# The ETFIQ Skeptic: what an ETF's name leaves out

> The ETFIQ Skeptic on the IQ Investment Committee: 13 criteria sourced to ETFIQ's own measures, their methodology pages, and US law.

Data as of Oct 21, 2026. Rebuilt every trading night.

Version 2.1, published Oct 21, 2026: 13 criteria, 11 sources, 0 of the 13 measures backed by peer-reviewed research.

## The ETFIQ Skeptic: what an ETF's name leaves out

ETFIQ’s own seat tests the other seats’ findings with 13 criteria on ETFIQ data, version 2.1 of Oct 21, 2026. What does the name leave out?

13 Criteria

11 Sources

2.1 Version

### The The ETFIQ Skeptic seat

The ETFIQ Skeptic is one seat of the IQ Investment Committee. It applies ETFIQ's own measures, their methodology pages, and US law to ETFIQ data.

### Scope

Every ETF. The Skeptic sits by default and can be removed.

What it must read first: The Skeptic reports no objection only after reading the measure that matters for the fund's category: the Promise check for a leveraged, buffer, coin or alternatives fund (the daily multiple compounded, the buffer delivered, the coin tracked); payout coverage for an income fund; the gap to its own mix for an allocation fund; the gap to its own benchmark for a stock, bond or commodity fund; overlap with its benchmark for a theme fund. Until that measure is read it says Not enough data, never no objection.

Source strength: ETFIQ’s own seat, tied to the Promise Index tests. ETFIQ's own seat, using only ETFIQ's measures, each cited to the ETFIQ methodology page that defines it, and the two tests US law sets for a diversified fund. It has no tradition and no byline. Four of its measures (the Promise badge; liquidity, spread and premium from the Trading hub) are pending: until their pages publish them they are printed as pending and left out of what the seat must read to clear a fund.

Lenses apply publicly known investing principles to ETFIQ data. No investor named is affiliated with or endorses ETFIQ. Not investment advice.

### What ETFIQ’s methods say

In ETFIQ’s words, with the source of each. Sourced to ETFIQ's own measures, their methodology pages, and US law.

- Every figure ETFIQ publishes says what it is, where it comes from and when; the Skeptic reads only those figures. (ETFIQ, Methodology: how every ETFIQ figure is built (ETFIQ, methodology page, 2026))

- A fund's own benchmark is the index it names, else the broad index its own report names; the gap is measured against that benchmark only. (ETFIQ, How a fund is placed (categories methodology) (ETFIQ, methodology page, 2026); ETFIQ, Income ETF methodology (ETFIQ, methodology page, 2026))

- Overlap counts, for every security two books share, the smaller weight; cash is never counted. (ETFIQ, Portfolio methodology (ETFIQ, methodology page, 2026))

### Sources

| Source | What the seat takes from it | Edition | Link |
| --- | --- | --- | --- |
| ETFIQ, Income ETF methodology | The gap to a fund's own benchmark, and what an issuer's Rule 19a-1 return of capital estimate does and does not tell you | ETFIQ, methodology page, 2026 | source |
| ETFIQ, Portfolio methodology | Weight overlap between two books, cash left out, and how a portfolio is looked through to its securities | ETFIQ, methodology page, 2026 | source |
| ETFIQ, Fund flows methodology | Daily flows from the change in shares outstanding, and the funds whose flows are not read | ETFIQ, methodology page, 2026 | source |
| ETFIQ, Methodology: how every ETFIQ figure is built | The rule that a figure the data cannot support is left out and the gap said, never approximated | ETFIQ, methodology page, 2026 | source |
| ETFIQ, How a fund is placed (categories methodology) | Each core fund's own benchmark: the index it names, else the broad index its own report names | ETFIQ, methodology page, 2026 | source |
| ETFIQ, The ETFIQ Promise Index | One test per category of whether a fund delivers what its category promises | ETFIQ, methodology page, 2026 | page to come with the measure |
| ETFIQ, Trading and liquidity methodology | The IQ Liquidity Score, quoted spreads and premiums and discounts from issuers' Rule 6c-11 figures | ETFIQ, methodology page, 2026 | page to come with the measure |
| US Congress, Investment Company Act of 1940, section 5(b)(1) (15 U.S.C. 80a-5(b)(1)) | The definition of a diversified fund: at least 75% of assets in cash, government securities, other funds and positions each no more than 5% of assets and 10% of the issuer's voting shares | United States Code, current text, 1940 | source |
| US Congress, Internal Revenue Code section 851(b)(3) (26 U.S.C. 851(b)(3)) | The diversification a fund needs to be taxed as a regulated investment company: at least 50% of assets in positions each under 5%, and no more than 25% in any one issuer | United States Code, current text, 1986 | source |
| ETFIQ, ETFIQ benchmarks | The Strategic family, x% VT and the rest US bonds, at every stock share | ETFIQ, rulebook 1.3, 2026 | source |
| ETFIQ, Methodology: scores | The rate cushion: 12-month income yield over duration, the bonds score | ETFIQ, 2026 | source |

