Why Evaluations Fail: The Oversizing, Stop-Moving, No-Risk-Model Pattern

Guide - Prop Trading

You spend weeks tuning entries. You backtest until the numbers look clean. Then the evaluation account blows through its drawdown limit in three bad days, and you're left wondering what happened. Here's the uncomfortable answer: it probably wasn't your strategy.

The evaluation isn't testing your strategy - it's testing your risk process

Most traders treat an evaluation like a strategy test. They think: if my edge is real, the account will grow and I'll pass. So they spend all their prep time on entries, indicators, and market structure, and almost none on how they'll size a position or where they'll cut a loss.

Then the account fails - not because the strategy stopped working, but because a handful of trades were oversized, a stop got moved once too often, or there was never a fixed rule for how much to risk in the first place. The strategy might have a real edge over 200 trades. It doesn't matter if the account doesn't survive the next 20.

This is the pattern that shows up in almost every failed evaluation, regardless of the market or the setup being traded:

• Oversizing relative to the stop and the account's remaining risk budget
• Moving the stop after entry instead of respecting the original plan
• No written risk model connecting size, stop distance, and account drawdown

These aren't three unrelated mistakes. They're the same root problem seen from three different angles. Let's break each one down before showing how they connect.

Symptom 1: Oversizing

Oversizing doesn't usually look reckless from the inside. It looks like conviction. "This setup is clean, I'm sizing up." The size gets decided by how good the trade looks, not by the distance to the stop or by how much of the account's risk budget is actually available that day.

Here's the problem: a trade that looks great can still lose. When it does, a position sized on conviction rather than on risk produces a loss that's disproportionate to the account's tolerance. One or two of those losses in a row, and a chunk of the available drawdown is gone before lunch.

Oversizing is especially dangerous because it works - for a while. A trader who sizes up on "good" setups will often get away with it for a stretch of winning trades. That reinforces the habit. Then the losing streak arrives, usually when confidence is highest, and the same sizing logic that felt smart now eats through the account far faster than a string of normally-sized losses ever would.

The math is simple even if it doesn't feel that way in the moment: size determines how fast you consume your risk budget. A trader risking a fixed, small percentage per trade can survive a long losing streak. A trader who doubles size on "high conviction" trades can undo weeks of careful trading in an afternoon, even with a decent win rate.

Symptom 2: Moving the stop

There's a real difference between adjusting a trade because new information invalidates the original idea, and moving a stop because the price is close to hitting it and you don't want to take the loss. The first is trade management. The second is denial.

Moving a stop after entry - just to give the trade "more room" - is often worse than not having a stop at all. A trade with no stop is at least an honest, if reckless, admission of unlimited risk. A trade with a stop that gets pushed back turns a defined, planned risk into an undefined one, right at the moment it's being tested. The trader thought they knew their risk. They didn't.

This interacts badly with drawdown structures that track the account's highest equity point. As a losing trade widens, the unrealized loss eats into the buffer between current equity and the drawdown limit faster than the trader tends to notice, because attention is fixed on the price action, not on the account-level number. By the time the stop is finally hit - now much further away than planned - the available risk margin has shrunk by more than expected.

The fix isn't complicated, it's just uncomfortable: the stop is a contract you sign before the trade, not an opinion you revisit while the position is open. If the stop level is wrong, that's a lesson for the next trade. It's not a reason to redraw the line while you're already in the position.

Symptom 3: No defined risk model

"Trading by feel" sounds harmless. In practice it means the size of every trade is decided in the moment, with no fixed rule behind it. One trade risks a small slice of the account, the next risks three times as much, based on nothing more than how the trader feels about the setup that day.

A real risk model answers a few questions before the platform even opens:

• What's the maximum risk per trade, as a fixed percentage of the account?
• What's the maximum aggregate risk allowed in a single day?
• How many correlated positions can be open at the same time before total exposure is capped?

Without answers to these written down somewhere, every risk decision becomes reactive. After a loss, a trader might size up to "win it back." After a winning streak, the same trader might size up because it "feels safe." Neither decision has anything to do with the actual state of the account or the actual risk of the next trade. That's not a strategy problem. It's the absence of a process.

Why these three symptoms always show up together

Here's the causal chain, and it's worth sitting with because it explains why fixing just one piece rarely works. No risk model means there's no fixed rule for size. Without a fixed rule, size gets decided emotionally - which usually means it's too big on the trades that feel most convincing. A stop placed under an oversized position feels "too tight," because the dollar amount at risk feels large relative to the position, not relative to the account. So the stop gets moved. The loss, when it comes, is both bigger than planned and further from the entry than planned.

These aren't three separate errors that happen to occur in the same account. They're the same missing piece - a defined risk model - viewed from three different moments in the trade: before entry (sizing), during the trade (the stop), and at the account level (no ceiling on aggregate exposure). That's why traders who fix only the size, without addressing the underlying model, often drift right back into the same pattern a few weeks later. The habit that caused the oversizing is still there. It just finds a new way to express itself.

How evaluation drawdown structures expose this pattern

Evaluation accounts are built around a small set of common mechanisms: a maximum drawdown limit, sometimes a daily loss limit, and in some structures a trailing component that moves with the account's equity, occasionally locking in place once it reaches a certain floor. These mechanics vary between providers and between account types, and they get updated over time, so the exact percentages and thresholds aren't something to memorize from a blog post - they're in the full challenge rules, and that's the page to check before funding any account.

What matters conceptually is this: a trailing structure punishes oversizing and moved stops faster than a static one. In a static drawdown, the distance to the limit is fixed from day one. In a trailing structure, that distance can shrink as the account's equity climbs, because the limit itself moves up with new highs. A trader who oversizes into a winning streak, pushing equity higher, can end up with less room to breathe than they think - the drawdown line has followed the equity up, even though the trader's mental model of "how much room I have left" hasn't updated.

Daily loss limits, where they exist, add another layer: they're often calculated against the account's highest reached equity for that day, not against the starting balance. That means a stop that gets moved mid-session, turning a small planned loss into a larger unplanned one, can eat through a daily limit far faster than the trader expects.

Not every model works this way. Some funding structures skip the daily loss limit entirely and govern risk purely through the overall maximum drawdown boundary, while others combine a daily limit with the overall one - and which mechanism applies, and how it's calculated, depends on the specific account type and can change over time, so it's never something to assume from memory. Where no daily cap exists, that doesn't remove the need for daily discipline - it just shifts the responsibility for it entirely onto the trader. Without an externally enforced daily stop, an unplanned bad day has more room to become a very bad day before any rule intervenes. Check the specific mechanics for each account type on the challenge rules page before assuming how any given structure works.

Building a risk model before your next evaluation

None of this requires anything complicated. It requires five decisions made in advance, in writing, before the platform is even open.

1. Fix the risk per trade as a percentage, not a dollar amount. A fixed percentage of current equity scales automatically as the account moves, instead of staying frozen at a number that made sense on day one and stops making sense after a drawdown or a run of gains.
2. Size follows the stop, never the other way around. Decide where the trade is invalidated first. Then calculate the position size that makes that distance equal to the fixed risk percentage. If the math produces a size that feels too small for the conviction level, that's a signal to skip the trade, not to widen the stop.
3. Cap aggregate risk for the day and the week. This applies whether or not the account has a daily loss limit built into its rules. A self-imposed ceiling - for example, no more than a fixed multiple of the per-trade risk lost in a single day - stops one bad session from turning into an account-ending one.
4. Write the rules down before opening the platform. Not after the first trade. Not halfway through a losing position. A risk model decided in the middle of a drawdown isn't a risk model, it's a rationalization.
5. Review the model only at the end of the day or the end of the week. Never adjust size, stop placement, or risk caps while a position is open. That's exactly the moment when judgment is least reliable.

A model this simple takes fifteen minutes to write and far less time to check before each session than the hours usually spent tweaking entry signals.

What changes with Instant Funding and multi-step models

Funding structures generally fall into two broad shapes: some rely only on a maximum drawdown limit, with no daily cap and no staged profit target, while others add a daily loss limit and a profit target split across multiple steps. Neither shape is inherently easier or harder - they just shift where the pressure sits.

A model without a daily loss limit puts the entire responsibility for daily discipline on the trader. There's no built-in circuit breaker stopping a bad morning from becoming a bad week. That's exactly why the self-imposed daily risk cap from step three above matters more, not less, in this kind of structure - the absence of an external limit doesn't mean the risk of a runaway day disappears, it just means nothing external will stop it for you.

A multi-step model, on the other hand, adds a different kind of pressure: the temptation to rush a stage by increasing size once the target feels close. This is oversizing wearing a different mask - sizing based on how close the finish line looks, rather than on the stop distance and the fixed risk percentage. The fix is identical to everything above: the risk model doesn't change because the target is close. Details on how each structure is built, including drawdown type and profit targets, are on the challenge rules page, and current account options are listed on the pricing page.

Conclusion

The pattern that fails evaluations is almost always the same one, regardless of the market, the timeframe, or the strategy behind it: no fixed risk model, which leads to arbitrary sizing, which leads to stops that feel too tight, which leads to stops getting moved. Break the chain at the first link and the other two stop happening on their own.

A risk model written before the session starts is the only real defense against this. Before starting or restarting an evaluation, it's worth reading the full challenge rules and the risk disclosure in full - understanding exactly how drawdown and loss limits are structured is the first step to building a risk model that doesn't get discovered the hard way, mid-trade.