Disclosures

Last updated: 2026-07-27

This is the page the Terms of Service refers to when it says the Disclosures page is part of your agreement with us. It collects, in one place, the things we think you should know before you rely on anything RingMaster shows you. It is written to be read, not to be survived.

1. What this site is

RingMaster publishes model portfolios and the analysis around them. You are not an advisory client of Ohey Inc, and no fiduciary or advisory relationship exists between us, unless and until you and Ohey Inc have entered into the written Investment Advisory Agreement. See section 5 of the Terms of Service, which governs on this point.

2. We publish no performance figures, anywhere, to anyone

Ohey Inc publishes no returns, no annualised figures, no risk-adjusted statistics, no win rates, no drawdown figures, no hypothetical results, no results reconstructed from historical data, and no illustrative example built from any of those. Not on this site, not in the product, not in an email, not in marketing, and not in a conversation. If you ever see a figure of that kind attributed to us, it did not come from us and you should tell us about it.

This is deliberate, and the reasons are worth stating because "we don't publish numbers" reads as evasive unless you know why.

A short record says almost nothing. A figure computed over a short period is dominated by luck, and publishing one invites exactly the inference it cannot support.
We cannot see your account. No figure we published could describe what any client actually experienced. It could only describe a model, and a model figure gets read as if it were a client's result. It is not.
The gap between a model and a filled order is real, and it is yours. Your entry price will differ from the price a model recorded, because of the spread, commissions, and the movement between your decision and your fill. A model figure quietly assumes that gap away.
A figure reconstructed from history is built with hindsight. It is constructed knowing what happened next, on a version of the method that may not be the version that runs today. See section 5.

So we publish none. The firm's software is built to enforce this: internal certification notes that do contain figures are held separately from anything a customer can see, and the payloads the product serves carry portfolio weights and never a performance number. If we ever do publish a figure, it will come with a full statement of how it was computed, over what period, net of what costs, and with what limitations, and it will appear in an amended Form ADV Part 2A before it appears anywhere else.

Past results do not indicate future ones. That is true of any figure anyone shows you, including ones that are not ours.

3. We hold no custody and exercise no discretion

This list is exact, and we intend to be held to it:

No custody. We never hold, receive, or have access to your money or your securities. We have no authority to withdraw anything from any account of yours and no standing arrangement with any custodian.
No discretion. We cannot make a decision for your account. You make every decision.
No execution. We do not place orders, ever.
No brokerage connection. We do not connect to your broker. We do not ask for and do not want your broker login.
No automated trading and no mirroring. Nothing on this site can trade on your behalf or copy a book into your account.
No account statements from us. Your broker is the only source of truth about what you hold. Reconcile against your broker, never against this website.

You are responsible for every order you place, for its size, its timing, the account you place it in, and the outcome. You are also responsible for your own taxes and for meeting your broker's rules, including margin rules and the pattern day trader rule.

4. What a rules-based mapping can and cannot do

The heart of the product is a mapping: your questionnaire answers go in, and one of a finite set of pre-built model portfolios comes out. It is worth being precise about what that is.

It is software, and there is no human in it

No person at Ohey reviews your assignment, approves it, or overrides it. The rules are applied in a fixed order and the same answers always produce the same book. That is a real advantage: the rule cannot have a bad day, cannot be talked into something, and cannot be nicer to a bigger account. It is also a real limitation: a defect in the rules would apply to everybody it touched, at once, silently, until somebody found it. We publish the rules, we content-hash them so a silent change is detectable, and we test every cell of them exhaustively, because an automated advisory feature run by a very small firm is a known way to hurt people. Controls reduce that risk. They do not remove it.

It knows nothing you did not tell it

The mapping has your answers and nothing else. It does not know what else you own, what you owe, what your tax position is, whether you are about to lose your job, whether this money is your emergency fund, whether you are running the same names in another account, or whether the answer you gave about sitting through a large fall is the answer you would actually live up to when it happened. If your answers are wrong, optimistic, or out of date, the book you are shown may be a poor fit for you and nothing in the system will notice. Update your answers when your circumstances change. You can do it at any time, and you do not need to give a reason.

What it does with a missing answer

It assumes the more restrictive thing, every time. An unstated risk tolerance reads as the most cautious. An unstated maximum loss reads as the smallest. An unstated account capability reads as cash-only. An unstated account size reads as the smallest band. This makes a half-finished questionnaire produce a cautious result rather than an unsafe one, but it also means a book shown against incomplete answers is not really a book chosen for you.

It can only ever move you to something more cautious

Every screen in the ladder removes options; none adds one. Your maximum loss answer is treated as a ceiling. A short time horizon removes books whose holding pattern does not fit inside it. A cash-only account removes every book that has to borrow shares. An account below a book's minimum size removes that book. A book that has not passed our internal certification review is removed for everybody. What you cannot do, by any combination of answers, is talk the rule into a riskier book than your starting point.

