Every funded-trading programme publishes a rulebook, and most traders read it the same way: scanning for the one number that could end the account. That reflex makes sense. It is also why accounts die to rules their owners had already read.
The problem is not the reading. It is the framing. A rulebook looks like a list of hurdles, so it gets treated like one — memorise the thresholds, avoid them, move on. But the thresholds are the output of the document, not its logic. Underneath, a rulebook is a risk contract, and every clause answers a question the firm has to answer before it puts capital behind a stranger on the internet.
There are four of those questions. Learn them, and any rulebook becomes readable in a single sitting.
Every rule answers a question about risk
The four questions are simple to state.
- Can this trader make money at all?
- How much can this trader lose before we stop?
- Is the money repeatable, or was it one lucky afternoon?
- What are we willing to let this trader do to get it?
Profit targets answer the first. Loss limits answer the second. Consistency rules answer the repeatability question. Everything else — time limits, style restrictions, news windows, weekend policy — answers the fourth.
Four families. That is the whole taxonomy. What follows is what each one actually measures, where the common misreadings live, and how the families collide in the moment that matters.
Family one: the profit target
A profit target is the entry ticket. It is the firm asking whether you can produce a return at all, and it is the least interesting rule in any rulebook — because it is the only one you fail by standing still, and the only one you can approach at whatever speed you like.
What matters is what the target is measured against. A target expressed as a percentage of the starting balance is a moving goalpost that does not move: the number is set on day one and stays there, whatever happens to your equity in between. That is the version worth wanting, because it cannot drift away from you after a drawdown.
At Carrot the 2-Phase Challenge splits the target across two stages — Reach the Phase 1 (Evaluation) profit target (5%) to advance to Phase 2 (Verification). — where Phase 2 asks for 8%. The 1-Phase Challenge asks for the same 8%, once: Reach the 8% profit target while staying within both equity limits.
Why targets get split into phases
Two phases are not two hurdles. They are one hurdle, sampled twice.
A single evaluation can be passed by a trader who took one enormous swing and got lucky. A second phase asks the same question again, after the luck has had time to run out. That is the entire argument for the two-phase structure, and Carrot's own numbers show the trade in miniature: 5% and then 8% across two stages, against 8% once. The staged route asks for less at a time, and asks twice.
The one-phase structure makes the opposite trade. Fewer stages, faster route, and the firm compensates for the lost sample somewhere else in the document. Which brings us to the families that do the compensating.
Family two: the loss limits
This is where accounts actually end. Everything in this family is a variation on one question — how much are we willing to lose before we stop — and the variations matter enormously.
Daily loss and overall loss are different instruments
A daily loss limit is a circuit breaker. It exists to stop a bad session from becoming a bad week, and it resets on a clock. An overall loss limit is the floor of the account. It does not reset. Touch it, and the evaluation is over.
Traders conflate them constantly, usually in the direction that hurts: treating the daily limit as if it were the floor, and discovering too late that a series of days well inside the daily limit has walked the account down to the overall one.
The reset clock is the part worth reading twice. At Carrot the daily limits sit at twoPhase: '5% of equity', onePhase: '4% of equity', recalculated each day at 00:00:00 UTC. On a venue that runs around the clock, that boundary is not the close of a session — it is a moment on a clock that arrives while the market is still moving. If your local trading day straddles it, you are trading against two different daily budgets in one sitting.
Equity or balance: the distinction that decides your day
A loss limit measured on balance counts closed trades. A loss limit measured on equity counts open ones too.
The difference is not academic. Under an equity-based limit, an unrealised drawdown on a position you fully intend to hold is a real drawdown, right now, against a real threshold. There is no waiting it out. Carrot states the arithmetic plainly — Equity = Balance + Unrealized P&L. — and both daily limits above are written against equity, not balance.
If you take one thing from this section, take this. Check which of the two your limits are measured on before you size a position, not after.
Anchored or trailing
The last variable in this family is whether the overall loss limit stays where it started.
An anchored limit is set once against the starting balance and stays there. A trailing limit follows your equity upward as you profit, which means the floor rises underneath a winning account — protecting gains you have made, and tightening the room you have to give them back. Carrot's overall limit trails upward from your high-water mark and stops once it reaches the starting balance — so it can rise while you are in early profit, but it never climbs past the level you opened at. For the 1-Phase Challenge the limit is stated as onePhase: '8% of starting balance'; the figures for both routes are laid out side by side in the CarrotFunding Rulebook.
Anchored or trailing changes how you should behave after a good week. Under an anchored limit, early profit is a genuine buffer. Under a trailing one, it is not — it is a new floor.
