Hacker News new | past | comments | ask | show | jobs | submit login
The Best Time to Post on Hacker News (hevenet.com)
138 points by Nathanael on Nov 8, 2012 | hide | past | favorite | 46 comments

I have quite high karma on HN and I got it two ways:

1. By writing original content that the HN audience is likely to like

2. By submitting that content at the right time

#2 is useful to know and anyone can do an analysis and figure the best time. I did it myself some time ago and came to my own conclusion about the best time (which is similar to the OP). But #1 is far more important than gaming the system. Focus on writing good stuff.

Fair, although I am not sure that the author was suggesting that this is all about personal karma. Rather, he seemed to be focusing on the best way to get one's content to the front of the page, in order to maximize exposure to readers.

There may be any number of reasons someone would want to do this, karma accumulation being only one of them. Product launches, Kickstarter campaigns, advice being sought or offered, etc., are equally valid and perhaps more likely motives.

Furthermore, I don't think it's fair to infer that the author advocates "gaming the system," or at least not the exclusion of writing good content. Rather, I inferred that he took writing good content as a given, and focused on the oft-pondered problem of how to break out that good content in a crowded news market.

All of that being said, your point #1 is extremely important. People should focus on quality first. It may be possible to game some spammy content to the top of the pile. Eventually, though, this strategy will do the poster more harm than good, whatever his intentions may be.

I'd imagine a corollary to #2 is by replying to comments at or near the top.

I see what you did there...

Just kidding. But to contribute something to the discussion: I was just recently wondering when the best time to publish a "Show HN" would be. In that case, your content is what it is. The time factor would definitely be important then to help you get the "early exposure" that you'd need to even arrive on the front page in the first place.

You can spend a lot of time creating content that you think HN is likely to like, and submit it at the optimal time, and still not get much karma unless that entry gets picked up within an hour. Otherwise it's just pushed down and out, and your time has been wasted (assuming the goal is to get karma).

Since there is a degree of luck involved in getting to the front page, I hypothesize that you can leverage the law of large numbers to maximize your chances of getting karma by also submitting HN-targeted content regularly/frequently.

Does your high karma impact ranking?

No, but fanboism does :)

any advice on doing "show HN" correctly?


The real answer, as anyone who reads HN knows, is "a few hours after an already high ranking post, on the same controversial subject, but taking the opposite stand" :)

You forgot a crucial part: reversing the original title.

"No, NYC is NOT playing second fiddle to silicon valley"

Or if you're a REAL expert, "Is NYC playing second fiddle to silicon valley?"

And then, for bonus karma, leave a comment to the effect of "The answer to any headline ending in a question mark is no. http://en.wikipedia.org/wiki/Betteridge%27s_law_of_headlines ", as if you're the first and only person to have ever heard of it.

Remember to include a snide remark with a vague reference to some sort of a logical fallacy, in order to demonstrate your superior mastery of logic. Bonus points for using pompous latin expressions.

ipse dixit?

Let's downvote those perfunctory mentions of Betteridge's Law (and similar). (I already do.)

Sorry, missed that bit yeah :)

There is already an interesting tool[1] out there for this. I am surprised it wasn't mentioned anywhere in the post. I would also be curious to know how similar or different they are.

[1] http://hnpickup.appspot.com/

Hey, author here. I'm going to transgress my no-HN rule this one time because I did in fact look at HN Pickup very closely and I already have a follow-up post planned in which I was going to mention it.

Now the difference between my post and HN Pickup is it's a prediction tool, so it's trying to pin-point precise moments when it's best to post, whereas I'm analyzing past data.

So in my post today I talked about the best time to post on average. But, I can also look at my data and say, for example, last Monday from 10 to 11 AM, 30% of submissions reached the frontpage and so it was, in effect, a great time to post.

And with that I could look at HN Pickup's past predictions and tell you how accurate they were, ie. at a time when HN Pickup was telling you it's a "very good time to submit a story", what percentage of stories actually made it?

It so happens that I've actually been able to do that. HN Pickup's creator was kind enough to provide me with a history of predictions for every 15 minutes of a whole week, and after comparing it to observed pickup percentages I found a rather weak correlation (0.25). So it would seem HN Pickup isn't a very effective tool.

Also its goal is only to look at frontpage pickups, it doesn't account for the duration of the stay on the frontpage or HN's traffic at the time, which I do in my analysis.

> Hey, author here. I'm going to transgress my no-HN rule this one time because I did in fact look at HN Pickup very closely and I already have a follow-up post planned in which I was going to mention

I would have thought by posting you would have already broken that rule:)

   for example, last Monday from 10 to 11 AM, 30% of submissions reached 
  the frontpage and so it was, in effect, a great time to post.
But it might also mean that less people see your post, which is still better than not making the frontpage at all, but not the same as getting there on "prime time", this should also be taken into account.

It is taken into account in my article...

