pull down to refresh

Want to know what it takes to be a serious stacker here on SN? You might not want to miss this one...

Stacker News Insights ๐Ÿ‘€

6 weeks ago I promised in the Stacker Saloon that I would review a big batch of SN post history to:
  • draw some conclusions of trends to help understand better what makes a successful post.
  • explore how the SN community is evolving.
  • salute & celebrate what Stacker News has become.
  • share insights for territory founders to improve engagement.
Today I am finally getting my finger out. This is the 1st of a 3 part series, to chunk-up some of the post data available on Stacker News in a responsible manner. The other two parts will drop later this week if there's enough interest from Stackers. Today we're going to focus on the overriding numbers in aggregate across dates, times & post types.
Above is a word-map from the post titles of >1k sat posts here on SN. In the coming days, we may also explore some territory data, to see which areas of the site are making a success of things. Finally in Part 3 we will credit those Stackers who are putting in the leg work, with the aim of helping others find who we should all 'subscribe to'. Encourage others to step-up their stacking game and post original content.

Notes

  • Data is taken from stacker.news/top/posts/forever?by=sats
  • Posts with over 1k sats total (8,657 posts) are included.
  • This initial snapshot was taken at 22:45 EST on Monday 26th February.
  • It only captures headline post data based upon homepage information visible to everyone - no SQL database access.
  • Satsraisers have been removed from chart data for now, to prevent skewing the totals generated from posts.
  • Want to be tagged in future editions/parts? Or do you have suggestions for new ~charts to follow-up with? Let's see suggestions in the comments below...
Check-out some similar discussions & graphs from the @sn crew:

Satistics โšก

By Month

In 2024, with not even 2 full months completed, there are already way more >1k sat posts on Stacker News (1,816) than the entire year of 2022 combined (1,009). 2024 has started off with a bang๐Ÿ’ฅ
Still, there are less >10k sat posts in 2024 (83) than 2023 (402), but every indication suggests that record could be broken this year. Let's hope this post can help get us there.
The average sats per post (3.8k) is impressively nearly-matching 2023 figures (4.0k), despite each satoshi being far more valuable today than last year. Reminder: this is not the average for all of SN posts but just the average for posts above 1k sats total.
The number of comments per post is intriguing also, since there isn't a huge difference between years. The curve seems to be flattening between months, with 2024 having an average of 16.1 comments per 1k sat post. On track to be between 2022 (17.2) & 2023 (16.6).

By Day of Week

The above graphs are suggesting we have ourselves some sticky users, with consistent sats every day of the week in 2024. Stacker News is definitely gamifying our stacking habits, damn you @k00b, @kr, @ek & team. 2023 shows an opportunity on Sundays for posting, with fewer high-value posts in number at that time. Saturdays and Mondays stackers are slacking.
It looks like posting on Wednesdays, Thursdays & Sundays is a great way to get an additional boost of sats when sharing engaging content on the site. For discussions & comments, Mondays and Fridays are the standout candidates.

By Time Of Day

It used to be the case that posting early morning for European hours would generate the most impact on the site. For 2023 & 2024, posting between 9am and noon Eastern Time is CLEARLY optimal, leading to 4,523 sats per post on average, vs 3,978 sats on average for other times overall. That's an extra 14% for posts with 1k or more sats. For posts with 10k or more sats, it seems that stackers have generated an extra 24.8% by simply posting between 9am and noon ET ๐Ÿ˜….
The same is true for comments. 21.1 average comments when posting during 9am to noon ET, vs 16.5 comments on average any other time of day (27% up).

70.2% of sats earned from posts (over 1k sats total) have come without sharing any link whatsoever. Just from pure chat & community alone. For 2024 that figure is 66.7%. Stacker News is becoming the source of bitcoiner content, not just a news aggregator no more.
Sharing a link results in less sats overall (3,365) when compared with no links (5,713). The gap has shrunk significantly since post descriptions became possible when posting SN links. Comments are drastically higher on posts with no links (22.1) vs those with (8.1). So again if you want more sats and engagement, take the time to write something original.

