More candidates fail CFA Level I than pass it. In the May 2026 window, 39% of the 31,566 candidates who sat the exam passed. That is not an unusual result: CFA Institute puts the long-term average at 40%.
Numbers like that invite a comfortable explanation. The exam is brutal, the curriculum is vast, and only the most gifted get through. It is an explanation that protects everyone's ego and helps nobody prepare.
Our view is different. Most candidates do not fail because they lack intelligence. They fail because their preparation system breaks down. The breakdown usually takes one of four forms: passive reading, a late start on practice questions, time spread evenly across unequal topics, and no review loop.
Many failing candidates worked hard. They worked hard on the wrong loop.
CFA Institute publishes no data on why candidates fail, so the four patterns below are observed patterns in how candidates prepare, not measured causes. Each comes with the signal that exposes it and the fix.
The Pass Rate Reality
A single window can mislead, so look at a run of them. The official Level I results over the last five windows tell a consistent story.
| Window | Pass rate | Candidates sat | First-time | After a deferral |
|---|---|---|---|---|
| May 2026 | 39% | 31,566 | 45% | 25% |
| February 2026 | 45% | 24,006 | 50% | 30% |
| November 2025 | 43% | 26,752 | 49% | 29% |
| August 2025 | 43% | 26,192 | 50% | 29% |
| May 2025 | 45% | 24,227 | 52% | 28% |
Two things stand out. First, no window reached half. The 10-year average CFA Institute quotes is 40% in some releases (May 2025, February 2026) and 41% in others (August and November 2025). Second, the gap between first-time candidates and those who had deferred at least once appears in every window, and it is large: in May 2026, 45% against 25%.
That gap does not prove cause. Candidates defer for many reasons. But it is hard to explain as a difference in intelligence, and CFA Institute does not try to: "Remaining on track with CFA Program study and exam preparation remains the clearest indicator of exam success." That is a statement about process, not ability.
The scoring structure points the same way. The pass mark, the minimum passing score, is set by CFA Institute and maintained across administrations by equating; it is not published. There is no minimum passing score per topic, so strong topics can offset weak ones, and the result depends on the overall score. What decides the outcome is total points, and total points are the product of a system.
Failure Pattern 1: Passive Reading Without Retrieval
The passive reader reads every reading, highlights carefully, makes neat summaries and then rereads them. The effort is real. The problem is that almost none of it requires producing an answer.
The research on this is unusually clear. A major review of ten learning techniques by Dunlosky and colleagues (2013) rated practice testing and distributed practice as high utility, and rated rereading, highlighting and summarization as low utility. Karpicke and Roediger (2008) found that repeated studying after items were learned had no effect on delayed recall, while repeated testing had a large positive effect.
The same study found something more troubling: students' predictions of their own performance were uncorrelated with how they actually did. That explains why this pattern survives so long. Rereading produces fluency, and fluency feels like knowledge. The passive reader does not notice the problem until the exam asks for recall instead of recognition.
The signal: you can follow a worked solution but cannot reproduce it with the text closed. Your study log shows many hours of reading and few questions answered.
The fix: end every reading with closed-book questions on that reading, and treat rereading as a response to a wrong answer, not a default activity. The mechanism is explained in more depth in Why Rereading Notes Fails.
Rereading measures how familiar the page looks. The exam measures what you can produce without it.
Failure Pattern 2: Practice Questions Started Too Late
The late starter plans to finish the curriculum first and "move to questions" in the final phase. It sounds logical: learn, then test. In practice, it means the first honest measurement of readiness arrives when there is no time left to act on it.
No published data link the date a candidate starts practice questions to their chance of passing, so treat this as observed practice rather than a measured effect. The logic, however, follows directly from the retrieval research above. Questions are not only an assessment. They are among the highest-utility study techniques in the research, and delaying them means spending months on lower-utility ones instead.