### Research behind the measures

Peer-reviewed research on how each measure is taken. It never adds a criterion, and where it weakens one the criterion says so.

- Michael J. Cooper, Huseyin Gulen and P. Raghavendra Rau (2005), Changing Names with Style: Mutual Fund Name Changes and Their Effects on Fund Flows , The Journal of Finance 60(6), 2825-2858 ( doi:10.1111/j.1540-6261.2005.00818.x ). Funds that renamed themselves after a popular style drew about 28% more money the next year with no better performance, and flows rose whether or not the holdings matched the new name.

- Kenneth R. French (2008), Presidential Address: The Cost of Active Investing , The Journal of Finance 63(4), 1537-1573 ( doi:10.1111/j.1540-6261.2008.01368.x ). From 1980 to 2006 US investors spent 0.67% of the market's value a year on fees, expenses and trading costs in search of better returns; holding the passive market portfolio instead would have raised the typical investor's return by about that much.

- William F. Sharpe (1991), The Arithmetic of Active Management , Financial Analysts Journal 47(1), 7-9 ( doi:10.2469/faj.v47.n1.7 ). Before costs the average actively managed dollar must earn what the average passive dollar earns, because together they are the market. Active management costs more, so after costs the average active dollar must trail, in every period.

### Definitions and method

Each criterion reads one ETFIQ measure against two thresholds. The Skeptic objects once: any criterion that Challenges makes the seat Challenge, whatever else it read. It reports no objection (Supports) only when every criterion it could read supports and it read at least half of them; with less read it says Not enough data rather than clearing a fund it could not check. Anything between is Mixed.

The weights. The Promise check carries the most (14%) because it tests the fund's own promise; the other criteria weigh 6% or 8%. The weights only order the score: one Challenge is an objection whatever its weight. They are ETFIQ's and were not set to land any fund.