Domain>1k PostsAvg SatsAvg Comments
--none--51315,56522.2
twitter.com2913,03611.1
github.com2733,0846.8
youtube.com2583,2487.2
substack.com1092,5337.9
bitcoinmagazine.com1053,3186.7
nobsbitcoin.com862,4783.7
bitcoinops.org682,3280.7
medium.com593,7284.2
lists.linuxfoundation.org563,5615.4
thebitcoinbugle.com472,1437.6
zerohedge.com362,3397.5
coindesk.com312,1899.3
cointelegraph.com222,2427.8
nitter.net183,04010.1
cnbc.com183,1998.7
primal.net179,34716.1
thrillerbitcoin.com172,4869.5
stephanlivera.com162,3402.6
forbes.com162,0206.2
mempool.space155,74513.5
blog.lopp.net152,9627.0
blog.mutinywallet.com137,56014.4
Not much can be derived from the domains shared each month, but I'm still a fan of these sat-ghetti graphs myself. It should be noted similar domains have been merged (twitter.com with x.com) wherever possible. Primal had a recent flurry of sats thanks to @Darthcoin's Zeus anniversary post here on SN. Since Q2 2023, people have been surprisingly focused on mempool.space, be it sharing specific mining blocks, transaction fees or specific transactions like this one from @BlueSlime with increasing popularity.

Stacker Summary ๐Ÿ˜ด

1. 2024 for Stacker News looks like being a record-breaking year to sat-earn.