It is a selection, not a portfolio built for you

The books themselves are not individualised. Each book holds the same names at the same weights for every client who is shown it. Your answers decide which book you see, not what is in it. Nobody at Ohey builds or adjusts a portfolio for any individual.

It is not a plan

It is not financial planning, tax advice, legal advice, retirement planning, or advice about anything you hold elsewhere. It does not consider your whole situation, because it does not know your whole situation.

5. Data and model limitations

The ranking is a rank statistic, not a forecast

The model orders a broad universe of US-listed equities relative to one another for the session ahead. What it produces is a position in a queue. It is not a forecast of a price, not a forecast of a return, and not a probability that any particular thing will happen. The product deliberately does not display an expected move, a confidence score, or a model agreement figure, because those outputs are not meaningful and showing them would imply a precision that is not there. If you find yourself reading a rank as a prediction, that is a reading the number cannot carry.

Missing inputs are filled in silently

The model expects a large set of inputs for every name it scores. When an input is not available at the time of scoring, the pipeline does not fail and does not warn: it substitutes the median value that input took during training, and the substituted column arrives identical for every name in the universe.

Two consequences follow, and they compound:

A column that is the same for every name cannot tell one name from another, so it contributes nothing to the ordering. The ranking is therefore driven by a smaller set of genuinely varying inputs than the model was designed around.
A substituted median is usually not the value the input would actually have had. So the model is not merely working with less information: it is running a slightly different ranking function from the one that was studied, and applying it with full confidence.

There is no observable difference between a healthy model ranking on its full input set and a hollow one ranking on a fraction of it. Both produce a confident-looking book. We are telling you this because it is material, it is the kind of thing that never appears in a brochure, and you cannot see it from the outside.

The books are filtered, but not by everything

Before a name may enter a book it passes a universe quality filter that removes instruments that are not ordinary company shares, including funds and notes a book could not reliably hold or borrow. That filter is necessary but not sufficient: two further screens used by our own execution venues, a minimum share price and an exclusion list for names that are hard or dangerous to borrow, depend on data that is not part of what the website receives, so the website cannot reproduce them. For that reason a book reconstructed on the website is never described as a book that was actually placed.

A risk overlay applies to some books, not all

Each evening a risk state is computed for certain books and published beside them: at full size, at a reduced size, or standing down for the session, and a book that is not at full size must say why. The rules behind it differ by book. Do not read an overlay on one book as protection on another. When the file that carries the risk state is missing or does not parse, the site says the risk state is unavailable rather than assuming everything is normal. Unavailable means unknown; it does not mean safe.

A book we have not certified is not shown to anyone

A book is served only after it has passed our internal certification review. A book that has not passed is hidden from everybody, and a client whose answers point at it is moved on to the nearest book that has passed, with the substitution stated. One of the defined books is failing that review today, because the one remaining check requires real trading over a stretch of live sessions and no amount of historical testing can substitute for it. We do not say when, or whether, that changes.

6. Market data may be delayed or stale, and here is how we judge that

Market data shown on this site may be delayed, incomplete, or wrong. We take data from sources we do not control and we do not warrant any of it.

Beyond ordinary data quality there is a specific risk: the site serves what the nightly pipeline last published, and if that pipeline does not run, or its output does not reach the server, the desk can be showing an out-of-date session while the market has moved on. That has happened, and it is why the product measures staleness the way it does.

Staleness is judged on sessions missed, not on hours elapsed. Wall-clock age cannot tell a weekend from a fault: the generator runs on weekday evenings, so a Friday session is legitimately more than two days old by Monday morning, and any hours-based threshold loose enough to tolerate that is too loose to catch a genuinely missed session. So the product computes which session it ought to be holding, from the generator's known schedule, and compares it to the session it does hold. That comparison is exact. There is no warning band, and no amber state, because a warning that fires every weekend is not a warning.

Two consequences you should know about. The true age of the data is always displayed, because that is a fact and we do not round facts; only the judgement is calendar-aware. And a market holiday reads as "behind" on purpose, because the product has no trading calendar and will not invent one. Erring toward "I may be stale" is the safe direction.

7. Conflicts of interest

We intend to make money in one way: a subscription fee paid by you for access to the website. There is one paid plan on two billing schedules, $19 a month or $190 a year, and no other fee of any kind. Twelve months bought one at a time would be $228, so paying yearly saves you $38, and it unlocks nothing the monthly plan does not. Annual is billed a year in advance; if you cancel part way through we refund the whole months you have not used, at one twelfth of the annual price each, and we keep no cancellation fee. There is also a free plan, which is permanent rather than a trial: it carries everything we publish, including everything our models produce, one trading session behind, and the subscription carries the same material for the current session. We take no commission, no payment for order flow, no referral fee, no soft dollars, and no money from any third party in respect of you. We do not sell your data, and there is no advertising network, data broker, or third-party analytics service on this site.

