Spaced Repetition Calculator

Spaced repetition is a learning technique that schedules reviews at increasing intervals over time, exploiting the psychological spacing effect. Rather than cramming, you review material just as you're about to forget it, which forces deeper memory consolidation. Research by Ebbinghaus and later by Cepeda et al. confirms that spaced practice can improve long-term retention by 200% or more compared to massed study.

The Ebbinghaus Forgetting Curve, discovered by Hermann Ebbinghaus in 1885, describes how memory strength decays exponentially over time without reinforcement. It shows that we forget roughly 50% of new information within an hour and up to 70% within 24 hours. Spaced repetition works against this curve by scheduling reviews at the moment retention drops to a target threshold, rebuilding the memory trace each time.

The classic Ebbinghaus approach uses fixed exponential intervals (e.g., 1, 2, 4, 7, 15, 30 days) regardless of how well you know each card. SuperMemo 2 (SM-2), developed by Piotr Wozniak, introduces an individual Ease Factor (EF) per card that starts at 2.5 and adjusts up or down based on your recall quality rating (0–5). Easier cards get longer intervals, harder cards get shorter ones — making the schedule adaptive to your actual memory performance.

FSRS (Free Spaced Repetition Scheduler) is a modern open-source algorithm developed by Jarrett Ye that models memory using parameters for Stability (S) and Difficulty (D). It is more mathematically rigorous than SM-2 and is trained on millions of real review logs, making its interval predictions more accurate. FSRS is now the default algorithm in Anki and is considered the current state of the art in spaced repetition.

The default ease factor of 2.5 works well for most average-difficulty material. Use values closer to 1.3 for highly complex content like medical terminology or advanced mathematics where you find yourself forgetting frequently. Use values closer to 3.5 for simpler material such as basic vocabulary or well-familiar concepts. In practice, algorithms like SM-2 and FSRS will automatically adjust this value per card as you provide recall ratings.

A target retention rate of 85–92% is the sweet spot recommended by memory researchers. Going higher (e.g., 99%) means reviewing much more frequently, which increases your daily workload significantly. Going lower (e.g., 70%) reduces workload but means forgetting more material at each review. For critical professional or exam-based learning, 90% is the most common recommendation.

Most learners can review approximately 1 card per minute including thinking time, so 20 minutes of daily study allows roughly 20 reviews. New cards take longer (2–3 minutes each) because learning requires more effort than reviewing. A sustainable deck size for a beginner is 10–20 new cards per day, with total daily reviews growing to 50–150 cards as the deck matures. This calculator estimates your reviewable cards based on your available study time.

For language learning (vocabulary, kanji, phrases), FSRS or SM-2 are ideal because they adapt to the irregular difficulty across thousands of cards. For structured exam preparation with a fixed deadline, the Ebbinghaus-based fixed-interval approach can be simpler to plan around, since you can align the review schedule to your exam date. For any serious long-term learning project, FSRS currently offers the best retention-to-effort ratio.