@drl21, Yeah, I'm not really sure how they could effectively deal with the problem, but it exists. It seems like the main points would be: 1) Letting people know there has been a change affecting the game somehow (i.e. expansion, release on Steam, etc) 2) Allowing the GLS prediction to change even well after a year or two
I still don't agree with awarding the full DKP based upon the prediction alone though. Say there is some release that fuels a significant shift in sales after the stock has been delisted, you can't take the money back that you've already paid out. A worse scenario would be predictions vastly overshooting actual sales, and it's only confirmed after the stock has been delisted (i.e. the speculation was completely wrong).
All that is definitely true, but it would be impractical to keep every listing open for years and years on end on the off chance that any given game will be one of the few with an a-typical sales pattern. As you said, these things are hard to predict, and that serves as a double edged sword, so it were. (I hate that analogy...anyone who knows how to use a sword would never be dumb enough to cut themselves with it.... but I digress)
Anyway, my point is that the vast majority of the time the SE is fairly accurate. But there's no denying that there are examples where SE can't do the job, because things do sometimes happen later in a games life. That's why I think there needs to be some sort of system where under special circumstances a game can be RE listed. Although that would pose a complication in and of itself....notifying people that a game has been relisted, giving people their shares back, etc. Or maybe go through the whole process of price determination a second time? I don't know.
Or maybe it could be similar to how we can sometimes bid on the sales of games for specific months. What if even a year and a half after a PC game has been delisted, with the news of an expansion we could bid on how many units the expansion will add to the sales of the original, via boxed sets etc. Not actually bidding on the sale of the expansion mind you, but on the effect that it will have on the sales of the original.
@drl21, I'm a mathematics major so that stuff unfortunately leaks out every now and then.
I think that's pretty much the main concern as well. Take Diablo II for instance, it's seemed to tail off and continue at what would presumably be a fairly slow by steady rate, but when Diablo III was announced it rocketed back into the top sellers list. The Sims also had a very jumpy sales distribution, which roughly correlated with each of it's billion expansion packs. These kinds of things are hard to predict unless you have insider information of what will be released in the future. I also think that the introduction of old classic titles on Steam and the just announced Good Old Games will have a significant impact on older titles.
When the reward for predicting GLS roughly correlates with the actual GLS, delisting will work by virtue of reflecting reality, but in the cases where something changes the sales quite a bit later the payout doesn't really work in reinforcing the goal of the SimExchange. Hence I don't feel like delisting is fair because of situations like I described above.
Grinding out the numbers will be a pain in the ass, assuming I can get enough historical data, even with something like R, so I'll leave it to the SimExchange staff if they want to comment on their accuracy (that's also part of the reason I left it to you -- there was a remote chance you'd want to grind through all those numbers).
It seems like a hard problem to correct. In the real world markets operate on currency that has significant value -- it costs you something of great worth to make the wrong call about a company and companies can go under entirely. There is no such threat at all on the SimExchange, which makes it far easiest to artificially inflate GLS predictions.
Think back to Diablo 2 again, what do you think the SimExchange would have predicted for it's sales? Most games are done almost all their sales within a year, and the last sales numbers before now for Diablo 2 were about 4.5 million. I don't think the SimExchange would have produced good results.
If you look at The Sims 3 discussions here, people can't even agree on what the last known Sims 2 sales numbers were let alone predict what the current sales could be. One person claims 50 million, another person 6.3 million, someone claims another random number -- that seems insane and is, but if you look back at the sales figures from the first year or two there would have been complete consensus on GLS.
Even the relatively stable Assassin's Creed predictions will probably quite a bit off in the end given the current prediction has settled near the last known figures, and the eventual release of Assassin's Creed 2. Maybe it will change, maybe not -- the point being there's quite a few titles that the SimExchange seems ill equipped to predict.
Woa!...Standard deviation....normal distribution....I'm having Quantitative Analysis flashbacks here. Crap....now that I think about it, I start Quantitative Analysis 2 in the fall.....I wonder if i can use the SE for homework....haha
Anyway, you make great points. Especially about how people tend to estimate what peoples average estimation will be rather then what the game will actually sell. However, even that estimation is affected by sales trends, as if a game becomes very popular, one would estimate that peoples estimations would increase, so essentially your doing the same thing. In general, the longer a game is listed, the smaller the standard deviation from the mean the sample will become, baring drastic changes in the market.
So say a game sells for 5 years. I game delisted 1 year after it comes out will be less accurate then a game that gets delisted 2 or 3 years after. However, I would say that 2-3 years after a game comes out, the market will have (generally) stabilized enough, and sales will be fallowing a predictable pattern, that a game delisted at that point in time will probable be 95% accurate. I can not support this with numbers, as I don't have any, this is just what I would imagine would be true based on experience. So if you have numbers to prove me wrong, by all means post them, I won't take offense.
