A team of bot-hunting players made the news last month when they took down eight ridiculously expensive supercarriers being controlled by bots, exposing just how big the scale of the problem is. The EVE security team responded with a ban wave hitting over 1,800 bot accounts in January and promises that they are “coming for the bots,” but one expert admitted in a recent interview that the war on bots may never be won. So just how difficult is it to tackle botting in EVE Online, and what could CCP do to improve things?
In this edition of EVE Evolved, I look at the difficulties in detecting and shutting down botters, how extensive botting may be in nullsec, and some things developers might have to do in order to solve the problem.
It’s hard to pin exact figures on the extent of bot use in any online game, moreso for EVE Online as it can be difficult to even distinguish between the bots and legitimate players. Some players will farm PvE content for hours each day, use multiple accounts at the same time, reject attempts to communicate, and run to safety at the first sign of danger — that doesn’t mean they’re bots. You can’t just look for sequences of repetitive actions because EVE PvE is inherently repetitive in nature, and most bots operate in otherwise empty star systems where the risk of being spotted and reported is low.
CCP has access to other tools and data in the battle against the bots, but the bot developers have some serious tools at their disposal too. Modern bots can read the game state directly from memory, inject code into the game client to execute commands without using the game UI, and even read and write packets directly to bypass the client entirely. Even if CCP manages to detect those types of bots, there are other kinds that don’t mess with the client at all. They can visually process the screen to locate elements such as buttons or NPCs to click on, simulate human-like mouse movements using curves and random noise, and even chat in fake channels to throw off GMs. Identifying bots is definitely not a trivial matter.
The recent news about a group of supercarrier bots being killed helped to expose the scale of the botting problem in EVE‘s nullsec regions, as someone was clearly making enough ISK to warrant spending over 200 billion ISK on the ships. The reality is that a supercarrier can pump out over 100-150 million ISK per hour farming an endless stream of nullsec anomalies, so they pay for themselves after about 200 hours of continuous operation. The group of bots were thoroughly tested by players and repeatedly made mistakes no player would make, getting caught in the same obvious log-off trap over and over again.
When news of the destruction of the eight bots was made public, several players independently confirmed that they’d seen the bots operating in those same star systems and had reported them multiple times. A look at the players’ killboards showed they’d been farming with supercarriers since at least 2013, so they may have been botting with impunity for years despite being reported to CCP. These cases also represent only the very upper end of botting in nullsec, and bot hunters routinely run across groups of faction battleships or Navy Issue Vexors that are obviously automated.
Discussions about botting in nullsec recently blew up in the EVE community after a video was leaked of Imperium alliance leader The Mittani aggressively threatening his alliance members for reporting friendly bots. Player Markus Hayabusa asked whether he should report any allied bots he finds to the alliance diplomats or send a report to CCP, prompting The Mittani to launch into a tirade suggesting that reporting innocent players would get them banned. CCP Guard challenged this assertion in a message to the community shortly after, encouraging players to report suspected bots and stating that “[b]ot reports from players are not acted upon unless we’re really certain and other evidence lines up.”
In a recent reddit post, player TikkTokk made a pretty convincing case for the fact that EVE‘s botting problem is in fact a PvE game design problem. Many forms of PvE in EVE are extremely automatable, and it doesn’t have to be that way. Alliance-owned nullsec systems can be upgraded to provide endless streams of “cosmic anomaly” combat sites for farming ISK, for example, and that’s part of the problem. Players can literally warp to the next cosmic anomaly in a list, launch drones, and kill wave after wave of NPC ships without even moving from the warp-in point.
If developers want to make botting difficult, they need to make the gameplay itself more difficult to automate. Create procedural environments with navigation hazards players have to manually pilot through, emergent situations the player has to respond to, branching scenarios that require the player to make good decisions on the fly in order to reap the largest rewards. The new Resource Wars gameplay has some hints of this, but this design philosophy would need to permeate all PvE activities if it’s to have an effect on botting. This would be a monumental development undertaking, to the point that I don’t even think it would be feasible.
CCP Guard downplayed the impact of botting in a recent interview with PC GamesN when he said that “a few botnets running will not put a great dent in the EVE economy,” and yet there’s no doubt in my mind that the existence of bots cheapens the accomplishments of legitimate players. It’s also difficult to shake the implication that a significant amount of the wealth from botting is going straight into RMT trades and driving up the cost of PLEX, which has an indirect impact on players and on CCP’s bottom line.
Detecting bots is not a trivial matter, and it’s not made any easier by the political greed of some nullsec alliances that advocate turning a blind eye to cheating allies. The best way to fight bots would of course be to make the gameplay itself difficult to automate, but that may not even be feasible in a 15-year-old game packed with existing PvE game mechanics and there are also market bots and spam bots to deal with. CCP is at least getting harder on botters and increasing the punishment, but this may be one fight that no developer can ultimately win.