Perceptle stats from the public API
Snapshot 2026-10-23T02:13:53.094Z from perceptle.com/api/stats. An earlier read of #97 at 13:10 UTC on 22 October was 8,358 plays; the row below is the later pull. This is not a leaderboard, and it is not your percentile.
I wanted a simple answer to “is 65 good?” The API does not give a distribution, so I will not invent one. What it does give is a daily player count and six averages. From 9 to 23 October the total averages cluster from about 60 to 71, except two Fridays: 16 October fell to 48.8 and 23 October (#98) is at 50.0 with only 2,036 plays so far. A 65 on a normal day is sitting on the crowd, around a C. The letter C starts at 55 and B starts at 70, so 65 is a C with room under the next grade.
Tempo is the outlier I keep coming back to. In this window it is often the only average under 50. On #97 the later snapshot shows Tempo at 31.6, Split at 77.6, and Track at 78.3, across 20,937 plays. When I played, Count and Track were readable and Tempo was where the run collapsed. That matches the table better than any story about “people are bad at counting.” Friday #98 is harsher still: Tempo is 13.7 and the total is 50.0.
| # | Date | Players | Count | Track | Angle | Tempo | Split | Total |
|---|---|---|---|---|---|---|---|---|
| 84 | 2026-10-09 | 6251 | 84.9 | 52.7 | 63.6 | 24.8 | 78.9 | 61 |
| 85 | 2026-10-10 | 4233 | 76 | 84.2 | 76.2 | 37.3 | 65 | 67.7 |
| 86 | 2026-10-11 | 4449 | 71.6 | 82 | 68.5 | 53.5 | 77.5 | 70.6 |
| 87 | 2026-10-12 | 6791 | 72.5 | 79.4 | 64.3 | 45.7 | 69.5 | 66.3 |
| 88 | 2026-10-13 | 7280 | 71.1 | 77.1 | 74.4 | 35.1 | 62.8 | 64.1 |
| 89 | 2026-10-14 | 7928 | 74.5 | 79.9 | 85.5 | 51.8 | 61.8 | 70.7 |
| 90 | 2026-10-15 | 15960 | 62.7 | 72.8 | 61.7 | 47.3 | 64.4 | 61.8 |
| 91 | 2026-10-16 | 46393 | 79.8 | 35.2 | 55.4 | 24.6 | 49.2 | 48.8 |
| 92 | 2026-10-17 | 25050 | 64.5 | 72.2 | 54.5 | 45.7 | 63.5 | 60.1 |
| 93 | 2026-10-18 | 20380 | 65.5 | 79.8 | 71.6 | 40.6 | 53.7 | 62.2 |
| 94 | 2026-10-19 | 23556 | 66.6 | 78.2 | 65.2 | 35 | 72.8 | 63.6 |
| 95 | 2026-10-20 | 22133 | 70.8 | 77.3 | 64.7 | 33.5 | 81.6 | 65.6 |
| 96 | 2026-10-21 | 22338 | 68.3 | 76.2 | 57.8 | 43.8 | 79 | 65 |
| 97 | 2026-10-22 | 20937 | 64.2 | 78.3 | 71.7 | 31.6 | 77.6 | 64.7 |
| 98 | 2026-10-23 | 2036 | 84.2 | 42.1 | 63.6 | 13.7 | 46.6 | 50 |


How to read a row
- Players is a count of submitted runs, not unique browsers. The API does not say which.
- Each test average is already on the 0–100 curve. It is not a raw error.
- The total is the average of the five tests, as the client defines it.
- A blank day on the archive means we did not save an API row. It does not mean nobody played.
- Your streak lives in the game’s local save. This page cannot see it.
The daily update script refreshes this snapshot from the same API. It will not add a rank, a percentile, or a name. If you want to turn your own five test scores into a grade, use the calculator. The grade cuts do not move when the averages move.
What is the average Perceptle score?
In this 9–23 October 2026 snapshot, daily total averages were mostly in the low 60s. Friday 16 October was 48.8 and Friday 23 October is 50.0 so far. That is a snapshot, not a lifetime average.
Can I see a leaderboard?
Not here. perceptle.com can show a “beats this percent” line after you submit, and that line changes as more people play. We do not copy it into a ranking.