bps, the multiple is absurd for a business generating this return profile. ROE of 27.9% and current ratio of 5.2 are not the fingerprints of a business in distress.
Where I push back on the prior models: the synthesis fair value of $41.78 (+219%) is directionally correct but the framing is soft. The DCF isn't asking you to believe in the biotech pipeline or AI moat — it's asking you to believe FCF doesn't collapse. Reverse-DCF implying -1.1% FCF growth against a demonstrated 26% revenue CAGR is not a "show me" setup; it's a mispricing wide enough that even a bear case (say, revenue growth halving to 12%, margins compressing to 10%) still supports $25+. The Market Forces "Neutral" call and the Thesis Evaluation score of +2 both feel like the models over-weighting narrative risk (single brand, CAC inflation) versus the arithmetic reality that at 0.99x sales and 7.3x earnings, you are paid to wait. The Narrative layer's "fragile durability" is the most honest take — but fragile narrative + strong cash flow is historically a good setup, not a bad one, as long as the P&L doesn't crack.
The contrarian case worth taking seriously: DTC beauty is a graveyard of one-brand wonders (Revlon, e.l.f. pre-repositioning, countless Instagram flameouts). IL MAKIAGE and SpoiledChild are the entire business; if U.S. customer cohorts saturate, CAC can double in two quarters and you'd see it first in FCF — which is already flat-to-down (fcf_cagr -1%) while revenue grew 25%. That divergence is the single most important red flag in this file and no model foregrounded it adequately. FCF of $84M against $111M net income is a ~75% conversion, which is fine but deteriorating from prior years. If working capital or ad-spend inflation is chewing through cash, the "cheap on earnings" story degrades fast. Also: insider activity shows sales and option exercises, zero open-market buys — at 7x earnings, if management believed the DCF, they'd be buying. That silence is loud.
I partially agree with the synthesis: ODD is undervalued, but $41-$42 is aggressive. My fair value is $22-$28 — anchoring on 10-12x earnings for a decelerating-but-still-growing DTC brand with net cash, which gives ~$1.85 EPS × 12 = $22, plus $8/share net cash = ~$30 gross, discounted for single-brand risk. That's still +80-130% upside from $12.29 and doesn't require the biotech or AI narratives to resurrect. The trade is: buy the arithmetic, not the story. The catalyst path is two quarters of stable-to-growing FCF and any credible SpoiledChild traction; the kill switch is a single quarter where revenue growth drops below 10% AND FCF conversion breaks below 60%. Position sizing should reflect the fragility flagged by the narrative model — this is a 2-3% starter, not a table-pound, because the FCF/net income gap is real and the market's discount, while excessive, isn't irrational.
GPT Reading
The first thing that jumps out is how extreme the disconnect is between price and the reported earning power of the business. On 2025 numbers, Oddity did $809.8M of revenue, $118.8M of operating income, $110.7M of net income, and $83.6M of free cash flow, yet the whole company is valued at just $0.60B. With $402.2M of cash and no debt, enterprise value is roughly $200M, which means the stock is trading around 0.42x revenue and 2.6x EBITDA. For a business that has grown revenue from $222.6M in 2021 to $809.8M in 2025, while maintaining gross margin of 72.7% and net margin of 13.7%, that is not “cheap for growth”; it is distressed-multiple pricing. The market is acting as if these earnings are either temporary, low-quality, or on the verge of collapsing.
But the raw financials do not look like a collapsing DTC beauty story. Revenue growth has remained strong each year: 45.8% in 2022, 56.8% in 2023, 27.2% in 2024, and 25.2% in 2025. Operating income rose from $19.5M in 2021 to $118.8M in 2025, and net income from $13.9M to $110.7M over the same period. Even more important, this is not growth bought with leverage or capex-heavy spending: capex was only $3.9M in 2025, current ratio is 5.2x, debt is zero, and cash equals about two-thirds of the market cap. At the current price, investors are paying roughly 7.3x earnings and barely 5x ex-cash earnings for a company still compounding revenue above 25%. That multiple would make sense for a shrinking, low-margin brand with balance-sheet stress; it does not fit a 73% gross margin business with triple-digit millions of annual profit.