The Promise badge (ETFIQ’s measure) Does the fund deliver what its category promises? Measure: the Promise check. The fund's result on its own category's Promise Index test: delivers, partly or fails; null where the test cannot be read. The test, its measure and its thresholds are the Promise Index's, never the Skeptic's own. Supports where the Promise check equal to delivers. Challenges where the Promise check equal to fails. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. The Promise Index's own test and thresholds: delivers, partly or fails. Pending: its measure is not published yet, so the criterion is printed as pending and left out of what the seat must read. Weight: 14%. Sourced to: ETFIQ, The ETFIQ Promise Index (ETFIQ, methodology page, 2026). Tests the finding of: every seat. Research, context, not a test of the measure: Michael J. Cooper, Huseyin Gulen and P. Raghavendra Rau (2005), Changing Names with Style: Mutual Fund Name Changes and Their Effects on Fund Flows , The Journal of Finance 60(6), 2825-2858. Investors follow a fund's name whether or not its holdings match it; the paper studies flows after name changes, not whether a fund delivers what its category promises. Full cost to own against the benchmark ETF (ETFIQ’s measure) With the spread added to the fee, how much more does a year cost than owning the benchmark ETF? Measure: the full cost against the benchmark. Full cost to own for a year, assuming one round trip a year: the net expense ratio plus the median quoted bid-ask spread over the last 30 trading days as a percent of price, minus the same figure for the ETF the fund's own benchmark is measured through, percentage points. Null where the fund has no benchmark ETF or either spread is not read. Supports where the full cost against the benchmark at or below 0.05 points. Challenges where the full cost against the benchmark at or above 0.2 points. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. Within 0.05 points of the benchmark ETF, and 0.20 points or more above it, are ETFIQ thresholds. Pending: its measure is not published yet, so the criterion is printed as pending and left out of what the seat must read. Weight: 8%. Sourced to: ETFIQ, Trading and liquidity methodology (ETFIQ, methodology page, 2026); ETFIQ, Income ETF methodology (ETFIQ, methodology page, 2026). Tests the finding of: Low Cost. Research, context, not a test of the measure: Kenneth R. French (2008), Presidential Address: The Cost of Active Investing , The Journal of Finance 63(4), 1537-1573. Trading costs belong in the cost of investing alongside fees, which is why the measure adds the spread; the paper measures the cost, it does not test this comparison. Research, context, not a test of the measure: William F. Sharpe (1991), The Arithmetic of Active Management , Financial Analysts Journal 47(1), 7-9. Against the same holdings every point of extra cost is a point less return; a fund and its benchmark do not always own the same things, so the arithmetic is context. The gap to its own benchmark (ETFIQ’s measure) Over a full year, how far did the fund finish from the benchmark it is measured against? Measure: one year against its benchmark. Total return minus the fund's own benchmark's over a full year, percentage points (core and income funds). Supports where one year against its benchmark at or above -0.5 points. Challenges where one year against its benchmark at or below -2 points. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. No more than half a point behind, and 2 points or more behind, are ETFIQ thresholds. An allocation fund is not measured against an all-stock index: it has its own criterion, against the ETFIQ benchmark at its own stock share. Weight: 8%. Applies to: the category one of stocks, bonds, commodities, income. Sourced to: ETFIQ, Income ETF methodology (ETFIQ, methodology page, 2026); ETFIQ, How a fund is placed (categories methodology) (ETFIQ, methodology page, 2026). Tests the finding of: Low Cost, Trend Following. Research: none cited; the criterion rests on the source alone. The gap to the ETFIQ benchmark at the same stock share (ETFIQ’s measure) For an allocation fund: over a year, how far did it finish from a simple low-cost mix with its own stock share? Measure: one year against its stock-share benchmark. For an allocation fund: its 1-year total return minus that of the ETFIQ Strategic benchmark whose stock share is nearest the fund's own look-through stock share, over the same days, percentage points. The benchmark is named on every reading. Null where the fund has under a year of prices or its stock share is not read. Supports where one year against its stock-share benchmark at or above -0.5 points. Challenges where one year against its stock-share benchmark at or below -2 points. