Canonical page: https://sequoia-research.com/writing/delaware-statutory-trust-reviews-performance-denominator-checker Author: Sequoia Research Published: 2026-10-17T01:36:44.432Z Updated: 2026-10-17T01:36:44.432Z Interactive tool: https://sequoia-research.com/writing/delaware-statutory-trust-reviews-performance-denominator-checker#research-tool # Delaware Statutory Trust Reviews: Performance Denominator Checker ## What Is This Article About? This guide explains a tool for checking the program counts behind a reported performance percentage. It reconciles the stated population, exclusions, included cases, and cases meeting a defined outcome. It also flags differences between the groups you choose to compare. A fictional example shows why a percentage can change when the denominator changes, even when the reported outcome count stays the same. ## Table of Contents - [What the Performance Denominator Checker Does](#what-the-performance-denominator-checker-does) - [Define the Claim Before Entering Counts](#define-the-claim-before-entering-counts) - [Enter the Population, Exclusions, and Included Cases](#enter-the-population-exclusions-and-included-cases) - [Keep Unknown Counts Separate From Zero](#keep-unknown-counts-separate-from-zero) - [State the Outcome Being Counted](#state-the-outcome-being-counted) - [Read the Population Reconciliation](#read-the-population-reconciliation) - [Check the Reported Percentage at Its Displayed Precision](#check-the-reported-percentage-at-its-displayed-precision) - [Work Through the Fictional Program Tables](#work-through-the-fictional-program-tables) - [Compare Denominators Without Ranking Sponsors](#compare-denominators-without-ranking-sponsors) - [Record Source Dates and Exclusion Rules](#record-source-dates-and-exclusion-rules) - [Understand Count Limits and Rounding](#understand-count-limits-and-rounding) - [Save the Check and Use the Open Questions](#save-the-check-and-use-the-open-questions) - [Frequently Asked Questions](#frequently-asked-questions) - [About Sequoia Research](#about-sequoia-research) - [Sources](#sources) ## What the Performance Denominator Checker Does The checker tests arithmetic and selected fields in a claim record you enter. It does not gather sponsor history, review program lists, or decide whether a return presentation is fair. Use it after identifying the counts and definitions in the source you are reading. The key question is simple: which cases sit below the line in the percentage? Nine programs meeting an outcome out of 12 included programs gives a different figure from nine out of 15. The numerator is nine in both cases. The denominator is the group being used as the base. The SEC warns that selected performance can exclude unfavorable periods or highlight only profitable investments. That is a reason to understand the population behind a claim. It is not evidence that a particular sponsor has omitted unfavorable results. [SEC performance-claims bulletin, September 15, 2022](https://www.investor.gov/introduction-investing/general-resources/news-alerts/alerts-bulletins/investor-bulletins-47). This tool produces a count record and questions, not a score. A row with no flagged fields is not an approved performance claim. The arithmetic can balance while the underlying list remains incomplete or uses a definition that does not answer your question. ## Define the Claim Before Entering Counts Give each claim a name that identifies its source or purpose. “Initial completed-program table” is more useful than “good sponsor.” You can enter up to 12 claims. Use separate rows when two documents use different populations or periods. Describe the included cases in words. State the program type, exit status, relevant period, and other rules that determine inclusion. A table of completed cash exits is different from a list of all launched programs. A table limited to one property type is different from a firm's entire history. Do not enter return percentages as program counts. A reported 8% annual return is not eight successful cases out of 100. This checker is designed for counts and the percentages formed from those counts. It does not calculate internal rates of return, equity multiples, or returns weighted by invested capital. Use consistent units within a claim. If the population counts programs, the exclusions and included cases must also count programs. Do not mix properties, offerings, investor accounts, and transactions without resolving the definitions first. The software cannot identify a unit mismatch from numbers alone. ## Enter the Population, Exclusions, and Included Cases The stated total population is the full group from which the source says it selected cases. Exclusions are cases left out under a stated rule. Included cases are the cases used for the reported outcome percentage. For reconciliation, the tool assumes the included and excluded groups do not overlap. Each case should appear in one group at most. It cannot inspect a roster to check that assumption. If one program appears twice, a balanced total may still hide the duplicate. Enter a source's stated counts without silently repairing them. If the numbers do not add up, the resulting flag helps you ask for an explanation. The tool accepts a population mismatch so it can show the size of that mismatch. It does not rewrite the supplied count. Exclusions should have a clear rule. “Outside the stated property group” is a scope description. “Not shown” leaves the reason open. If the excluded count is positive and no rule is entered, the tool flags the missing explanation. It does not classify the excluded cases as losses. A count of included programs is required because it is the base for the claim being examined. The other count fields may be left blank when they are unknown. ## Keep Unknown Counts Separate From Zero Blank means unknown in the population, exclusion, and outcome fields. Zero means a known count of none. Those are different statements and produce different results. If a source lists eight included programs but says nothing about the total population, enter eight as included and leave total blank. Do not infer that the full population is eight. The tool can still calculate an outcome rate within those eight if you supply a valid outcome count. It cannot reconcile the wider population. The same rule applies to exclusions. A source that does not mention exclusions has not necessarily established that there were none. Leave the field