There is a second cost. A candidate who meets questions late also meets the exam's way of asking late: how a concept is phrased in a question stem, which wrong answers are built from predictable errors, and how quickly a decision has to be made. None of that is learned from reading. It is learned from answering, getting it wrong and understanding why.
CFA Institute's own advice supports an early, steady approach. It says to start studies as early as possible and to "Leave a minimum of 4 weeks at the end for review." It also warns: "Do not focus solely on drilling with practice exams." The point is not a late block of mocks. It is questions throughout.
The signal: your first question set came weeks into your preparation, and your first full mock is planned for the final month.
The fix: answer questions from the first week, on every reading, and track results by topic. The guide to using practice questions properly covers how to turn those results into decisions.
Failure Pattern 3: Equal Time on Unequal Weights
The weight-blind planner follows the order of the curriculum and gives each topic roughly the same time. It feels fair and thorough. The exam does not reward fairness.
For the 2026 Level I exam, Ethical and Professional Standards carries 15-20% of the exam, while Derivatives carries 5-8%. Financial Statement Analysis, Equity Investments and Fixed Income each carry 11-14%. Quantitative Methods, Economics and Corporate Issuers each carry 6-9%, Alternative Investments 7-10% and Portfolio Management 8-12%.
The table shifts for 2027. Ethics falls to 10-15%, Quantitative Methods rises to 11-14%, and Derivatives and Risk Management moves to 6-9%. A candidate using an old plan for a 2027 sitting is misallocating before the first reading. This is where candidates make mistakes: they plan for the exam in the book, not the exam on the calendar.
Because there is no per-topic pass mark, time is a portfolio decision. An extra week on a 5-8% topic you already understand buys fewer points than the same week on an 11-14% topic you are weak in.
The signal: your hours by topic look flat, or follow reading order, and you cannot say which three topics carry the most weight on your exam.
The fix: allocate time by weight first, then adjust by your own weakness. The topic weights breakdown shows where the points actually come from.
Failure Pattern 4: No Spaced Review, So Knowledge Decays
The no-review crammer moves steadily through the curriculum and never looks back. Each topic is learned once, well enough to pass the end-of-reading questions, then left untouched until the final weeks. By then, the topics from month one have largely faded, and the last month becomes an attempt to relearn half the exam.
CFA Institute's advice on this is three words long: "Cramming doesn't work." The same review that rated practice testing highly gave distributed practice, spreading study of the same material over time, the same high-utility rating. Knowledge that is not revisited decays. Knowledge that is revisited at intervals holds.
This pattern often hides inside a plan that looks disciplined. Hours are logged every week, the curriculum advances on schedule, and the candidate feels on track. The flaw only shows when a mock tests early material.
The signal: there are topics you studied more than six weeks ago that you have not answered a single question on since.
The fix: build review into the weekly plan. Every week, a fixed slot goes to questions on older topics, weighted toward the ones where your results are weakest.
A Self-Diagnostic: Five Questions to Test Your System
Each pattern leaves evidence in how you study. Answer these five questions honestly, using your study log rather than your impression.
- In the last week, did you spend more time answering questions than reading? If not, you are at risk of pattern 1.
- Can you solve a typical problem from your last topic with the book closed? Following a solution is not the same as producing one.
- Did you start answering practice questions in your first week of study? If your first questions came late, you are at risk of pattern 2.
- Do your hours by topic reflect the weights of the exam you will sit? Check against the 2026 or 2027 table, not a general impression. This tests pattern 3.
- When did you last answer questions on the first topic you studied? If the answer is "months ago", you are at risk of pattern 4.
Two or more weak answers do not mean you will fail. They mean your system has a leak, and leaks compound over months. Each fix above is small on its own. Together, they turn effort into points.
The exam does not reward effort on its own. It rewards retrieval, allocation and review, repeated over months. WSU exists to keep that system from breaking down: its study engine plans each day around the exam weights, schedules reviews at expanding intervals, and tracks which topics your own results have verified.