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. The same lines as the benchmark gap. A 30/70 fund is measured against a 30/70 mix, never against an all-stock index such as ACWI. Weight: 8%. Applies to: the category equal to allocation. Sourced to: ETFIQ, ETFIQ benchmarks (ETFIQ, rulebook 1.3, 2026); ETFIQ, How a fund is placed (categories methodology) (ETFIQ, methodology page, 2026). Tests the finding of: Low Cost, All-Weather, Diversification. Research: none cited; the criterion rests on the source alone. Return of capital share (ETFIQ’s measure) How much of what the fund paid out was the holder's own money handed back? Measure: the return of capital share. The share of the fund's distributions over its latest twelve months of notices, weighted by amount, that its issuer's Rule 19a-1 notices estimate as return of capital, percent, printed with the two dates its notices cover. Printed only between 0 and 100, with the number of notices it rests on. Null where no notice was published or read, never 0, and null where the twelve months' notices carry no dates: a window is never printed as twelve months without them. Supports where the return of capital share at or below 10%. Challenges where the return of capital share at or above 50%. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. 10% or less, and half or more, of distributions estimated as return of capital are ETFIQ thresholds. Weight: 8%. Applies to: the category equal to income. Sourced to: ETFIQ, Income ETF methodology (ETFIQ, methodology page, 2026). Tests the finding of: Cycles & Risk. Research: none cited; the criterion rests on the source alone. The payout covered by total return (ETFIQ’s measure) For an income fund: did the year's total return cover the cash it paid, or did the payout eat capital? Measure: income coverage. 100 times the 1-year total return over the 1-year cash paid (income funds). Supports where income coverage at or above 100%. Challenges where income coverage at or below 50%. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. A total return at least equal to the cash paid, and under half of it, are ETFIQ thresholds. A payout the return did not cover came out of the holder's capital. Weight: 8%. Applies to: the category equal to income. Sourced to: ETFIQ, Income ETF methodology (ETFIQ, methodology page, 2026). Tests the finding of: Cycles & Risk, Trend Following. Research: none cited; the criterion rests on the source alone. The rate cushion (ETFIQ’s measure) For a bond fund: does a year's income cover the loss from a rise in rates? Measure: the rate cushion. The 12-month income yield divided by duration: how far rates can rise in a year before the year's income is gone (bond funds). Supports where the rate cushion at or above 1 points. Challenges where the rate cushion at or below 0.25 points. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. A year's income covering a 1-point rise in rates, and a quarter point or less, are ETFIQ thresholds. It is ETFIQ's bonds score measure. Weight: 6%. Applies to: the category equal to bonds. Sourced to: ETFIQ, Methodology: scores (ETFIQ, 2026). Tests the finding of: Low Cost, All-Weather. Research: none cited; the criterion rests on the source alone. Overlap with its benchmark ETF (ETFIQ’s measure) For a fund sold as something different: how much of it is the benchmark ETF you could own instead? Measure: the overlap with its benchmark. Weight overlap between the fund's book and its benchmark ETF's book: for every security both own, the smaller weight, summed; cash rows left out (the overlap rule /methodology/portfolio states). Supports where the overlap with its benchmark at or below 40%. Challenges where the overlap with its benchmark at or above 70%. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. 40% or less in common, and 70% or more, are ETFIQ thresholds. Printed with: the share of the book matched to its benchmark's: The share of the fund's weight that joined the benchmark ETF's book at all, printed beside the overlap so a 0 that is a true 0 can be told from a join that found nothing. Weight: 8%. Applies to: the category one of themes, alternatives. Sourced to: ETFIQ, Portfolio methodology (ETFIQ, methodology page, 2026). Tests the finding of: Disruptive Innovation, Diversification. Research: none cited; the criterion rests on the source alone. The largest single company (ETFIQ’s measure) Does one company carry more of the fund than its name suggests? Measure: the largest issuer. The weight of the largest single issuer, over the same issuers as the ten-largest measure. Null on the same terms. Supports where the largest issuer at or below 10%. Challenges where the largest issuer above 25%. Mixed between. Threshold: Supports: ETFIQ’s reading; Challenges: set in law (US Congress, Internal Revenue Code section 851(b)(3) (26 U.S.C. 851(b)(3)) (United States Code, current text, 1986)). 