Miss pattern analyzer
Most missed questions are not random. They cluster in a few question types: the flaw you keep misreading, the inference you keep rushing. This analyzer reads your own drill and timed-section history, groups your misses by question type, and ranks your weakest types first, so your next study block goes where the points are leaking.
Most missed questions are not random. They cluster in a few question types: the flaw you keep misreading, the inference you keep rushing. This analyzer finds those clusters in your own drill and timed-section history, so your next study block goes where the points are leaking.
It reads your drills and timed sections, groups misses by question type, and ranks your weakest types first.
How the method works
- It reads your attempts. Every drill and timed-section answer you have submitted, newest first.
- Your most recent answer decides. If your latest attempt at a question is wrong, it counts as a miss. Answer it correctly later and the miss clears. This keeps old mistakes from haunting you after you fixed them.
- Misses group by question type. LR types (flaw, strengthen, necessary assumption, and the rest), RC question kinds. Each group gets a miss rate: misses divided by questions attempted in that type.
- Weakest first. Types rank by miss rate, so the top of your list is the type costing you the most points right now.
What you get
- Miss rate per question type. Not just a count: five misses out of six attempts is a bigger problem than five out of fifty.
- Only types with enough data. A type needs at least three attempts before it gets a rate. With fewer, the number is noise and we say so.
- Nothing invented. No modeled probabilities, no predictions. Counts from your own answers.
What to do with the results
- Take your top type and re-drill it: your missed questions deck is already grouped for this.
- Before you re-attempt, write down in one sentence what the type asks. If you cannot, study the type first.
- Fix one type at a time. Watch its miss rate fall over the next week, then move to the next.
Frequently asked questions
How does the analyzer decide what I missed?
It looks at your drill and timed-section attempts and uses your most recent answer on each question. If your latest attempt is wrong, it counts as a miss. A later correct answer clears the miss.
How many attempts do I need?
About five answered questions before the page shows anything, and at least three attempts in a question type before that type gets a miss rate. Fewer attempts than that is mostly noise.
Is my drill data shared with anyone?
No. The analysis runs on your own attempts inside your account. Nothing about your scores is published or shared.
Does this replace blind review?
No. Blind review tells you why you missed one question. The analyzer tells you where your misses cluster across weeks. Use both: the analyzer picks the target, blind review fixes it.
Take the free diagnostic. It gives the analyzer its first real data and gives you a baseline score.
Last updated October 12, 2026