Attitude to market abuse


#1

So there is clearly some market abuse going on with COLIQ and so on, especially in the last couple days. I was just wondering if the devs had plans to combat this going forward.

In my experience, actively policing it never works and instead the mechanics have to be made “abuse-proof”… Honestly I don’t have a good idea of how to fix COLIQ, but I’d be interested to know what ideas are out there.


#2

The initial idea behind the COLIQ command is to allow new players to get into the game and to make errors. Once they are stuck we want them to give the possibility to start over.

Right now the system is exploitable and we know that. Restricting the COLIQs is on my immediate todo list.

Here is how I would try to minimize the problem right now: The times between COLIQs will be restricted on a user level. The first COLIQ is immediately possible (ie a user has misclicked in the company setup), the second COLIQ is only possible after 3 days (ie a user has sent his ships into nowhere with no fuel left), the one after that will be possible after 21 days (a user has decided that he wants to venture into another industry) and finally any COLIQ after that will have a cooldown of 60 days.

That won’t solve all possible exploits but will keep the market distortions to a minimum.

What do you think?


#3

Also keep in mind that this kind of abuse only really makes sense in combination with another account. If that account is a classic multi-account (same user, different accounts), we will be able to automatically detect and police most of these cases. If the accounts belong to different users, it becomes a lot harder to automatically detect patterns of misuse. But it’s on our list of things to look into before an actual release.


#4

I think that some sort of “reset” button has to be easily reachable, otherwise it will be a huge turnoff for new players. Plenty of people would need to coliq more than once to get the hang of stuff.

Perhaps a better solution is to allow new players up to a certain amount in special currency, that can only be used in a “government store”, which sells

  1. completed base item
  2. “completed buildings + a small amount of starting ingredients for that building” items

which obviously then cannot be sold on the open market.

Also, I’m really dubious about you automatically detecting alts; if someone wants to hide it, they can do so, and automatic detection often needlessly harms people in the same household playing the game, which is just irritating.

Edit: We are currently discussing this in the global chat, and seem to have decided that (as in my suggestion) the problem isn’t really COLIQ but how start-up is currently implemented.


#5

Just making starting items untradeable would fix most of the colliq issues.

Starting packages also need to be balanced significantly, and everyone receiving the same “starting currency” would definitely be a step in the right direction.


#6

There is still the issue of starting credits, even if the items are untradable. Intead, design could reduce the amount of materials needed to build all pioneer-powered structures, balanced by an increase the area they use - basically make the starting package a thing of low value relative to what can be made with a few days of production.

I’m strongly in favor of everyone getting the same starting package. Even though this is just a test run, I’m surprised at how tilted I am that I haven’t been able to accumulate enough wealth in a month to equal that which I could have had on day one with a different starting package.


#7

It would always be a pattern detection. For example, the majority of the resources of one company are in some way transferred to a particular other one, then the first one is deleted and both were run from the same IP…in that case things are pretty obvious. Even if it isn’t an actual multi (because the other player is the cheater’s neighbor or nephew or dog) :slight_smile:

That. This is a trading sim. We at least have to give people money and we can’t restrict how this money is spent without considerably hampering the game. This riddle is very hard to solve and as said above, I think eventually the only way around these problems will be strong anti-cheat detection algos in combination with a game design that “simply” makes it less attractive to misuse COLIQ.


#8

Pattern-based detection of cheating is a losing game for developers. It takes an ever-increasing amount of effort to do, and the developers will always lose–I’ve seen it happen in literally dozens of games over more than 25 years of online gaming, including in environments that had multiple full-time developers devoted solely to trying to stop abuse.

It’s easy to catch blatantly obvious cheaters, but those aren’t the ones you need to worry about. The tools that catch the blatant cheaters also tend to generate a whole lot of false positives, and they’re trivial to circumvent by any cheater who puts even a tiny amount of effort. They’re also easy for malicious actors to abuse to frame legitimate players. Only the most clumsy and incompetent of cheaters will use alt accounts from the same IP address and directly funnel resources to their “main” account (and I’m pretty sure that’s already been happening!).

The “riddle” has been solved time and time again in many other games and environments. There are many possible solutions, and which one(s) you pick depend on what your vision is for the game, what you want the new player experience to be like, how much staff/developer time you want to commit, etc. We, as players, can’t make those decisions for the developers.

If you feel that you “have” to give people money, you can make all money traceable and add mechanisms for unwinding fraudulent transactions. If you want the effect of giving people money without actually giving them money, there are ways to do that too. If you make the choice of handing out a big bag of cash (and that is a design choice, not something intrinsic to trading simulators), then the rest of the game needs to have systems in place to limit or eliminate the degree to which that can be abused. If you feel any kind of “reset” mechanism (COLIQ) is too abuse-prone or it’s not worth the effort to prevent abuse, then alter the other systems in the game so a reset mechanism isn’t needed (e.g., fewer company failure modes, a better new player experience, ways to recover from early mistakes). If there’s too much of an advantage to having alt accounts, increase the barrier to entry. And so on.

The lowest-cost approach to cheating in many systems is to clearly define what’s not allowed, tell the community that it will not be tolerated, and make it clear that actions are being taken. When blatant cheaters get to brag about what they’ve done, that sends the message that cheating is just fine.


#9

A possible solution would be to add a flag variable to the starting resources, marking them as resources that can’t be sold, kinda like banks marking stolen money.

The biggest problem right now is COLIQing is hurting the PP1 manufacturers as they can be easily undercut, which is hindering the economy.


#10

I totally agree with this statement and would really caution the dev team against going down that path.

Edit: What has kindof worked in another alpha game I used to play (Haven and Hearth, if you’re interested) is to say that if a bug or flaw is found by a player, it should be reported: at that point, the devs can say if it can or cannot be used until patched. If something is not reported, using it for gain/griefing is bannable if discovered, at the discretion of the developers. Discovering a bug and using it once in an economically irrelevant way for testing should be encouraged (sometimes with some cosmetic reward). This requires a public bug tracker, but it adds incentives for the players to test the game in a sensible way and allows the devs to clearly state what specific behaviours are not permitted (temporarily or otherwise).


#11

As you might have seen in the Release Notes we added a cooldown as a first measure against COLIQ abuse. The cooldown times are:

  • First COLIQ: immediately
  • Second: 3 days
  • Third: 21 days
  • every subsequent: 60 days

We intent to include a manual check by a moderator later.


#12

This is good for now, but I still don’t think a cooldown is a good long-term solution; I guess you can look at churn metrics and see if its correlated with hitting one of the cooldowns, to see if I’m wrong.