10% or less is ETFIQ's line. More than 25% in one issuer fails the tax code's single-issuer cap (26 U.S.C. 851(b)(3)(B)), which is law. Weight: 6%. Applies to: the category one of stocks, themes, income, allocation, alternatives. Sourced to: ETFIQ, Portfolio methodology (ETFIQ, methodology page, 2026); US Congress, Internal Revenue Code section 851(b)(3) (26 U.S.C. 851(b)(3)) (United States Code, current text, 1986). Tests the finding of: Diversification. Research: none cited; the criterion rests on the source alone. Diversified under the law's two tests (ETFIQ’s measure) How much of the fund sits in companies each over 5% of it, against the limits the law sets? Measure: the weight in issuers over 5%. The summed weight of every issuer above 5% of net assets, over the same issuers as the ten-largest measure. US government and agency securities are left out, as the 1940 Act's test leaves them out. A holding that is itself a fund is read through to what that fund owns: the Act would leave the fund's own row out, while ETFIQ reads the issuers the holder is exposed to, and says so. Null on the same terms as the ten-largest measure. Supports where the weight in issuers over 5% at or below 25%. Challenges where the weight in issuers over 5% above 50%. Mixed between. Threshold: Supports: set in law (US Congress, Investment Company Act of 1940, section 5(b)(1) (15 U.S.C. 80a-5(b)(1)) (United States Code, current text, 1940)); Challenges: set in law (US Congress, Internal Revenue Code section 851(b)(3) (26 U.S.C. 851(b)(3)) (United States Code, current text, 1986)). Two tests, each US law, read on today's weights. Supports: at most 25% in issuers above 5% of assets, the 1940 Act's 75% test. Challenges: more than 50%, which fails even the tax code's 50% test for a regulated investment company. Between the two, Mixed: the fund passes the tax test but not the Act's on today's weights. The Act applies its test when a fund acquires a security, so a fund that drifted past 25% has not broken the law; the reading says how concentrated it is now. The 10% voting-share limit is not measured. ETFIQ reads weights as shares of net assets, while both laws use total assets. The tax test, like the Act's, allows for drift: it is read at each quarter's end, with time to cure a failure (26 U.S.C. 851(d)). So a Mixed or Challenges reading measures concentration on today's weights and is never a finding that a fund has broken either law. Weight: 6%. Sourced to: US Congress, Investment Company Act of 1940, section 5(b)(1) (15 U.S.C. 80a-5(b)(1)) (United States Code, current text, 1940); US Congress, Internal Revenue Code section 851(b)(3) (26 U.S.C. 851(b)(3)) (United States Code, current text, 1986); ETFIQ, Portfolio methodology (ETFIQ, methodology page, 2026). Tests the finding of: Diversification, Low Cost. Research: none cited; the criterion rests on the source alone. The IQ Liquidity Score (ETFIQ’s measure) Can a holder trade the fund without paying away the cost the other seats counted? Measure: the IQ Liquidity Score. The IQ Liquidity Score on its fixed 0 to 100 scale with published anchors. Null until the Trading hub publishes it for the fund. Supports where the IQ Liquidity Score at or above 60. Challenges where the IQ Liquidity Score at or below 30. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. 60 or more, and 30 or less, on the fixed scale are ETFIQ thresholds. Pending: its measure is not published yet, so the criterion is printed as pending and left out of what the seat must read. Weight: 8%. Sourced to: ETFIQ, Trading and liquidity methodology (ETFIQ, methodology page, 2026). Tests the finding of: Low Cost, Macro & Liquidity. Research: none cited; the criterion rests on the source alone. Premium and discount to NAV (ETFIQ’s measure) Does the price stray from what the holdings are worth? Measure: the median premium or discount. The median absolute premium or discount of the closing price to NAV over the window the Trading data states (its last twelve months), percent, as the Trading hub publishes it. Supports where the median premium or discount at or below 0.1%. Challenges where the median premium or discount at or above 0.5%. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. A median of 0.10% or less from NAV, and 0.50% or more, are ETFIQ thresholds. Pending: its measure is not published yet, so the criterion is printed as pending and left out of what the seat must read. Weight: 6%. Sourced to: ETFIQ, Trading and liquidity methodology (ETFIQ, methodology page, 2026). Tests the finding of: Low Cost, Margin of Safety. Research: none cited; the criterion rests on the source alone. Flows and closure risk (ETFIQ’s measure) Is the fund small and shrinking enough that it could close? Measure: the closure risk. high where net assets are under $25 million and the last 3 months saw net outflows; elevated where net assets are under $50 million and it is not high; low where net assets are $50 million or more. Null where net assets are not published, or where they are under $25 million and the 3 months of flows were not read. Supports where the closure risk equal to low. Challenges where the closure risk equal to high. Mixed between. Threshold: ETFIQ’s figure, because the investor published none. Under $25 million with 3 months of net outflows, and $50 million, are ETFIQ thresholds. Weight: 6%. Sourced to: ETFIQ, Fund flows methodology (ETFIQ, methodology page, 2026). Tests the finding of: Macro & Liquidity. Research: none cited; the criterion rests on the source alone.

