Prop Firm Resets Explained: When to Reset, When to Ride, When to Walk

Your evaluation sits $1,800 into a $2,500 drawdown and the target is still $2,600 away. The reset button costs $85 and wipes the damage clean. Most traders treat that button as an admission of failure, so they grind the crippled account instead, and the math says the grind usually costs more. One note before any of it: the $85 is illustrative. Reset fees vary by firm and account size, so check yours before acting on any number in this article.

What a reset actually is

A reset is a paid restart of the same evaluation. Your starting balance, your drawdown limit (including the level a trailing drawdown has ratcheted up to), your progress toward the profit target, and usually your trading-day counters all go back to day one. Topstep's help center, for example, lists account balance, Maximum Loss Limit, consistency target, and trading days as everything a reset sends back to the start. The exact list varies by firm, so read your own firm's version, but the shape is consistent: account state resets completely, your strategy does not.

Resets are also near-instant. Topstep applies them immediately, and MyFundedFutures advertises accounts restored in minutes. You keep your existing login and platform configuration, and at some firms the reset touches billing: Topstep pushes the rebill date out 30 days when you reset, for example. Billing mechanics change often, so verify the current rule with your firm before you lean on it.

Three things a reset is not: a funded-account rescue, a strategy fix, and a free action. Each of those misreadings costs real money, and the rest of this article prices them one by one.

Reset vs a new evaluation

The reflex assumption is that the reset is always the cheap option. It usually beats full list price for a new evaluation, and it is instant with the same account settings. But futures prop firms run aggressive promotions on new evaluations almost continuously, and during a promo a discounted fresh eval can undercut the reset fee. That leaves exactly one correct behavior: compare both prices on the day you decide, every time.

Here is the reset landscape as of mid-2026. Every figure below is time-sensitive and firm-specific. Treat the table as a snapshot and verify against the firm's own pricing page before paying anything.

Two corrections to the folklore. First, "resets cost tens of dollars" is only true on small accounts; large evaluations reset for low hundreds, and at some firms a reset now costs more than a discounted new evaluation. Second, the fine print diverges: a reset may preserve or push your billing date and can consume a free credit, while a new evaluation restarts everything, including minimum-trading-day counters, and may re-trigger setup steps. Neither option is universally better. The two prices and the fine print decide it, that day.

Apex traders face a different question entirely. With no reset on 4.0 at the time of writing, a failed evaluation means a fresh purchase, which changes risk management from the first trade rather than after the first bad week. Our guide to passing the Apex evaluation covers the one-shot model in detail.

The buffer math: what you are really buying

Back to the opening scenario: $2,500 maximum drawdown, $1,800 already gone, so the remaining buffer is 2,500 minus 1,800, or $700, with the target still $2,600 away and an illustrative $85 reset fee.

Look at what the reset actually purchases. It restores the full $2,500 of room, which means it buys back $1,800 of survivable drawdown for $85. That is 85 divided by 1,800, about 4.7 cents per dollar of room. No other purchase in trading hands you risk capacity at that price.

Now model the grind honestly. Say you risk $200 per trade (1R), win $300 when right (1.5R), and win half the time. Expectancy is 0.5 x 300 minus 0.5 x 200, or +$50 per trade, so reaching a $2,600 target takes about 52 trades on average. Meanwhile the remaining buffer is 700 / 200 = 3.5R. One run of four losses (4 x $200 = $800) ends the account.

Four straight losses at a 50% win rate has a 6.25% probability in any given four trades (0.5 to the fourth power). That sounds tolerable until you stretch it across the 52 trades the target demands: the probability of at least one four-loss streak somewhere in that sequence works out to 84%. The grind survives roughly one time in six. After a reset, the buffer is 2,500 / 200 = 12.5R, a breach requires thirteen consecutive losses, and the chance of that across the same 52 trades is about 0.25%. For $85, survival moves from roughly 16% to 99.8%.

The expected-cost framing stings more. Riding costs nothing today, but at an 84% breach probability you should expect to pay for a restart later anyway: about 0.84 x 85, or $71 in expectation, plus the weeks burned grinding a doomed account. You pay nearly the same money either way. Only one of the two paths also buys you a survivable account.

A caveat before anyone quotes these numbers. This is a simplified streak model: it ignores that interleaved wins rebuild some buffer, and it ignores that trailing drawdowns ratchet the floor up behind winning trades, which cuts the other way. The direction of the conclusion is robust; the exact percentages belong to the model, not the market. For the deeper version of this arithmetic, see the drawdown recovery math for funded accounts.

When resetting is correct

Strip out the emotion and the decision reduces to two numbers you can compute before any session in drawdown: B, the dollars of buffer between your balance and the liquidation or trailing level, and R, the dollars you risk per trade.

