Why is the rep cap set at 200? Why not 2000?

Question

It's quite well documented that SO has a daily reputation cap of 200. Canonical and duplicate questions abound on the subject, and I understand the various aspects of the answers provided.

However, I'd like to know how the specific threshold of 200 was selected (why not 1000? why not 20?). Was it reached upon arbitrarily?

Or was it more methodical. Crucially, what was the reasoning behind this particular number?

Of course, this rep cap isn't the only moving part in what is a complex SO economy. Explaining it's magnitude independently of the magnitude of other moving parts in the economy would be perhaps incomplete. I suppose the 200 threshold is somehow dependent on how many other key interactions it can translate to (e.g. 20 answer upvotes, 40 question upvotes, etc). However, i) this is speculatory, and ii) then the question simply morphs into 'why was a rep cap chosen that exactly translates to 20 answer upvotes, why not 100 answer upvotes, etc'?

I'm a student of game economies and am highly interested in an involved, technical answer regarding the magnitude of these thresholds.

Answer 1

(tl;dr? Skip to the last three paragraphs.)

The reputation cap is one of the network's longest-running features, having been set at 200 since the month Stack Overflow launched, if not earlier. Records from back then are... spotty.

Background

Many features we take for granted today, like restricted editing, review queues, and association bonuses, didn't exist when the site was born. (On the other hand, closing was much crazier, with only one person needed to shut something down, so there is something to Glorfindel's answer.) Based on vague memories of the early days, observations of how the team operates, and general background, I'd say the rep cap made it into the system so early for two reasons:

1. Balance
   People are naturally going to think a lot about reputation scores. The underlying "game" of the site is designed to take advantage of that, for Pete's sake. But not everyone is going to be able to rack up massive scores, even if they're equally capable. Those pesky "jobs" and "real lives" get in the way. Having a cap reduces the amount to which simply spending time on the site correlates to rep, which ultimately helps maintain reputation's purpose as a coarse-grained measure of trust (both "how much trust should you put in this person's content" and "how much the system and community trust this person to behave").

2. User health
   I'm pretty sure I've seen something from Jeff Atwood about this, but I can't find it now. Instead, I'll quote cletus (source):

   It acts as a hint to limit yourself to a certain amount of time on the site each day. Of course you can stay as long as you like but the daily cap is a gentle hint that maybe you should be doing something else for awhile.

   This might be annoying in the short term, but keeps users coming back and being productive in the long run instead of getting addicted and burning out.

Honorable mention:
The rep cap is a failsafe against users gaining a boatload of reputation off of a single post that gets wildly popular for just a few hours/days. This, too, was more of an issue when the site was nascent, when more discussion-based, opinionated, and/or humorous posts (not to mention bug reports and feature requests) were permitted on the main site.

Data... kinda

To tie this all in with the actual magnitude question you asked... I again can't say for sure. I spent a lot of time trying to find the actual origin of the rep cap, and couldn't. The farthest I got is finding out that it originated as a UserVoice request back when Stack Overflow was in beta. For reference, see Bernard Dy's comment on Aug. 25, 2008 at UserVoice post fix this limit for voting, or a brief mention Jeff made in Podcast 019, which was released the following day:

we're currently recalibrating a bunch of things and I think we're about to piss off a bunch of our audience 'cos we're going to put in a whole bunch of reputation caps which they're not going to like!

One thing I do know is that many early site decisions (and, for that matter, some more recent ones) were made by gut feel. I strongly suspect that someone simply picked 200 as a number that felt not-too-high and not-too-low, and then when it seemed like things were going well, the value stuck.

This could be attributed to both the "if it ain't broken, don't fix it" factor, and users getting anchored to the starting value, such that any attempts to change it would have probably met with unhappiness. And it does seem like a nice middle ground, where you can get a few decently upvoted posts up in a day without "losing out" but not be encouraged to be reading SO 24/7.

One last bit of speculation: the cap was actually part of a response to complaints about the previous reputation control system (limiting people to just five or six votes per day), so it's possible that rep-related numbers from that era were factored in, at least qualitatively.

Answer 2

It's probably more related to what you can do with the reputation than what is necessary to reach the daily reputation limit. For instance, it takes 2.5 days (2 if you have the association bonus) to get access to the review queues (Meta Stack Exchange is an exception) which is one of the keystones of moderation on Stack Exchange. In addition, on beta sites, the same reputation level gives you close vote power.

If we would have a rep cap of 500, it might just take one good answer to a particularly hot network question and a user could access this 6-8 hours after they've joined a site. It's still possible but now it takes at least an extra day, after which the question usually receives less views (and the answer less drive-by upvotes).

Or, to quote an answer to the linked question:

I would not want someone who has been on the site for 2 days to suddenly have the ability to close questions or edit my posts, if he hasn't had time to understand the nature of the sites.