### Shown beside the seat, not scored

- Figures ETFIQ does not have for this fund (the figures missing: How many of six figures ETFIQ prints for a fund it does not have for this one, a stale figure counting as missing: fee, net assets, 1-year return (for a fund over a year old), holdings book (missing, or dated more than 100 days before the reading), daily flows (missing, or more than 10 days old), and its benchmark (for core and income funds). Reported as ETFIQ's own gap, never as an objection to the fund.) Printed beside the seat, never an objection: a gap is ETFIQ's own coverage, not a fault of the fund. A stale book or stale flows count as gaps.

### How the Skeptic objects

The Skeptic tests the other seats' findings, starting with the strongest supporting seat: of its criteria that Challenge, the minutes lead with the one whose tests list names the seat with the highest score, then the rest in seat order. Each objection names the figure, its date and the methodology page it comes from, for example: Low Cost supports, but full cost with spread is 0.21 points above the benchmark ETF.

### Compare and model modes

In model mode: In each sleeve, only funds the Skeptic reads with no objection.

### How the seat is tested

The cases this seat must pass before it ships, with the result each must give.

- A fund most seats support, whose full cost is 0.21 points above its benchmark ETF : Challenges (Full cost to own against the benchmark ETF: Challenges, The gap to its own benchmark: Supports). Low Cost supports on the fee, but the fee plus the spread costs 0.21 points a year more than the benchmark ETF: a real hole, found with ETFIQ's own figures.

- A clean index fund : No objection. Every check reads clean, so the Skeptic reports no objection rather than inventing one.

- An income fund paying mostly return of capital : Challenges (Return of capital share: Challenges). 62% of what it paid was the holder's own money handed back, and it trailed its benchmark by 4 points.

- A theme fund that is mostly its benchmark ETF : Challenges (Overlap with its benchmark ETF: Challenges). 78% of it is what the benchmark ETF already owns.

- A fund with only two of its checks readable : Not enough data. Two clean checks are not enough to clear a fund; the Skeptic says so rather than reporting no objection.

- An income fund whose payout was covered, and which partly delivers its promise : Mixed. One check sits between its thresholds, and none fails.

- A 30/70 allocation fund, 11 points behind ACWI and level with its own mix : No objection (The gap to its own benchmark: does not apply, The gap to the ETFIQ benchmark at the same stock share: Supports). Measured against the ETFIQ benchmark at its own 30% stock share, not an all-stock index: no objection.

- A Treasury fund paid for its duration : No objection (The rate cushion: Supports). A year's income covers a 1.2-point rise in rates, and nothing else objects.

- Two holes, answered to the seat that supports most strongly : the minutes lead with The largest single company. Diversification supports most strongly (Supports, 0.9), so the minutes lead with the objection that tests it, the 27% in one company, before full cost, which tests Low Cost (Mixed, 0.4).

- A 3x fund whose Promise check is not yet published : Not enough data. Its size reads clean, but the one test that matters for a leveraged fund, the daily multiple compounded, has not been read: the Skeptic does not clear it.

- MSTY : Challenges. An income fund whose year's total return was minus 135.9% of the cash it paid. Figures read on Oct 22, 2026.

- VOO : any result but Challenges. An S&P 500 fund at 0.03% with every figure published. Figures read on Oct 22, 2026.

### What this cannot tell you

- A promise made outside the fund's category, such as a word in its marketing that no test reads.

- The liquidity score, spread and premium, and the Promise badge, until their pages publish them.

- Whether another seat is right: the Skeptic tests a seat's finding against ETFIQ's figures, not against the investor's judgment.

- Whether to own the fund. The seat applies published principles to data and gives no advice.

### Changes to this seat

- Version 2.0, Oct 21, 2026: Allocation funds are measured against the ETFIQ benchmark at their own stock share, never an all-stock index; data gaps are reported as ETFIQ's, not as an objection, and stale figures count as gaps; payout coverage, the rate cushion and the law's two diversification tests joined the seat, each threshold side labelled; the order of objections is set.

- Version 2.1, Oct 21, 2026: The seat reports no objection only after reading the measure that matters for the fund's category; until then it says Not enough data. A return-of-capital share is read only with the dates its notices cover.

### How to check any of this

This seat’s criteria, thresholds, weights and cases are in the committee file , the same file the committee reads.

### Where to next

---

Source: ETFIQ, independent ETF data. Recomputed nightly from public filings and prices.
Cite: ETFIQ, The ETFIQ Skeptic: what an ETF's name leaves out, data as of Oct 21, 2026. https://etfiq.com/investment-committee/etfiq-skeptic/
Licence: https://etfiq.com/license/ Figures ETFIQ computes are free to use with attribution: ETFIQ (etfiq.com). Issuer data stays under its owner's terms.
HTML: https://etfiq.com/investment-committee/etfiq-skeptic/