The repeatability family: consistency rules
The third question is the one traders resent most, and it is the most reasonable question in the document: was the money repeatable?
A firm about to allocate capital is not buying your best day. It is buying your distribution. A trader who made the target across many ordinary sessions and a trader who made it in one spectacular one have produced the same number and told the firm two completely different things about what happens next.
What a best-day rule actually measures
Consistency rules formalise that intuition as arithmetic. The mechanic is a ratio: take your total profit, remove your single best day, and ask how much of the result survives.
Carrot's version is the Best Day Rule. The rulebook publishes the calculation itself — (Total Profits - Best Day Profit) / Total Profits x 100 — and the threshold that score has to clear is set out there alongside it. Two things about the rule are worth knowing before you plan around it.
The first is scope: the rule Applies to 1-Phase Challenge only. Not applicable to 2-Phase Challenge. — which follows directly from the logic above. A two-phase structure already samples you twice; it buys its repeatability evidence through the second phase instead of through a ratio.
The second is that a consistency rule is a gate, not a trapdoor. Falling below the threshold does not end a challenge. The rulebook is explicit that a score under the threshold is not a failure state — the challenge continues. It holds progression until the distribution catches up, and the way it catches up is by trading normally, because every ordinary day you add makes your best day a smaller share of the total.
That reframing matters. A trader who reads a consistency rule as a threat trades scared. A trader who reads it as a ratio just keeps trading.
Family four: the freedom rules
The fourth family is defined by what a rulebook chooses not to restrict. Time limits, minimum trading days, holding-period rules, style prohibitions, news blackouts, weekend flat requirements — each one is a lever a firm can pull to reduce its own variance, and each one is paid for by the trader in flexibility.
These rules are easy to skim past, because an absence reads as nothing. It is not nothing. A minimum-days rule forces activity that a patient strategy would not otherwise take. A time limit converts a risk decision into a deadline decision. A news blackout removes exactly the sessions some strategies exist to trade.
The rules Carrot does not have
The rulebook's position here is a single line: No time restrictions, no trading style limitations, no artificial barriers. No countdown. No minimum days. No rule on how long a position may be held.
News trading is not fenced off either — the venue documentation states it directly: We don't restrict trading around news events - you can trade through them. And because Hyperliquid trades 24/7, there is no weekend flat rule to observe, because there is no weekend close to observe it against. We wrote about what that does to the rhythm of an evaluation in trading a challenge in a market that never closes.
Leverage is the one dial that stays bounded: Leverage is up to 5x per position, or the maximum Hyperliquid allows for that market if that is lower. That is a ceiling, not an instruction — and the number of challenges you may run in parallel has no ceiling at all: There is no limit to how many challenge accounts you can hold or pass.
Where the families collide
Read individually, the four families are straightforward. The failures happen where they overlap, and two overlaps account for most of them.
Fees are part of your P&L
Trading costs are not a separate ledger that settles later. They land in the same P&L the loss limits are measured against — at Carrot, You pay standard Hyperliquid maker/taker fees with no markup, and those fees move your equity like any other debit.
For a high-frequency approach running close to a daily limit, that is not a rounding difference. Your costs and your risk budget draw on the same account, and a strategy sized as if they were separate is quietly overleveraged all day.
A breach is an event, not a negotiation
The last collision is procedural. A limit breach is evaluated by the system, at the moment of the touch, on equity that includes your open positions. It is not reviewed afterwards by someone weighing context.
Carrot's rulebook says so without softening it — No appeals, no resets, no exceptions. — and pairs it with the other half of the same sentence, which is the part traders should actually price in: Your only loss is the challenge fee. A breach is final, and it is bounded. Both halves are true, and a rulebook that states the first without the second is telling you half of the contract.
How to read a rulebook quickly
The taxonomy collapses into six questions. Ask them of any rulebook, in this order.
- Is the profit target measured on starting balance or on something that moves?
- Is the daily limit measured on equity or on balance — and when exactly does it reset?
- Is the overall limit anchored to the start, or does it trail your high-water mark?
- Is there a consistency rule, and does it gate progression or end the account?
- Which freedoms are restricted — time, style, holding period, news, weekends?
- What is the maximum I can lose, in currency, if every rule goes against me at once?
The last question is the one that puts the other five in proportion. Answer it first if you only have time for one.
Carrot publishes all of this in a single place rather than across a marketing page and a support article: the full parameter set for both routes is in the CarrotFunding Rulebook, the route from evaluation to a funded account is documented in how the challenge process works, and the phase-by-phase specifics live in the evaluation and verification account documentation.
Read the rulebook as a contract, not a warning label. The rules are not there to catch you out — they are the firm telling you, in advance and in writing, exactly what it is buying.
