The Dustbin

[paulmcneil]

It actually was three consecutive attacks. The fourth attack did not cause any checks. Doing the math of 6% x 6% X*8% you get the probability of about 1 in 3500. However I don’t think you took into account all three double drops were in range of a general in combat which gives the additional +1 to avoid the double drop. This significantly increases the odds I believe to well over 1 in 10,000. However just because an event is low probability does not mean it can’t happen. Someone has to win the lottery and I won the triple double drop one last week. This was the problem with playing dkalenda, these kind of low probability events happened nearly every game with him.

Harvey

Actually someone doesn't have to win the Lottery. Not unless there is a maximum pot cut off in EURO Millions anyway...

[paulmcneil]

Division B

paulmcneil (Polish) beat Nijis (Fatimid) 62:41

[snooky51]

Whatever the fallout from this scandal is, the remedy is going to have to serve Slitherine's interests and the competitive multiplayer community's interests. A workable solution for Slitherine will probably need to be inexpensive to implement and operate, in addition to restoring player confidence in their PBEM system. From reading this thread my understanding is that we want something that is transparent in how it operates, simple to understand and utilize, and most importantly effective in policing cheaters.

My understanding is the "anti-cheating" system in place as of now is not transparent, nor does it appear effective in policing cheaters. The idea of having the number of restarts placed in the in-game chat or in the AAR, seems to be inexpensiveness, transparent and simple to understand. However, the previous few pages suggest we are at a crossroads at how to then utilize that number of restarts to effectively police cheaters while keeping the peace in the community. I am in agreement with whoever objected on grounds that having the community i.e. the individual players police ourselves will probably not work and degenerate into a tool for undermining other player's credibility.

I am not aware of Slitherine's resources or how much time/money they reasonably should invest in preventing cheaters from gaining a leg-up in tournaments awarding $10 voucher prizes. However, on the tournament side if the tournament organizers create a rules-regime that accounts for number of restarts (idk like -1 point of your points for the number of restarts, so if you win but restarted twice, you only score 2 points for the win). Then those restart numbers are just placed in the score-reporting forum by the winner a (if he doesn't report his own restarts his opponent can always comment and so on), no different than the score of the game. That can just get recorded similar to the 3-1 scoring. If someone has restarts but for legitimate reasons (internet, etc.) its upon them to proffer evidence that their internet was down or something (I am pretty confident that information is available through one's ISP) and present that to the tournament organizers. From their they can adjudicate as they always have.

I know this does put more work on already overburdened people, but atleast keeps it to a minimum until someone present their evidence that they had legitimate restarts. We could even have a sub-forum that tracks these numbers for individuals, as someone suggested, to give people more information to work with.

Whatever is done will need to be a compromise, and somewhere someone will have a new job regarding the policing of cheaters.

As someone who found out about this publisher and this game just 3 months ago. I don't expect to have much insight, and if that is reflected in what I stated above, feel free to ignore it.

[Schweetness101]

I suspect that the technical difficulty of changing over the PBEM system to save every move as they happen, rather than every turn, would probably involve a greater development cost than Slitherine is willing to invest in. It's I think a 3 year old game, by a relatively small publisher, in a pretty niche genre (that I love, don't get me wrong...), where the great majority of players are singleplayer only, so a significant edit to the multiplayer system at this point is probably out of the question.

I'm guessing that the only technical solution they would be willing to implement is simply informing both players each turn if a reload has occurred, but then, as a few people have pointed out, that would result in a large number of poisonous and false accusations of cheating occurring, which might have a net more negative effect than just doing nothing. This is probably the internal argument Slitherine is having right now.

Perhaps a more effective variation on the reporting reloads would be to add some randomization into the reloading reporting. IE sometimes it sends a 'too many reloads' type message at 3 reloads, sometimes at 5, etc...in the same game, thus being a bit more intimidating for a potential cheater, because they aren't sure with any given reload if it will be the one to get them reported.

If there was also an open channel to Slitherine to make reporting suspicious luck easier, that could also serve to intimidate cheaters as well. That is, a player who feels they have been cheated in a given game can report it to Slitherine, and Slitherine can inform them of whether there were any reloads that game. Just knowing that that can happen more readily should also serve to intimidate cheaters.

The very existence of threads like this will also likely intimidate potential re-loaders from trying it again for some time, so at least there's that.

[Schweetness101]

Division B

Schweetness101 - Byzantine 551-578 AD defeats paulmcneil – Parthian 250 BC-225 AD with Seleucid 166-125 BC allies 65-41

I went mostly cavalry (mix of light and non light horse archers, byzantine lancers) with lots of foot lights, some massed archers, and a small corps of superior dismounted lancers. The Parthians deployed a solid block of very superior but slow Argyraspids, cataphracts, and pikes on their left, and a large number of unarmored horse archers and light horse on their right. The average and unarmored light and non-light horse archers of the Parthians could not compete in the ranged fire with a combination of massed archers and armored byzantine ranged horse, so the Parthian right of cavalry was rapidly routed before the slower units on their Seleucid left could catch up. When they did finally catch up, they did plenty of damage to the Byzantine lancers and dismounted lancers, and closed the score a bit, but it was not enough to change the outcome at that point.

gg to paul

[Schweetness101]

Division B

lydianed (Moors + Byzantines) defeated Schweetness101 (Byzantines) 62:49.

Oh Damn

sorry Cromlechi, I tried

javelin horse...javelin horse everywhere

[Ludendorf]

Fine work. Congratulations on the win.