blank unless the evidence supports a zero. Add the uncertainty to the source reference so a later reader can see why the field was left open. Unknown outcomes also stay unknown. A blank outcome count does not produce a zero-percent result. If there are no included cases, even an entered outcome count of zero produces no rate. Dividing by zero does not yield a useful performance percentage. ## State the Outcome Being Counted Write the outcome definition before entering its count. It might describe programs that returned at least original capital, programs with a stated positive return, or another defined result. The cases meeting that outcome must be a subset of the included cases. Those definitions are not interchangeable. Returning original capital alone is different from earning a positive total return after all relevant costs. A result measured before personal taxes is different from an investor's after-tax result. Keep the source's wording and limits visible. The checker does not decide whether the outcome is the best measure of performance. It checks whether the count can fit inside the denominator and calculates the stated fraction. The rest of the included cases are not automatically failures or losses. They may have another result or an unresolved status. If the outcome count exceeds the included count, the tool flags the error and withholds the rate. It preserves the entered counts for review. Ask whether the source uses a different unit, duplicates cases, or has a transcription mistake before changing the numbers yourself. ## Read the Population Reconciliation When all three counts are known, the tool calculates: **Unaccounted cases = stated population − excluded cases − included cases.** A positive result means some of the stated population is not accounted for by the two entered groups. A negative result means the groups together exceed the stated population. A zero result means the counts balance under the assumption that the groups are separate. For example, a population of 20, five exclusions, and 12 included cases leaves three unaccounted cases. A population of 20 with five exclusions and 15 included cases balances. These calculations say nothing about what happened to any individual program. The checker separately flags an included or excluded count that exceeds the stated population. A valid outcome fraction may still be displayed when the broader population does not reconcile. Read the fraction alongside the warning. Correct division within a table does not resolve an inconsistent population record. If either the total or the excluded count is unknown, the difference is not calculated. The tool does not replace a missing count with zero or reverse engineer it without telling you. ## Check the Reported Percentage at Its Displayed Precision The calculated outcome rate is the outcome count divided by included cases, multiplied by 100. The result is displayed to two decimal places. You can also enter the percentage printed in a source and select whether that source displays zero, one, or two decimal places. The checker compares like precision. Two outcomes out of three cases calculate to 66.67% at two decimals. A source displaying one decimal would show 66.7% under the tool's rounding convention. Those figures agree at the stated precision. A source displaying whole percentages would show 67%. The tool rounds half up. For example, one case out of eight is 12.5%, which rounds to 13% when zero decimals are selected. A source that uses another rule could differ without an underlying count error. Check its method before treating a rounding flag as a false claim. The entered reported percentage must fit the selected precision. For a one-decimal display, enter a value such as 66.7, not 66.67. If no percentage was reported, leave that field blank. The tool will still show the fraction it can calculate from the counts. ## Work Through the Fictional Program Tables The starting example contains three invented claim rows. All counts, definitions, dates, and source names are training entries. They do not describe an actual sponsor or offering. The first table states a population of 20 programs, five exclusions, and 12 included completed cash exits. Nine included cases meet the stated outcome: returning at least original capital before personal tax. Nine divided by 12 gives 75%, matching the entered whole-number percentage. The population check still leaves three cases unaccounted for. The second table uses the same stated population and exclusion count but includes 15 cases. The outcome count remains nine. Nine divided by 15 gives 60%, and the population now reconciles. The tool flags the different included counts and the unresolved count gap in the first table. | Entered claim | Population less exclusions less included | Outcome fraction | |---|---:|---:| | Initial table | 20 − 5 − 12 = 3 | 9 ÷ 12 = 75% | | Expanded table | 20 − 5 − 15 = 0 | 9 ÷ 15 = 60% | | Separate partial table | Not calculated; population and exclusions unknown | 6 ÷ 8 = 75% | The change from 75% to 60% does not establish that performance worsened over time. Both example rows use the same cutoff date. The denominator changed. The reason for adding three cases and the outcomes of those cases still need evidence. The third table shows six outcomes among eight included cases. Its 75% fraction is valid under those entries, but the full population and exclusions are unknown. It is left outside the comparison group because it describes a separate partial record. Sharing a percentage does not make it the same sample as the first table. Use the output to ask for the program lists, the inclusion rule, and a reconciliation of the three added cases. Do not use the lower or higher percentage as a sponsor ranking. ## Compare Denominators Without Ranking Sponsors Give rows the same comparison-group label when you intend to compare them. Labels match after ignoring letter case and extra spaces. A row without a group remains a standalone claim check. Within a group, the tool flags different included counts, included-case definitions, outcome definitions, known cutoff dates, population counts, and exclusion rules. It also notes missing population or exclusion counts, absent cutoff dates, unreconciled counts, and invalid outcome subsets. The text checks are literal after basic space and case normalization. They do not interpret meaning. Two differently worded definitions may mean the same thing. Identical wording