2. Thursdays are now the best day (16% extra sats) for posting. Wednesdays or Sundays are the next best options.

3. Posting between 9am and noon (Eastern) may net you as much as 14-25% extra sats.

4. Posting between 9-12 EST will also see 27% more comments on posts.


Preview of Territories - Part 2 ๐Ÿค 

Core Territories - ~bitcoin, ~nostr, ~meta, ~tech

Surprising to me, ~bitcoin (4.1k sats per post) is not much more popular than any of the other core territories for sats per post. ~tech is definitely trailing the Big 3 (in ~bitcoin, ~nostr, ~meta). On average making less than 3k sats per post vs 4k across the others, a 25% deficit.
We all likely knew this already but if you wish to stir a debate, the place to post it in is ~meta (with 25 comments per 1k post, vs 16 in ~bitcoin, 13 in ~tech and 10 in ~nostr. That's why I chose ~meta on this occasion over the increasingly popular ~charts.
More on that territory and others in the next edition...
Maybe the most interesting thing to me about this is how, as the btc price has exploded, stackers have become more generous, not less. I know this is true of me, as well. I attribute this to two things, basically -- as NGU, people in this space get excited and feel good, alive, like the slog and the obsession is paying off. People feeling good are generally kind and spread the goodness around -- which is a good lesson for life, I think -- and it's true here, too.
The second reason is perhaps a controversial hypothesis: stacking has little to do with price. People are not price sensitive in their bones. The cheap-asses going around zapping one sat are not doing it because they need every fraction of a cent to buy a micro-hamburger to keep from starving. I don't even think they're doing it because they imagine an outsized future. Even if the price goes up tenfold in the next decade, which I am confident it will not, your 100 sat zap -- which would be generous, by the standards of most stackers -- is now worth sixty cents. You're telling me some latent multi-millionaire is clutching his pennies this tightly? I just don't buy it.
So my hypothesis is that we should expect more of the same -- NGU leading to stackers making it rain.
reply
Word. I got a rude shock the other day when I went back to use Simple Bitcoin n Yzer and received a mere 1 sat in rewards. SNโ€™s relatively generous zapping culture made me forget how tough it is outside SN to stack sats
reply
142 sats \ 0 replies \ @k00b 28 Feb
That's what I felt like I saw last cycle. People spend more money when they have more of money.
reply
I have a completely contrary hypothesis to yours.
As the price (purchasing power) rises, stackers are acting more in line with the incentive structure of the site's rewards mechanism. That's exactly what we'd predict from rational economic agents.
reply
How could we disambiguate between our hypotheses? It strikes me that yours has not accorded well w/ reality, as demonstrated by your repeated reminders that people's normal zapping tendencies are un-economic.
reply
They're getting less uneconomic as the stakes increase though. We refer to that as "saliency" in experimental economics. Oftentimes experiments fail to find results because the design didn't create a big enough payoff spread between the different actions.
If I could earn another kilosat by zapping better, I'm more inclined to make the effort to do so when that kilosat is worth more.
reply
Ah -- that seems plausible. I might incorporate that into something like:
  1. the psych factors of all the exuberant feels lead to people zapping more
  2. this gives them the chance to observe economic payoff matrices
  3. increased value makes those payoffs salient enough to care about
If that's the generative model, it's not obvious how to realistically pry apart #1 from #2 and #3. You could run an experiment but it would be unnatural-seeming.
reply
Numbers 2 and 3 lead researchers to identify this problem. A bunch of experiments were putting people in unusual situation that nonetheless had very distinct theoretical predictions. However, due to funding constraints, the payoff differences were only like a nickel.
The participants quite rationally didn't find it worth their time to think through what was going on nor did they notice a material difference in the payouts they were getting so they couldn't learn anything anyway.
reply
what does NGU mean?
reply
Numero go up (price)
reply
Nรฃo sei
I'm eager for the ensuing discussions about what kind of rewards experiments should be tried based on this great work you've done.
Something that came to mind while reading through your post, is that we may want to have rewards over multiple timeframes: i.e. weekly rewards and monthly rewards, in addition to daily. I think that might help smooth out some of the morning congestion.
Also, we had discussed some form of peak-load pricing, which also might smooth out some of the morning posts.
reply
Peak-load pricing, as in making it more expensive to post something during prime hours on the platform? It reminds me of electronic road pricing, a scheme that charges road users money for using expressways during peak hours in my country haha
reply
Same idea. This came up before territories launched and we were trying to think through how to reduce the spammyness of the main feed.
My simple idea was to have posting fees = the number of posts in the past hour. Back then the posting fee was only one sat.
To update the idea, we'd probably want each territory owner to be able to set their own peak-pricing mechanism. Now it would help address the issue of low territory revenue, in addition to cutting down on spam.
reply
73 sats \ 1 reply \ @k00b 28 Feb
I think about your peak pricing algo all time.
reply
Now that we have different territory owners, it could eventually be a useful policy lever for them.
reply
deleted by author
reply
He highlights 9-12 EST several times in the post.
reply
ahh thanks! You were too quick off the mark! I saw that and deleted it quickly but you'd already replied ๐Ÿ˜† All very interesting, I had been suspecting this, as I'm in Europe currently and a lot of activity happens overnight for us.
reply
Yeah, I think anchoring to the Texas time zone might be leading to some gamification that works against our European and Asian Stackers.
I've seen sites randomize their equivalent of the reward deadline, in order to give everyone a fair shake.
reply
Sometimes feels like you could post something really awesome morning Europe time and it's tumbleweed time
reply
If I'm interpreting this post correctly, I think a lot of European morning posts get overlooked, because a flood of American morning posts bury them.