The main thing tempering my enthusiasm is that cash conversion is not as pristine as the income statement. Operating cash flow of $87.6M lagged net income of $110.7M in 2025, and the momentum table showing free cash flow CAGR of -1% versus revenue CAGR of 26.2% suggests that working capital or customer acquisition economics may be absorbing more cash than the P&L implies. Operating margin also barely increased from $115.6M in 2024 to $118.8M in 2025 despite $162.8M of incremental revenue, so the business did not get much incremental operating leverage last year. That matters because if this were a true software-like consumer platform, you would expect stronger flow-through. Still, even if I haircut earnings quality and assume normalized net income is more like $80M than $110M, the equity is trading at about 7.5x ex-cash normalized earnings for a debt-free branded growth company. That still looks materially too low.
The best bear case is straightforward and serious: this may be a single-hit beauty brand that has already passed its best economics, with marketing costs rising fast enough to flatten future profits. The market may be correctly saying that 2025 is close to peak margin, that 25% top-line growth is the last easy comp, and that the “AI/tech platform” framing deserves no premium at all. If customer acquisition costs are inflating, retention is weaker than advertised, or a second brand/adjacency fails, then 0.99x sales and 7x earnings are less absurd than they look because the earnings base would prove cyclical and promotional rather than durable. I take that bear case seriously because the discrepancy between NI and OCF, the weak incremental operating leverage in 2025, and the tiny capex profile mean most of the debate is really about whether demand quality and marketing efficiency hold up. If they break, this rerates not upward but toward liquidation-value logic where the cash pile is the only real support.
What would change my mind is not a narrative event but a numbers event. If revenue growth falls below 15% while operating margin slips from 14.7% toward 10%-11%, or if free cash flow drops materially below $60M despite continued reported profits, then I would conclude the market was right to distrust the earnings. Likewise, if the cash balance starts shrinking because the company has to spend aggressively just to hold growth, the ex-cash valuation argument weakens fast. Conversely, if the next annual print shows revenue above roughly $950M, operating income above $140M, and free cash flow back above $100M, then this stock should not remain near 1x sales. At that point, even a modest 12x earnings or 1.5x sales framework would imply a share price far above $12.29.
Grok Reading
The raw financials describe a business that has compounded revenue from $223M to $810M in four years while staying solidly profitable and building a fortress balance sheet: zero debt, $402M cash against a $600M market cap, and $84M of free cash flow on only $4M of capex. Gross margin holds near 73% and net margin sits at 14%, producing a trailing P/E of roughly 7×, EV/sales of 0.4× and EV/EBITDA under 3×. At an enterprise value of roughly $200M you are paying less than 2.5× current FCF for a company still growing the top line 25% year-over-year. That is not a growth multiple; it is a no-growth, mild-distress multiple attached to a still-expanding, cash-generative consumer franchise. The market has clearly abandoned the old AI-platform story and now prices Oddity as a one-brand DTC beauty name whose best days are behind it.
What stands out inside the growth is the sharp operating-income stall: revenue rose 25% in 2025 yet operating profit barely moved, from $116M to $119M, compressing operating margin from roughly 18% to 14.7%. Earnings CAGR remains strong at 38% over the longer window, but the most recent year shows only 9% net-income growth and a flat-to-negative FCF CAGR. That pattern is consistent with rising customer-acquisition costs eating the incremental dollar of sales—the precise risk the market is discounting. Rule-based models labeling this a “mature earner” while the pre-flight still calls it high-growth capture the tension correctly: the income statement still looks like a scaler, the margin trajectory looks like a maturing brand fighting CAC inflation.