may hide different underlying lists. Read the source and explain the difference instead of relying only on the flag. Equal counts also do not prove equal cases. Two tables can each include 12 programs while sharing only some of them. This version does not import case lists or match program identities. It cannot establish whether a fair comparison is possible; it shows selected differences that deserve review. ## Record Source Dates and Exclusion Rules Use the cutoff date for the record being described, not automatically the day you opened it. Put the relevant reporting period in the included-case definition where needed. A single date cannot describe every timing rule in a performance table. The tool checks that an entered date is a valid calendar date. It compares cutoff dates within a group. It does not decide whether a source is stale, check today's date, or verify that the source actually uses the entered cutoff. Use the source-reference field for a document title, version, page, and important limit. It is plain text. A web address entered there is not opened or checked. No private document needs to be pasted into the tool. FINRA's private-placement guidance calls for careful assessment of third-party research, including its independence and incentives. A count reconciliation is one narrow aid to reading such material. It does not replace investigation of the actual claims or the underlying offering. [FINRA Regulatory Notice 23-08, May 9, 2023](https://www.finra.org/rules-guidance/notices/23-08). ## Understand Count Limits and Rounding Each count must be a whole number from zero through one billion. Enter digits without commas, signs, fractions, or exponent notation. Use blank only in the fields that allow unknown values. The included count cannot be blank. Reported percentages must be from zero through 100, with no more than two decimals and no percent sign. Calendar dates must fall from 1900 through 9999. Text fields have length limits to keep the record focused. Counts are kept as exact integers. Percentage rounding uses integer arithmetic so familiar half-point cases are handled consistently. A rounded 100% can still represent a fraction slightly below 100%. Read the actual numerator and denominator, especially with large case counts, rather than treating the rounded display as a claim that every case met the outcome. ## Save the Check and Use the Open Questions Run the check after completing the entries. Any edit, addition, or confirmed removal clears prior results and disables download until you run it again. Removing a row uses an in-page confirmation. Cancel keeps the claim; confirming removes only that row and renumbers the remaining claims. The JSON download records the entered definitions and counts, calculated rates, flags, comparison groups, method, and generation time. That time is when the output was created, not when the evidence was verified. The file does not preserve an automatic history of earlier runs. On Writing, use the tool's Download control and then the Save file button below the frame when shown. Read the saved file to confirm it contains the intended record. This version has no import function, account storage, or automatic backup. Closing or refreshing the page clears your edits. FINRA's 2026 oversight report identifies failures to document investigation of performance representations. The practical next step is to preserve the question and the supporting answer, not just a clean arithmetic result. The report was published in December 2025. [FINRA oversight report, printed pages 53–55](https://www.finra.org/sites/default/files/2025-12/2026-annual-regulatory-oversight-report.pdf). ## Frequently Asked Questions ### Does a Zero Reconciliation Difference Prove the Record Is Complete? No. It shows that the entered included and excluded counts add to the stated population. The tool has not checked the program list, duplicates, definitions, or missing cases. A balanced calculation is a useful check, not proof of complete disclosure. ### Can I Use This to Compare Sponsor Success Rates? Only after defining what each rate counts and assessing whether the cases are comparable. The tool flags selected differences but does not resolve them or rank sponsors. Different property types, periods, exit rules, fees, and missing outcomes can limit a comparison even when the arithmetic is correct. ### Why Does the Tool Allow Inconsistent Population Counts? It preserves the source's stated counts so the mismatch can be measured and questioned. It flags an overcount or unaccounted cases instead of silently changing an input. An outcome count above the included count is different: it cannot form a valid subset rate, so that rate is withheld. ### Are Cases Outside the Reported Outcome Count Losses? Not necessarily. They may have a different result, be unresolved, or fall outside the outcome definition. This tool does not classify them. Request the relevant status and result records before calling any case a success, loss, or failure. ### Why Can a Rounded Percentage Match but Still Need Review? Rounding can hide small differences, and a matching percentage says nothing about source quality. The tool compares at the selected displayed precision. Read the exact counts and ask whether the chosen population and outcome actually support the claim you intend to use. ### Does the Checker Calculate Investment Returns? No. It calculates percentages of cases meeting a stated outcome. It does not calculate cash-weighted returns, IRR, equity multiples, or investor tax results. Use it to examine the population behind a claim, while reviewing the actual return method separately. ## About Sequoia Research Sequoia Research, LLC provides independent research on DSTs and other real estate investment offerings. This educational tool helps readers make performance populations, missing counts, and comparison limits visible. ## Sources - [SEC: Investor Bulletin on Performance Claims](https://www.investor.gov/introduction-investing/general-resources/news-alerts/alerts-bulletins/investor-bulletins-47). September 15, 2022; accessed October 16, 2026. - [FINRA: Regulatory Notice 23-08](https://www.finra.org/rules-guidance/notices/23-08). May 9, 2023; accessed October 16, 2026. - [FINRA: 2026 Annual Regulatory Oversight Report](https://www.finra.org/sites/default/files/2025-12/2026-annual-regulatory-oversight-report.pdf). Published December 2025; printed pages 53–55; accessed October 16, 2026.