That single revenue source still creates conflicts. These are the real ones:

The conflict	What we do about it
We decide which book your answers select, and we are paid by you. That gives us a reason to show you something that looks worth paying for.	There is one paid plan, so there is no dearer level to route you to and nothing we can withhold to sell you an upgrade: every book you are suited to comes with the same subscription. The rules also contain an explicit step that tells a small account the subscription may not be worth its cost, before you subscribe, and points you back at Free. We publish the reasoning so you can check what the rule actually did. That is a control, not a cure. Read the reasoning.
People at our firm may own the same securities that appear in a book. Since we place no orders for anybody, a person here who traded a name before it was published would be trading ahead of the clients who receive it.	Disclosed here and in Form ADV Part 2A Item 11. The firm's personal trading policy is a matter for that document.
Different clients see the same book at different times. A free account sees every book it is eligible for, complete and sized, one trading session behind the subscriber view.	We say so plainly rather than presenting a delay as a feature, and we label the session on every screen that carries it so a free client always knows which day they are looking at. We will not claim more separation than the delay actually provides: a book is held for several sessions at a time, so a one-session-old view will usually name the same positions a subscriber is holding. What a subscription buys is the current session and nothing else: not sizing, not extra books, not extra panes, and not privacy over the names, all of which the free plan now carries in full. The conflict this creates is a crowding one: because the two views name the same securities, a subscriber acting at the open may be trading alongside free readers acting a day later in the same names.
We run the same books ourselves, at less than full capital, as our own forward record.	That record is ours and is not a client account. It produces no figure that we publish.
Every client entitled to a book sees it at the same moment, so clients may compete with one another for the same liquidity.	We have no way to sequence, stagger, or allocate among clients, because we place no orders. We disclose it rather than implying it is managed.

Several conflicts an adviser normally has cannot arise here at all: we cannot trade against you, cannot front-run your order, and cannot cross a trade, because we place no orders, hold no assets, and see no account.

8. Risk of loss

Investing puts your money at risk. You can lose money, including all of the money you put in. Do not invest money you cannot afford to lose. If you are unsure, speak to a professional who knows your full situation.

Specific risks that attach to these books, briefly:

Model risk. A model that ordered names usefully in the past may order them uselessly, or harmfully, in the future, and can be confidently wrong for long stretches.
Concentration. The books hold a small number of names, so a single adverse event in a single company can move the whole book.
Short selling. Several books hold short positions. A short position has no ceiling on its loss: the share you are short can rise without limit and your loss rises with it. Borrow can become expensive or be withdrawn overnight, a lender can recall stock and force a position closed against you, and a crowded short can move violently for reasons unrelated to the company.
Margin. Books that hold shorts require a margin account. Margin magnifies both directions, carries interest, and exposes you to a margin call in which your broker may liquidate at a time and price you did not choose.
Pattern day trader restrictions. United States rules restrict frequent same-day trading in margin accounts below a stated equity minimum.
Execution and slippage. Your fills will differ from the prices a model recorded, sometimes materially, particularly in less liquid names, at the open, and on volatile days. Over a book with high turnover that difference can be the difference between a result and its opposite.
Taxes. Turnover has a tax cost in a taxable account, and neither the questionnaire nor the mapping takes your tax position into account.
Availability. The service can be interrupted. If the site is down on an evening when a book changes, you will not see the change.

9. Third-party content and links

The site may show market data or link to other websites. We do not control those sources and we are not responsible for them. Treat a link as a pointer, not as our endorsement.

10. Where these documents sit relative to each other

Document	What it is for
Terms of Service	The contract for using the website. It says when an advisory relationship does and does not exist.
Privacy Policy	What we collect, why, who sees it, how long we keep it, and what you can ask us to do with it.
This page	The disclosures the Terms of Service incorporates by reference.
Form ADV Part 2A	The firm brochure: the long-form description of the business, the fees, the methods and the risks.
Form CRS	The short relationship summary, with the questions you should ask us.
Investment Advisory Agreement	The advisory contract itself, and the conditions on which it takes effect.

Where this page conflicts with the Investment Advisory Agreement on the subject of the advisory relationship, that Agreement governs.

11. How to contact us, and how to reach a regulator

Write to [email protected]. That address reaches us for questions, complaints, account closure, and privacy requests. We aim to reply within five business days.

If we cannot resolve a complaint to your satisfaction, you are entitled to take it elsewhere, and nothing in any agreement with us takes that right away:

The SEC. Complaints and tips can be submitted through the Commission's website at sec.gov/tcr. Free tools for researching firms and financial professionals are at Investor.gov/CRS, which also carries educational material about investing.
Your state securities regulator. Every state has one, and it will take a complaint about a firm operating in your state.
Your state attorney general, for a privacy complaint.

One last thing, because it matters: investing puts your money at risk and you can lose money, including all of the money you put in.