My mane concern is, if a game get's delisted, and then something drastic happens that does change the market (game suddenly gets a cult fallowing and sales start going back up for example) is there a process on SE by which a game can get RElisted? There might be some occasions where that could be appropriate.
BTW, this is an awesome debate. That's why I like SE....intelligent people.
@drl21, Being better than analysts is one thing and being correct most of the time is another. What I'm saying is that there's typically a fairly large error with GLS prediction, not that they're worse than analyst predictions.
When I say delisting a game is unfair, I mean that delisting assumes that the GLS prediction is correct, though it almost always will have a relatively wide margin of error. The last article I read on the SimExchange quoted an error of 16%, which is hardly what I'd call correct. It's more of a ballpark estimate like most estimates of this kind.
As an aside, what are you basing your assessment of the SimExchange's accuracy on? You're saying that we're pretty close 90% time then what is standard deviation of the error assuming the GLS prediction error roughly follows a normal distribution?
When we're talking about inference, we typically want to know with at least 95% confidence that a particular hypothesis is correct. Say our hypothesis is that the SimExchange's GLS predictions within 1 year are within 10% of the actual final value, can you prove to me that this is true with 95% confidence?
Getting back to the topic at hand, predictive markets are useful when no better method exists. The most common critique is that there is a tendency for those participating in predictive markets to abstract their predictions. Instead of trying to predict say what the GLS will end up being, people will try to predict what the average prediction of the GLS will be -- or in essence people will attempt to game the market. That's why awarding correct predictions is important.
Saying that predictive markets are a decent way of predicting the GLS of a game is fine. But claiming that this same prediction should be how we measure the accuracy of our predictions is just flat out wrong. Delisting a GLS prediction before the GLS is known with a high degree of confidence is tandem mount to awarding speculative investment, attempting the predict the prediction instead of the actual GLS.
That being said, the methodology behind the shorter term measures seems fine since they're awarding DKP on the basis of an actual measure.
"There's a decent chance that we'll be completely wrong about what the GLS will end up being. "
Ah, but that's where your wrong. The whole point of SE is that statistically, we as a large group, are far far FAR more accurate and consistent in predicting the life time sales a game then any analyst in on the market. Cool, huh? Seriosuly though, it's true. In fact, game analysts will often use the data and numbers that we settle on as part of their determination. So while we may be off slightly, 90% of the time were pretty close. So once we settle on a number that everyone can agree on, theres almost no point in waiting, because the chances of us being off are fairly small.
I actually discovered the SE when I read an article detailing all of that. I thaught that sounded awsome, so I joined.
@apujanata, I don't really get that -- say everyone eventually settles this stock to say 3000 DKP, who's to say it really will reach that point? There's a decent chance that we'll be completely wrong about what the GLS will end up being.
It seems somewhat unfair, or maybe cheap, given the Metacritic score guessing is pretty hard and fast -- whatever score it reaches it reaches; no if, and, or but. I can see how de-listing the average games would be reasonable since they have fairly quick right tailed distributions, but some games like Diablo have less predictable distributions hence it's far more likely that we'll just get it wrong.
I see how it's pragmatic in a sense, but it seems inelegant.
I believe that lots of simEx stock will get delisted about 1-2 years after release, unless there are still lots of activity in that stock (market still couldn't agree on the right price). However, don't worry about the actual delisted date, since we will always refer to Lifetime #, not current #.
Even if, say, by 2 years after release Diablo 3 only get 5 Million sales, if all of us agree that it will reach 7 Million eventually (another 8 years), it will get delisted at 700 DKP pricepoint (last price point when delisted).
While that's true, and while I apologize if my terminology was misleading or incorrect, Blizzard is a well known developer and Diablo 3 will be disc based. Therefore, based on your own argument, my comments are still legitimate with reference to this situation. Regarding the market that Diablo 3 will enter, the PC game market is shrinking.
Question: How long does GLS last? It took Starcraft ten years to almost hit 10 million units sold, and if my break down way below is right, it will have taken Diablo II at least 8 years to have almost reached 10 million units.
I don't know about you, but I do not see myself frequenting this site a decade from now.
4
Yeah, I'm not really sure how they could effectively deal with the problem, but it exists. It seems like the main points would be:
1) Letting people know there has been a change affecting the game somehow (i.e. expansion, release on Steam, etc)
2) Allowing the GLS prediction to change even well after a year or two
I still don't agree with awarding the full DKP based upon the prediction alone though. Say there is some release that fuels a significant shift in sales after the stock has been delisted, you can't take the money back that you've already paid out. A worse scenario would be predictions vastly overshooting actual sales, and it's only confirmed after the stock has been delisted (i.e. the speculation was completely wrong).