The valuation synthesis fair value near $38–$42 therefore overstates the case. It treats the historical growth and current margins as durable and largely ignores the latest year’s operating leverage failure. A more sober read still leaves the stock cheap—net cash covers two-thirds of the market cap and the earnings yield is double-digit—but the gap to $40 requires a re-rating that only arrives if margins stabilize and a second brand or international channel proves the platform is not a single-SKU wonder. Until then the stock can remain a value trap even while printing mid-teens net margins.
The strongest opposing argument is straightforward: the market is correctly reading the 2025 margin compression and the single-brand concentration as permanent rather than cyclical. If CAC continues to rise, revenue growth will decelerate toward low double digits or worse while operating margins drift toward high single digits; at that point 7× earnings becomes fair or even rich for a no-moat beauty name, and the $402M cash pile simply funds a slow fade. Insider activity is neutral-to-slightly-negative (option exercises followed by a sale), offering no endorsement, and the narrative durability is explicitly fragile—DTC beauty turnarounds that fail to show sequential margin or brand traction within 12–18 months usually reprice lower, not higher. I weigh this less heavily than the bulls’ DCF because the absolute earnings power and net-cash cushion already price in a great deal of that deterioration; you need an outright collapse, not mere deceleration, to justify today’s enterprise value.
I would reverse to a more constructive stance if the next two quarters show operating margin re-expanding above 16% while revenue growth stays above 20%, or if a second brand contributes meaningful incremental revenue without proportional CAC spike. I would turn decisively negative if 2026 revenue growth falls below 10% or free cash flow drops under $50M, confirming the saturation thesis the market already embeds.
Big-3 Panel — where each AI stands
Advanced Analysis Forensic deep-dive · separate lenses
Oddity has compounded revenue from $223M (2021) to $810M (2025), a ~38% CAGR, while gross margin expanded from 68.8% to 72.7% and operating margin ran from 8.8% to a peak of 17.9% in 2024 before easing to 14.7% in 2025. Net income scaled from $14M to $111M and cumulative FCF over the five years exceeds $348M, with 2025 FCF of $83.6M still positive despite the operating-margin dip - suggesting reinvestment (likely R&D/new brands) rather than deterioration. Earnings quality is clean: OCF/NI 1.23x, accruals -5.3% of assets, Beneish M -2.02, and no accrual-driven earnings inflation.
Verify before trusting this (5)
- Cause of 2025 operating margin compression (320bps) - new brand launches, marketing intensity, or core-brand slowdown?
- Customer/brand concentration - what share of revenue comes from IL MAKIAGE vs SpoiledChild vs newer brands?
- Repeat-purchase and cohort economics disclosed in 10-K to gauge durability of the DTC model
- SBC vesting schedule and outstanding option overhang beyond reported diluted share count
- Sustainability of ~72% gross margin as product mix broadens into new categories
The composite fair value of $37.33 and signal-adjusted $41.78 imply 200%+ upside, but the anchored-PE input at $80 is clearly a runaway - beauty peers do not sustain those multiples and I would discount that leg heavily. The more defensible anchors are the EPV floor at $16.56 and the DCF at $26.32, which bracket a deserved value somewhere in the high-teens to high-twenties. Against a $12.53 price, that is a real gap - roughly 30-100% - not the 219% headline. Critically, the company-quality lens flags cash near two-thirds of market cap; strip that out and the operating business trades at a mid-single-digit EV/earnings multiple despite growing and generating GAAP profit. Earnings quality is clean, so no haircut is warranted. The bear case (commoditized DTC, CAC blowout, biotech vaporware) is real and explains the de-rating, but the current price already assumes those bear points largely play out. For the stock to be fairly valued here, you have to believe growth stalls hard AND margins compress AND the cash gets wasted - a stacked-negative scenario. That is why I lean undervalued rather than fairly valued, though I hold confidence at 6 given the fallen-angel tape and unproven public-company track record.