HN felt kind of frozen last week, maybe due to many people being out of power on the east coast. I imagine the amount of traffic gotten was lower, making the easy front page perhaps less cost effective. It might have been better for authors to hold articles until more audience was back online...

The best time to post on HN is when you have something interesting to share that will benefit the community.

Same holds for your email marketing, social media, etc. Good content is an order of magnitude more important than time-of-day optimizations (which in my experience, only helps some people some of the time).

In my experience, that's not always the case: sometimes interesting posts get drowned out.

I've seen this happen a couple of times. An article is submitted once only to drown with around five upvotes, and then reach the frontpage on a second submission.

There's some truth there, but if you send an email to the East Coast US at 2am, there's going to be a small chance people will read it when they wake up (they'll already have tons of other stuff competing at the top). Same goes with Twitter. You can't tweet something for everyone to see at a time when your target audience is not watching twitter. You need to test for that.

Either way, the essence of your argument is correct. Useful content fares better that fluff.

It depends on the source. Articles move up two ways. One is by votes, the other is by resubmission by different users. Peak times also mean peak competition.

If you are the first to post an article from a commonly frequented site such as TechCrunch, then it will tend to gain exposure because others are submitting it.

On the other hand, I've found that posting articles from less mainstream sources benefit from sitting longer on the "New" page. This means posting during times of lower activity. I have woken up to find something I posted just before bed has gained significant traction.

Interesting. I wasn't aware of this resubmission mechanism.

It's probably the way that some blog posts make it.

Some other suggestions to get you on the frontpage:

1. Posting anything by Marco/Gruber/Atwood/37Signals/Stripe.

2. Posting anything making fun of PHP.

Don't forget Github. Most of their blog posts that aren't regarding new hires hit the front page.

I would like to see a scoring algorithm that dampens (if not totally eliminates) time-of-day effects. Essentially, clock time shouldn't be a factor at all: just some other synthetic 'tick'. That tick is probably 'sitewide votes' (but might be 'legitimate votes' or 'net upvotes' or 'homepage views').

Secondarily, I would like to see a decay function that gives an older story with a current surge as much credit as a new story with similar recent surging.

(For example, imagine story A submitted at hour 0 which receive no votes for 6 hours. At hour 6, story B is submitted. Between hour 6 and hour 7, story A gets 12 upvotes, and story B only 8. Over the current-comparable period, A is hotter... but I'm pretty sure the HN algorithm and most similar sites penalize A for the earlier-submission time. And yet due to things like headline corrections, the eventual arrival of a story's natural audience, etc. the A story may in fact now be 'better', in those attention-deserving qualities that the ranking/decay function is trying to detect.)

As long as the quantity of good submissions varies consistently depending on time of day, there will be times when more good submissions are competing for a fixed amount of frontpage time. Thus it's unlikely you could eliminate time-of-day effects.

I like your second suggestion though.

Posted at 9:04am Eastern it would appear.

I just posted something about the impact we humans are having on planet earth. It was a documentary I listened to on the BBC world service in the wee small hours 3 days ago during my customary insomnia and about which I've not been able to stop thinking. I'd love it to spark a discussion but it looks like it's dropping into oblivion

This is probably not the best way to try to hijack this thread. If it was interesting and/or warranted discussion I'm sure folks would have started discussing it.

I'll concede your first point but I don't think your second always holds true.

I've seen posts of the same link with the same title only a few hours apart get dramatically different traction (5 vs 285 up votes). There are tricks to try and game the system but it also involves more luck than most people seem to think. That said, posts that are truly insightful and anything from PG will make it to the top, it seems, every time.

This is the third or fourth article I've read on HN about the best time to submit content on HN.

Fair warning: the next time I read this topic here, I am going to write a comment about the best time to submit comments on articles about the best time to submit articles.

This is basically true of all US oriented news sites. Also, generally monday and tuesday morning are ahead of the rest of the week. The best time to try to get something to 'go viral' is about 10 - 11am EST tuesday morning.

I'd be interested to see the analysis with the day of week included.

If the 7am-noon trend is caused by the social factor of EST to PST working coming in to work, it'll likely be mon-fri, with spikes on mon and thu.

That is the usual pattern for sites that people check when they come in to work.

I did a similar post about a year ago, though the focus was on other things than best time to post: http://foundation.logilogi.org/2011/5/9/the-best-time-to-pos...

It would be interesting to include day of week as well. I have read that Sunday is a good time for example, but taking into account viewer traffic, time on front page etc, (as the OP does of course) that might well be different.

What's the deal with EST or PDT? Can't you Americans just use some GMT-n so we can all convert it? Despite most stories being American-based or focussed here, the world does not center around you. Yeah it does on your maps - it's confusing, I know - but no the world is not actually about you.

Hi. I'm the author and I live in France. I used EST for 2 reasons.

1/ Most HN readers can relate to that easily.

2/ GMT doesn't account for daylight savings time.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact