Big move in this one today.... I don't see any news on it.

These huge bounces are getting annoying.
I repeat my suggestions from the 8th.

@Bradley Baker - might these bouces be caused by new users voting? like me, for example =)
As for the topic itself - i believe that Facebook definetely has a greater potential than Myspace... However my intuition tells me - not this month...

I think the huge bounces are due to day traders who throw in $500,000 at a time to sway the % up or down and then hope that others see the huge sway and vote accordingly in the meantime dropping the immediate cashout value of those who are already invested. That's what I do at least.

Looking into it.

I'm not sure how that can be accomplished.

What's the difference between trying to create a daytrading wave to ride and feeling the community is too far to one side on the issue, jumping on it, and then when the community shifts towards the other direction, getting back off? While Aaron's expressed motivations are counter to the idea of predicting outcomes, the described behavior could be completely justified. Other than through first-person posts it's not clear how rationale can reliably be devined.

Part of the problem might be that the total money in the system has been growing rapidly (assumption, but warranted I think). What made sense for tuning when nearly everyone was near the starting amount of S$100,000 might not make sense with a few people in the S$900k-S$1.5mil range (or my former position (yesterday) around S$2.5mil) and a dozen more at S$500k+. Finding a mechanism that works for both and will continue to work as the largest portfolios continue to grow is decidedly non-trivial. We do need SOME responsiveness or every prediction just sits near 50% no matter how clear the eventual resolution becomes. Finding that happy medium sweetspot.... not so easy. Especially given that it likely changes over time.

Going back to this prediction, rather than general site discussion for a moment.... Does comScore have a regular release schedule? Do we have any indication that there will be updated information prior to the closing of this prediction? Without any such release we're still with the most recent information, which, of course, leaves MySpace on top, correct? So part of this prediction is whether or not new information will come out in the next 3(ish) days. And THEN whether or not facebook has passed.

@Bradley - what I'm talking about is a scenario where one user creates multiple accounts to attempt to "game" the system. We have alerts in place and are upgrading the sensitivity as I write this. We currently believe in a policy of going after the individual wrong-doer(s), rather than halting a bet and hurting everyone involved. It's a tough call, as these markets are very fluid (and, it's not always 100% cut and dry as to what is going on, but rest assured we are monitoring this very closely).

We will know quite soon who (if anyone) has attempted to "cheat" their way to some easy money. Slapping a bunch of money down at the last minute is totally fair game, however. Creating multiple aliases to achieve those goals is not, and grounds for significant penalties, and possible site banishment, loss of S$/status, etc. You'll see more soon on that in an update to our Terms and Conditions.

That's what I was talking about.

@Bradley and @Daniel have really hit the nail on the head in terms of the sweetspot tuning. It's a bit of an art form. I really like this idea: ""it would also be nice to be able to pare positions without completely cashing out." We had that in place on the site prior to our launch, and should be able to bring it back in the very near future as we finish off some Trading Page updates.

I know it was wrong but I just couldn't help myself. I wanted to see if it could be done (and to be honest, I wanted more $). I filled out a bug report afterwards explaining what I had done and have been waiting for the sky to fall since then..

Here's the lowdown:

I don't know how closely you've been following this prediction but last week I created a couple of extra accounts and gamed it. I simply created a few dummy accounts and purchased the stock to push the cashout yield higher. I then cashed out with my main account (followed by the dummy accounts) and then bought again with the main account (and then the dummy accounts). I'm not sure but I think I did this 3 times and within 10-20 minutes I had more than doubled my $.

The jumps in the odds were insane.. Just by dropping a few hundred thousand on the table I could push the odds between 10% and 60%..

This is simple stock market manipulation and -obviously, shouldn't be allowed. The thing is, I didn't really need to make bets with the dummy accounts. I could also have just posted 5-6 messages that could sway other bidders. I can imagine that the multiple betting can be detected and penalized, but how are you going to fix the second issue? (It's not easy, stock markets still haven't found a surefire way to detect this..)

A more logarithmic (?!) algorithm is a must in my opinion.. I didn't understand how the concept of time can be incorporated into this though..

@Eric: Got it. Hadn't assumed you meant multiple-account manipulation (what Can K describes). I thought you were referring to wave creation/riding (what Aaron describes).

Thanks for both the continued responsiveness and what appears to be continuing efforts to a) improve the site, and b) listen to user feedback.

It'll hopefully make your lives easier once we have the user forum. I'm amazed that you all can effectively glean useful feedback from what start as prediction-specific comments threads and then occasionally drift into site feedback and site feedback discussion, but it appears that you manage to. Must require a lot of reading through all kinds of comments.

While we've converted this (partially) into a feedback/suggestions thread, is there a way (either currently or planned) to get a list of the N most recent comments posted to any prediction thread? I'd certainly browse such a list and it would help find what could be otherwise obscure comments, making them more useful for all involved.

@ Can Koklu: Merely posting comments shouldn't have the type of concentrated effect that you describe. Either the comments are pointing out real news -- in which case it's reasonable for the market to react and all you're doing is helping that happen sooner (ie be more responsive in the correct direction) -- or you're posting false information -- in which case a) regular users will start to ignore you and/or warn off other, and/or b) you'll be sanctioned much the way people are in the real world. If false information becomes common place (a la pump and dump posts on Yahoo! Finance message boards or stock picking email spam) then it'll merely result in all such information being viewed as unreliable. People may use Google News, etc. to verify or they may ignore such posts, but either way the false information should have limited effect. Other than getting repeat offenders banned.

Multiple accounts working in concert, on the other hand, can easily move stocks. Especially if the amount of money available in a reasonable number is enough for a significant move. As the site population grows and individuals get less power vs. the community (which may be difficult when the top people have the port size of dozens of average people) this will, hopefully, become more difficult. And in any case should be fairly easily detectable (or at least difficult to make hard to detect).

@Bradley - regarding comments, what you'll see in about 3 weeks (fingers crossed) are notifications when someone replies to you, or notifications if someone comments on a suggestion that you created (you'll be able to toggle the notifications on/off if you find them annoying). I'm not sure I 100% follow your request, though. Do you mean a list of, say, the 5 most recent comments posted on any prediction? We have a couple of comments widgets on the site, but nothing quite like that. It could be a useful tool to be placed in everyone's portfolio (My Predictions link). If that's what you're proposing, I'll float that idea over to the crew.

Also, I want to let you all know a little bit about some of what's in development to whet your appetite. We're busily working on a "karma" system which will allow the community members to value others' contributions, and intersect this with your rank. I don't want to give too much away for now, but Can's point #2 shouldn't be too much of a concern when we roll that out. We'll eventually have a team blog, too, so we can fill you all in there about what's going on and not fill up our comments section with these types of tidbits. Our newsletters also will pick up some of the load when important new site features roll out.

Oh, and the first phase of the prize program is just a few weeks away.

By the way, you guys simply rock!

Very interesting, Bradley. We've gotten several requests for notifications, but your ideas are very intriguing indeed.

Regarding the original prediction, Industry Standard contributor Melissa Chang has analyzed international competition between MySpace and Facebook, and predicted a winner in Facebook vs. MySpace: The battle for global social network dominance

And Bradley, Josh here -- to answer your question about comScore, we have a login and are able to pull the data needed to judge this prediction. Can't disclose at this point obviously how it is looking as of this morning. comScore does do regular press releases announcing notable trends (which are free to everyone). It will be interesting to see if they do a release on this item as an update to their initial one.

Okay, some more details. In terms of international traffic, Facebook was at 100.7 million uniques in January, and MySpace was at 109.3 million.

According to comScore, as of Friday, Facebook is now at 109.2 and MySpace is at 116.6. The gap is still decreasing, but the incumbent remains in the lead.

The community was, as usual, correct. Thanks for your patience!