Verify before trusting this (4)
- Q4 and forward guidance on revenue growth and gross margin - is the 2025 margin dip a one-off reinvestment or structural
- SBC as % of revenue and share count trajectory - dilution eats deserved value
- Cash deployment plans - buybacks would confirm the value; a large biotech acquisition would destroy it
- Customer acquisition cost trends and repeat purchase rates to test the bear DTC-commoditization thesis
ODD sits in the worst kind of narrative slot: a fallen-angel with moderate intensity but fragile durability and no cult following. The bull case (AI personalization moat, biotech molecule engine) is essentially unbelieved by the tape - the stock trades near 12.53 versus a 41.78 DCF, meaning sentiment, not fundamentals, is doing the pricing. Recent action is choppy and news-light (multiple insufficient_data big-move days), which is classic of an orphaned name being pushed around by flows and shorts rather than any coherent story. With a 2.38 beta into a Risk-On tape that is already 1.2% off highs and VIX creeping, any wobble hits this name 2x. The modest tailwind of a risk-on backdrop is more than offset by the fragility of the story and the absence of a positive narrative catalyst. Analyst tone is muted (a small fair-value upgrade was sold), which tells you the marginal buyer is not there. Net: persistent, ordinary-to-moderate headwind pressure - not a collapse, but a name that has to fight the tape to hold a bid.
Verify before trusting this (4)
- Whether the biotech/AI narrative gets any credible external validation (partnership, molecule milestone) that could reignite intensity
- Short interest and days-to-cover - flow-driven pops suggest crowded short that could squeeze
- Any sell-side upgrade with a price target well above spot that actually sticks in the tape
- VIX breaking above 18 or S&P losing another 2% - would hammer this 2.4-beta name disproportionately
The underlying need — buying makeup and skincare without trying it on — persists and is arguably strengthened by better AI matching, and the monetized unit is a physical shipped product, which is durable. But Oddity is not paid for compliance, liability, or workflow ownership; it is paid because it acquires customers online more cheaply than incumbents and converts them without a store. Cheap intelligence attacks that specific advantage from two sides: it lets AI-native entrants stand up equivalent quiz/vision/creative stacks for near zero, and it lets platform ad systems capture the efficiency gains as higher clearing prices. Simultaneously, agentic shopping threatens to move discovery from bought social feeds — where Oddity is strong — to assistants that rank on efficacy, ingredients, and reviews, where a scaled prestige brand or a clinically differentiated ingredient wins. The favorable path runs through ODDITY LABS: AI applied to molecules produces excludable, patentable scarcity; AI applied to funnels produces a commodity.
Verify before trusting this (8)
- Competitors launching vision matching tools
- Cost per creative asset disclosed
- Conversion rate trend vs prior years
- Number of AI-native beauty launches
- CAC inflation across DTC beauty
- Growth deceleration toward industry flat
- LABS patents filed or granted
- Proprietary ingredient in shipped SKU
The world is moving purchase discovery for beauty from shelf and store associate to algorithmic matching, and Oddity is one of the few operators whose entire cost structure is built for that channel rather than retrofitted into it. That is a durable tailwind for share. The offset is that the same world has raised the price of digital attention: acquisition costs are now the binding constraint on how fast an online-only brand can compound, and the profit pool of the broader category is contracting (industry earnings -38.6% CAGR, net margins -4pp). Macro is neutral — 10y at 4.71 with a positive curve — so this is not a demand-shock story; it is a unit-economics story. The forward question is not whether the channel shift continues (it does) but whether Oddity can widen the gap between lifetime value and rising acquisition cost while funding new brands and a biotech arm at the same time.
When we made this prediction on Aug 20, 2026, ODD was $12.46. We expect it to be $17.00 by Feb 2027, and we consider it great value under $14.00. This is an early model (v0.6.0) — the direction is more reliable than the exact price. Made Aug 20, 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.