I guess the issue is that people show up in the morning with a post in mind, If they looked for content to read first, then those European posts would get more attention.
reply
Do the morning European posts get overlooked and zapped less as there are rewards for zapping top posts and top comments early? So, when there are more stackers online are they going for the more recent posts, rather than older posts? that's the problem with rewards, it nudges and changes behaviour. I mean, that's the whole point of it. There are always consequences
Would love to be able to schedule posts too ๐Ÿ˜
Whilst there are no doubt tweaks that can be made, I donโ€™t think the issue is as pronounced as my post makes out.
Weโ€™re talking about a small uplift of 14%. 4,523 sats during 9-12 EST, vs 3,978 sats on average other times. 24% higher specifically for 9-12 vs 3-6am EST (3,654 sats).
I still think genuine proof of work rises to the top.
I posted at just about the worse time possible last night and it still showed up in the top 5-10 posts by mid morning the next day.
a long waiting post:) ๐Ÿฅ
  1. Thursdays are now the best day (16% extra sats) for posting. Wednesdays or Sundays are the next best options.
It matched my observations. Also Fridays and Mondays are the worst.
  1. Posting between 9-12 EST will also see 27% more comments on posts.
it also depends on the OP.
  1. Sharing something original WITHOUT a link may result in 69.7% more sats. 44% was the additional sats earned last year.
Yes - stickers should always post their original work here instead of post links - it's no longer in fiat social farming attention, you got to stack real sats here!
reply
Great post. You have already done a ton of work, but I wonder if you have the tools to adjust for zapping habits? During SN's first year or two I think top posts garnered far less sats. Perhaps a pre and post @nemo adjustment? ๐Ÿ˜€
reply
Graciasig.
2022 showed something contrary to that. There were certainly far fewer posts and sats thrown around total (there are more cowboy hats these days) but the average zap, at least on a per 1k post basis, was higher.
Donโ€™t unfortunately have access to zap data, only in the aggregate post totals. And sats from comments on each post.
reply
Interesting. It's great that you have accessed this data. Maybe I'm wrong. It's just anecdotal, but top posts seemed to garner fewer sats when I first joined SN.
reply
I may have a deeper look into that.
It certainly was more challenging to get to 1k sats back in 2022, but once you did it seems you were heavily rewarded.
reply
90 sats \ 1 reply \ @siggy47 28 Feb
Some more off the top of my head, but it seemed (counterintuitively) like stackers were zapping more as the fiat price of bitcoin rose in 2023.
reply
100%. That was definitely visible in charts for all 3:
  1. number of 1k sat posts
  2. total sats from posts
  3. total sats from commentsโ€ฆ
reply
During SN's first year or two I think top posts garnered far less sats. Perhaps a pre and post @nemo adjustment?
it could be more zappers now ( especially some big ones) and also the rewards? ๐Ÿค”
reply
Maybe. I can't prove it, but I felt a distinct culture change in late summer 2023 when it came to zapping. I always wondered if I was right.
reply
Developers should introduce a new OP code: SATS_RETURN :)
reply
Sharing something without a link almost zaps nearly 70% more!!! That's one big takeaway for Stackers here. If you really wanna make it here, as with the SN community now more mature, you need to open some kind boggling threads rather just posting a link from some common mass media. Twitter (x) and other social platforms can be escaped as only sometimes because sometimes they have worth sharing ideas.
That's a great post... Thank you so much.....
reply
Fantastic work, my friend!
reply
Really impressive, and thank you for sharing all this work with the community.
Can I ask what was your process doing so? Which tools did you use? How did you design those clean and tidy charts? Did you access SN data via APIs or somehow elsewhere?
reply
I could certainly improve the workflow and automate a lot more if I wanted a live dashboard or to do this regularly, but without any API this is what was used to get the job done for now:
  1. visit stacker.news front page as I mentioned above under 'notes'
  2. tap 'more' button to see more than 21 posts at a time. repeat 390 times
  3. right-click on page and save .html file locally
  4. created a python script that recursively spits out and renders all the front-end post content (title, author, territories, sats, cowboy hat, comments etc) into a csv
  5. copy that CSV into a spreadsheet with new columns with formulas to sort and categorise content better
  6. create a bunch of pivot tables & graphs in said spreadsheet
  7. screenshot those graphs & upload to SN
If I was to make this a continued process, I would automate Steps 2 and 5 - making sure the data got pushed to Tableau or something as capable. For a one-off exercise like this however, it's more than sufficient.
And next time I want to click a button 390+ times to get an updated view, I can simply paste the data over the existing spreadsheet data and all the pivot tables automatically update.
reply
Try this to automate step 2. For $12/y the PRO version could be really useful and can save you a lot of time (and clicks!)
reply
220 sats \ 0 replies \ @kr 28 Feb
great insights, thanks for putting this together!
reply
Very good post. You answered many questions, GRACIAS!
reply
152 sats \ 0 replies \ @mo 28 Feb
Just... wow! Thanks ๐Ÿ™
Looking forward for next two remaining parts
reply
Some excellent work and breakdowns here - thank you!
reply
Congratulations! You concretised the best practices to stack sats. But since I already post every day, I doubt this will impact my behaviour haha.
I feel that Mondays and Fridays may gravitate towards discussion because of Meme Monday and Fun Fact Friday respectively. These are evergreens that pull in a crowd, week in week out
reply
152 sats \ 0 replies \ @bshell 28 Feb
Commenting is key to getting OP To return. This data is beautiful. Nice work!
reply
Epic! Thanks for sharing ๐Ÿ‘Œ
reply
Thursdays are now the best day (16% extra sats) for posting
hehehe more people want to get into SNL "top of the week". I don't care too much "when" I should post. I just post when somebody (a shitcoiner or gov agent) piss me off.
Great analysis !
reply
Trigger happy DC.
I agree - I also donโ€™t like sitting on stuff when itโ€™s ready to go. Just found it an interesting datapoint when there seems to be more generosity / traction.
reply
Thank you for making this available
reply
bookmarked to read later. congrats and thanks
reply
Very detailed, THANKS. <
reply
Good info, many surprises.
reply
๐Ÿ‘๐Ÿ‘๐Ÿ‘
reply