Update Site:
Rolling average:
Rolling average:
Analyzing the data yourself
You can download the raw data from:
Here is an example Python script to list all update sites with more than 5000 hits in the year 2022:
import json
year = 2022
cutoff = 5000
with open('stats.json') as f:
stats = json.load(f)
recent_totals = {
site: sum(v for date, v in counts.items() if date.startswith(f'{year}-'))
if isinstance(counts, dict) else 0
for site, counts in stats.items()
}
popular = dict(sorted(recent_totals.items(), key=lambda x:x[1]))
for site, total in popular.items():
if total > cutoff:
print(f"* {site} = {total}")