• Reset when B/R is smaller than a normal losing streak for your win rate. Near 50%, a five-or-six-loss streak is not an anomaly, it is a scheduling question: in the model above, at least one six-loss streak appears within 50 trades with 31% probability, and at least one four-loss streak with 83% (model outputs, not market statistics). If B/R sits below 5 to 6 and you are not within a day or two of target, reset.
• Reset after a tilt or rule-violation day, even if B/R looks fine. An account state that invites revenge trading has zero psychological buffer left. The reset restores your head along with the balance.
• Reset when the price is trivial next to the time at stake. If you hold a free reset credit, or the fee is a rounding error beside the weeks a doomed grind would burn, the decision is already made. Check credit expiry too; at Topstep, credits issued on or after December 11, 2025 lapse a year later (verify the current terms).

When riding it out is correct

The reset-early logic has boundaries, and pretending otherwise would turn this article into a sales page for reset fees.

• Shallow drawdown. If B/R comfortably exceeds your realistic worst streak, say 10R or more of room around a 50% win rate, a reset buys you very little. Keep trading.
• Striking distance of target. Within a day or two of passing at normal size, with buffer that covers an ordinary streak, finishing beats restarting. A reset here throws away progress you already paid for in time.
• No reset exists. On one-shot products (Apex 4.0-style evaluations, at the time of writing), the only lever left is position sizing. Sizing down stretches B/R without a fee, at the cost of slower progress to target.

One warning about the middle ground. If the buffer is thin and the honest fix is cutting size so far that the target moves months away, that is not riding, it is deferring the decision while the subscription meter runs. Do the math, then act on what it says.

The psychology of the reset button

The reset button is one of the few things in prop trading that is mispriced in the trader's favor, and only when used unemotionally. The $1,800 already lost in our scenario is gone whether you press the button or not; the only live question is what the next dollar of risk costs. Yet traders will burn three weeks grinding a 3.5R account to avoid an $85 admission that the attempt failed. The fee was never the expensive part.

Nowhere else in trading can you buy survivable risk at five cents on the dollar. The fee is not what makes a reset expensive. The ego is.

The flip side is one the firms understand perfectly: resets and rebills are a core revenue engine of the evaluation business. Free reset credits on renewal, like the ones Topstep banks with every rebill, are sharp retention design, because a "free" reset makes staying subscribed feel costless, and that keeps marginal traders on the treadmill. For the full picture of that engine, read how prop firms actually make money. And firms keep re-engineering the lever: Apex deleting resets from its 4.0 one-shot model proves availability itself is a moving target. And if the treadmill itself is what exhausts you, instant funding versus evaluations walks through the alternative honestly.

Resets on funded accounts

Here the defaults flip. Resets apply to evaluations; a breached funded account usually cannot be reset and requires passing a new evaluation from scratch. Topstep, for instance, restricts resets to active Trading Combine subscriptions, and its Express Funded and Live Funded accounts are ineligible.

Exceptions exist, and they are priced like exceptions. Topstep sells Back2Funded, up to two reactivations of a lost Express Funded Account, available only before your first payout, at roughly $499 to $729 by account size, with the offer lapsing 30 days after the breach. TakeProfitTrader sells PRO resets at roughly $399 to $1,499 by size, capped at three (published figures conflict on the top tier), while PRO+ has no reset at all; losing one means starting over from the test. MyFundedFutures documents a sim-funded reset on its Flex plan, again with a short claim window. All of these are time-sensitive; read the firm's current rule page before assuming any of it still holds.

The takeaway is blunt: funded resets are rarer, far more expensive, and heavily conditioned. Assume a breach means a new evaluation until your firm's rules say otherwise, and run your funded buffer math against that cost, not against an $85 evaluation reset.

A reset decision rule

Everything above compresses into six steps. Compute B and R before every session in drawdown, then walk the list.

1. Reset if B/R is smaller than a normal losing streak for your win rate (below roughly 5 to 6 near a 50% win rate) and you are not within a day or two of target.
2. Reset after a tilt or rule-violation day regardless of B/R. Zero psychological buffer is still zero buffer.
3. Ride if drawdown is shallow, B/R comfortably exceeds your realistic worst streak, or you are close to target at normal size.
4. Before paying, compare that day's reset fee against that day's discounted new-eval price, and check for a free credit. Buy the cheaper restart.
5. Walk if this would be roughly your third reset on the same strategy. You are no longer buying buffer, you are subsidizing a negative edge. Go to sim, collect 50 to 100 trades of data, and come back only if the numbers say the edge exists.
6. Funded breach: do not assume a reset exists. Read the rule page first; the default outcome is a new evaluation.

A word on copiers, because we build one. Thor copies trades across futures and CFD accounts for funded traders, so take this from the people with the least incentive to say it: a copier is not a fix for a failing evaluation, and it can make failure more expensive. Mirroring one strategy across several eval accounts means one bad day breaches all of them at once, multiplying reset fees across perfectly correlated failures; several firms restrict cross-account copying anyway (verify before you copy). A copier scales an edge across accounts; it scales the absence of one just as efficiently. Prove the edge on one account, then multiply it.

A reset fixes account state, never expectancy. If the strategy loses money, a fresh $2,500 of buffer just takes longer to burn; if it makes money, the reset is the cheapest insurance the industry sells. Know which trader you are before you press the button, then press it without ceremony.

Frequently asked questions