20

Due to the recent discussion about edit flooding, and one possible counter measure mentioned by @Grace Note ...

You may consider that it is often wiser to perform massive edit jobs during low period hours.

... and @random asking in a comment:

Can we get stats on when a site has its lowest views?

I hereby officially ask: When is the low period hour during the day for Super User? Are there stats or any insightful data dumps?


By the way: From my personal experience, it seems that Monday (08:00 until 15:00 UTC) is relatively "boring", probably because most U.S. people go to bed early or are asleep.

4
  • 1
    +1 interesting question, I look at the activity at intervals but it seems that statistics might work out much better; but in the end we should solve the actual problem somehow... Aug 10, 2011 at 22:11
  • 1
    Yup. It's no absolute answer to the original problem, but it would at least mitigate some of the issues.
    – slhck
    Aug 10, 2011 at 22:12
  • 2
    +1 because I'm interested in what this question asks for. I'm not favourable to the idea of limiting edits though, because I think that editing is an important feature that gives this community more depth by allowing everyone to participate. Aug 11, 2011 at 2:49
  • 2
    Just asked this - meta.stackexchange.com/questions/102012/… Aug 11, 2011 at 23:31

3 Answers 3

12

Any time during the week from around 8 pm Pacific (this is around 3:00 UTC) to 12 am Pacific is usually fine if you're trying to target low traffic times.

The absolute least traffic is Saturday night, same time.

As far as editing goes, it's best to just do it in smaller chunks over time, if you have, say, more than 10 edits to do at once.

12

Here are some usage statistics from Quantcast, which is used to track usage of SE sites (you'll see it as a tracking pixel when you check the loaded resources for each page).

enter image description here

Unfortunately, I haven't found how to get better resolution than day.


Using variants of @DaveDuPlantis' query, I created the following charts, that display the requested information in some more detail than the answers provided so far:

All posts on SU by hour of day:

enter image description here

select
datepart(hour, Posts.CreationDate) Hour,
count(Posts.Id) Questions
from Posts
group by 
datepart(hour, Posts.CreationDate)
order by Hour asc

All posts of SU by hours of day and day of week:

enter image description here

select
datename(weekday, Posts.CreationDate) Weekday,
datepart(hour, Posts.CreationDate) Hour,
count(Posts.Id) Questions
from Posts
group by datename(weekday, Posts.CreationDate),
datepart(hour, Posts.CreationDate)
order by Weekday, Hour

I manually fixed the output to bring order to the weekdays


Charts for 2011 only (using Oct 4 data set)

The following are the same charts, but limited to posts created in 2011 only, by adding the following to the queries above:

where YEAR(CreationDate) = 2011

enter image description here enter image description here

Both charts for 2011 only are basically identical to those for the whole lifetime of the site. The absolute numbers are way down though, so it's not the domination of 2011 in the first charts that causes this: It's just that there was no shift in the times people post to the site.

2
  • It's possible I'm really wrong in my conclusion, since I know nothing about statistics.
    – Daniel Beck Mod
    Oct 15, 2011 at 10:08
  • 1
    great post, UTC 3:00 is basically what I indicated in my answer, and that's true here. (though technically UTC 4:00 and UTC 5:00 are even lower traffic.) Jul 30, 2012 at 7:02
5

Saturday and Sunday (UTC time) are the lowest activity days of the week on Super User.

3
  • 2
    Any specific times? Or aren't these statistically available to moderators? Aug 10, 2011 at 23:32
  • 1
    I believe this is in the morning of UTC people, as USA people are still sleeping at those hours... Aug 11, 2011 at 11:00
  • Well, the fewest questions are asked from 0200 through 1059 UTC on Saturday and Sunday, if that helps any. (We had a similar question on Gaming, so I modified an existing query to run for that site and used it again for SU.) Aug 12, 2011 at 17:26

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .