Delhi Noon Archive Statistics
Descriptive statistics for the Delhi Noon archive: how often each number 00 to 99 appears, how records spread across bands, and how complete each year is.
In short
These statistics describe what the archive already contains: how many times each value from 00 to 99 has been recorded, how entries spread across the ten bands, and how complete each year is. They summarise history and predict nothing about any future entry.
These figures describe the past only
Frequency counts summarise what the archive already contains. They carry no information about any future entry, and reading them as a forecast is the gambler’s fallacy. They are published here because a record set should be measurable, not because the measurements predict anything.
- Records analysed
- 1,361
- Distinct values
- 100
- Average per value
- 13.6
- Observed range
- 6 to 21
Every archived entry
From 00 to 99
If spread perfectly evenly
Least to most recorded value
How often has each number been recorded?
All one hundred values, laid out in rows of ten. A darker cell has been recorded more often across the archive; the count sits under each value.
Shade is relative to the most recorded value in the archive (21 occurrences), not to any absolute scale.
Most and least recorded values
Recorded most often
Recorded least often
Ties are broken by the value itself so this ranking is identical on every request. A gap of a few occurrences across 1,361 records is ordinary variation in any finite record set, not a pattern.
How do records spread across the ten bands?
Grouping the hundred values into ranges of ten gives a coarser view, and a clearer answer to whether the archive is evenly spread.
How many records does each year contribute?
Completeness matters before any other statistic: a year with fewer records contributes proportionally less to every count above.
2026 is short because it is still in progress, holding records only to 22 Sep 2026.
How these statistics are produced
Every figure on this page is computed at request time from the same stored entries that the chart pages render. There is no separate statistics database to fall out of step, and no rounding beyond the decimal places shown.
A value is counted once per dated entry. Dates with no archived record contribute nothing rather than being imputed, which means the counts describe the archive as it actually is, including its gaps. Where a correction changes an entry, these counts move with it on the next request.
What these counts do not support
Every figure above describes entries that have already been recorded. None of it carries any information about a future date, and a low count does not make a value “due”. If that conclusion feels wrong, it is worth reading why past charts cannot predict the next number, which sets out where the intuition comes from and why it fails here.
Frequently asked questions
What is a frequency table?
A frequency table counts how many times each possible value appears in a data set. Here it counts each number from 00 to 99 across every archived entry, so a value with a count of 18 has been recorded on 18 separate dates.
Does a high frequency make a number more likely to appear again?
No. Counting how often something occurred in the past tells you about the past only. Treating a frequency table as a forecast is the gambler's fallacy, and this site publishes these tables strictly as a description of the archive.
What do the bands 00-09, 10-19 and so on show?
Bands group the hundred possible values into ten ranges of ten. They give a coarser view than the full frequency table and make it easy to see whether the archive is spread evenly across the range or clustered in part of it.