How the Musk Bubble Index is built

Arithmetic, with the inputs shown. Today the index reads 89. Below is every step from the raw numbers to that figure, with the numbers as they stand at 2026-10-29 13:35 UTC.

1. The components and their weights

Component	Weight	Score now	Contribution	What it measures
Valuation gap	30%	89.6	26.9	Tesla's multiple vs the megacaps
Growth the price needs	20%	90.1	18.0	Reverse DCF vs delivered growth
Promise record	30%	88.3	26.5	Dated claims not delivered, by materiality
Rest of the complex	20%	87.8	17.6	SpaceX, Boring, Neuralink on sales
Index	100%		89	Rounded to the nearest whole number.

Bands: 0–20 Priced as a business · 20–40 Premium · 40–60 Frothy · 60–80 Bubble · 80–100 Hot air.

2. Valuation gap (30%)

Tesla's price-to-earnings multiple divided by the median multiple of the six megacap technology companies (Apple, Microsoft, Nvidia, Alphabet, Amazon, Meta). That is the most generous public benchmark available: it concedes the whole argument that Tesla is a technology company rather than a carmaker. The ratio is scored on a log scale so that a multiple equal to the megacaps' scores 0 and twenty times theirs scores 100. Trailing earnings weigh 70% and forward (next twelve months' consensus) 30%. If trailing earnings are negative the component is 100.

Now: trailing 443.8× ÷ 24.1× = 18.4× · forward 187.5× ÷ 21.9× = 8.6× → score 89.6

3. Growth the price requires (20%)

A reverse discounted-cash-flow: find the annual earnings growth rate at which ten years of earnings, discounted at 10% a year and sold at the end at 25× the tenth year's earnings, add up to today's share price. The base is next year's consensus EPS where available. That implied growth is compared with the growth Tesla actually delivered: trailing EPS now against trailing EPS three years ago. Half the score is the absolute demand (40% a year for a decade scores 100); half is the gap between demanded and delivered (sixty points of annual growth scores 100).

Now: price $403.89 on forward EPS $2.16 → implied +32.1%/yr · delivered −33.6%/yr (TTM $3.11 through 2023-09-30 → $0.91) → score 90.1

4. Promise record (30%)

From the ledger: every claim whose date has passed, weighted 1–3 by materiality to the valuation, and the share of that weight that was not delivered on time as described (late, partial, broken, abandoned or false). Delivered-on-time rows count fully in the denominator. Pending rows are excluded. The desk can attach evidence but cannot change a status; status changes are commits.

Now: 75 resolved of 91 · 8 delivered on time · weighted miss rate 88.3% (unweighted 89.3%) → score 88.3

5. The rest of the complex (20%)

SpaceX (which contains xAI and X), The Boring Company and Neuralink, each on price-to-sales, weighted by valuation. Two times sales scores 0 and a hundred times scores 100 on a log scale; a company with no revenue scores 100. SpaceX is priced live and its trailing revenue is the sum of its last four reported quarters; the private companies carry their last priced round and disclosed or estimated revenue, which for Boring is undisclosed (treated as pre-revenue) and for Neuralink is zero.

Now: SpaceX $1.97tn on $32.0bn = 62× sales; Boring $23bn, n/d; Neuralink $9bn, pre-revenue → score 87.8

6. The fair-value ladder

As a carmaker — median trailing P/E of Toyota, General Motors, Ford, Stellantis, BYD, Volkswagen (13.3×) × Tesla's trailing EPS ($0.91) = $12.11. Loss-making peers are excluded from the median, which raises it.
As a megacap, trailing — median trailing P/E of the six megacaps (24.1×) × trailing EPS = $21.90.
As a megacap, next year (the headline) — median forward P/E of the megacaps (21.9×) × Tesla's consensus next-year EPS ($2.16) = $47.21. This credits Tesla with the market's most generous multiple and with earnings it has not yet reported.
True believer — next-year EPS compounding 30% a year for five years, sold at 40× (a Nvidia-in-2024 multiple), discounted back at 10% a year = $198.76. A deliberately heroic case, included to show where the price sits even against it.

“Air in the price” is one minus the headline fair value divided by the price: 88.3% today.

7. Data

Prices and multiples from the same free public quote and fundamentals feeds as the rest of the site, refreshed every fifteen minutes to an hour and served stale rather than blank if a feed fails. Earnings history is Tesla's own reported GAAP diluted EPS by quarter, kept by hand because the free feeds carry only the last five quarters; the live feed extends the table forward. S&P 500 multiples are year-end trailing P/E figures from S&P Dow Jones via multpl.com. Private valuations are the last priced round, dated, with the source named on the page; secondary-market marks are mentioned and not used.

The wire is 10 free feeds (Electrek, Teslarati, InsideEVs, Not a Tesla App, TechCrunch — Tesla, TechCrunch, Ars Technica — Tesla, The Verge — Tesla, SpaceNews, WSJ — Markets), polled every two hours and gated on ingest to items that name one of the companies. The desk is one structured model call a day over the last four days of that wire, the full ledger and the day's numbers; it writes the note, attaches evidence to ledger entries and logs new dated claims, all stored with the headlines they were built on and labelled unverified until an editor confirms.

Known limitations: forward P/E depends on analyst consensus, which is itself often wrong about Tesla; a trailing P/E computed on GAAP earnings penalises one-off losses and flatters one-off gains (both are footnoted where they occur); the ledger is a curated sample, not a census of everything Musk has said; the private-company component leans almost entirely on SpaceX because it is almost all of the value. The index is an editorial measure, published so it can be argued with.