[devoncop]

I suspect that the technical difficulty of changing over the PBEM system to save every move as they happen, rather than every turn, would probably involve a greater development cost than Slitherine is willing to invest in. It's I think a 3 year old game, by a relatively small publisher, in a pretty niche genre (that I love, don't get me wrong...), where the great majority of players are singleplayer only, so a significant edit to the multiplayer system at this point is probably out of the question.

I'm guessing that the only technical solution they would be willing to implement is simply informing both players each turn if a reload has occurred, but then, as a few people have pointed out, that would result in a large number of poisonous and false accusations of cheating occurring, which might have a net more negative effect than just doing nothing. This is probably the internal argument Slitherine is having right now.

Perhaps a more effective variation on the reporting reloads would be to add some randomization into the reloading reporting. IE sometimes it sends a 'too many reloads' type message at 3 reloads, sometimes at 5, etc...in the same game, thus being a bit more intimidating for a potential cheater, because they aren't sure with any given reload if it will be the one to get them reported.

If there was also an open channel to Slitherine to make reporting suspicious luck easier, that could also serve to intimidate cheaters as well. That is, a player who feels they have been cheated in a given game can report it to Slitherine, and Slitherine can inform them of whether there were any reloads that game. Just knowing that that can happen more readily should also serve to intimidate cheaters.

The very existence of threads like this will also likely intimidate potential re-loaders from trying it again for some time, so at least there's that.

All good points except the last sentence.

In fact at this moment in time there is less incentive for a cheater not to cheat than at any previous moment as the inadequacy of the current meAsures have been made glaringly obvious and unless and until the system is improved that position sadly won't change.

[Schweetness101]

I suspect that the technical difficulty...

All good points except the last sentence.

In fact at this moment in time there is less incentive for a cheater not to cheat than at any previous moment as the inadequacy of the current meAsures have been made glaringly obvious and unless and until the system is improved that position sadly won't change.

oh yeah lol

I guess I was thinking in the sense that people might be thinking that a general review of reloading is coming and that at least for now they'll back off of cheating that, although conversely they may be thinking '255 reloads without getting banned! time to cheat to the max while the getting is good!' although I hope not...

[paulmcneil]

Div B

for Gdod our challenge is up pw = norman

[paulmcneil]

Div B

for Macedonczyk our challenge is up pw = byzantine

[blobka]

The insane amount of spotted reloads suggests, that you not gonna get the desirable outcome that easily, it can turn for the worse as well, reloading once-twice can be considered as abusing the flaw 'on paper' but pretty useless in reality. I had in my experience at least once such a story with, when I did realize it can be misused as it doesn't save the seeds for the rolls onto the server and the appropriate turn sequence 'on the fly', but in my case it turned for the worse, I recall.. .

PS: Division C-D not an A but an average player's opinion.

[NikiforosFokas]

Thanks guys for your kind words...

[nyczar]

Division B

nyczar - Spartan 221-146 BC with Roman 280-220 BC allies defeats paulmcneil - Etruscan 330–280 BC with Samnite 355-272 BC allies 44%-16%

[MikeMarchant]

The insane amount of spotted reloads suggests, that you not gonna get the desirable outcome that easily, it can turn for the worse as well, reloading once-twice can be considered as abusing the flaw 'on paper' but pretty useless in reality. I had in my experience at least once such a story with, when I did realize it can be misused as it doesn't save the seeds for the rolls onto the server and the appropriate turn sequence 'on the fly', but in my case it turned for the worse, I recall.. .

PS: Division C-D not an A but an average player's opinion.

Everyone's opinion is valid, Blobka, no matter which division you're in.

Says a guy in division C whose cheating skills are clearly as poor as his playing skills.


Best Wishes

Mike

[bomber23]

Division D

bomber23 (Carthage Hannibal in Italy) beat Bluefin (Syracuse) 51-26

[SimonLancaster]

Human nature suggests that an individual is more likely to attempt an act if it is viewed as easy than if it is viewed as difficult. If making cheating more difficult does not deter cheating, why would so much effort be put into making cheating difficult?


Best Wishes

Mike

In this current situation I really don’t see much difference between easy and difficult. You cheat 5 times. Next game you cheat 8 times. What is an acceptable number of reloads for one opponent might not be for another.

I do think some kind of reload counter would be a start.

Overall, though, server side calculations have to be the way forward. That might not come until the next game..

If by 'this current situation' you are talking about cheating in general, then whether it is easy or difficult to cheat is absolutely funadmental to the problem.


Best Wishes

Mike

Right, but at the moment there is no difference between easy and difficult. You just cheat. Whether it is a few times or 225 times.

What would make a difference is some kind of clear turn counter and then obviously, it would be quite difficult to make up a story that accounts for 225 reloads....... it is a sad story to be honest.

[Barrold713]

Well done Richard. Congratulations

BDH

[Nijis]

Division B

Nijis (Fatimid Egyptian 978-1073 AD) defeats nyczar (Frankish 751-887 AD with Croatian 625-849 AD allies) 60 to 58.

The score says it all. This was extraordinarily close. The Fatimids took an early lead, the Franks came surging back, and than the Egyptians finally inched ahead again.

(3-1)

[IainMcNeil]

We're currently collecting data on all games to find out what "normal" behaviour is to be able to set the thresholds that will prevent cheating.

This particular incident would have been averted if more time had been spent reviewing it but the system was designed years ago and has not really been reviewed as there is so much else to do so could definitely be better, so more man power is not a long term solution. After review we've noticed the system is better at picking up long term general abuse than spikes in cheating behaviour. When we have the full picture we'll work out how better to detect the cheating but we have first to collect and analyze that data.

We are very worried about false positives as this will kill the hobby quicker than failing to detect cheating. There is nothing worse than being accused of cheating when you didn't. We have probably been too reluctant to call people in the past but this has to be handled carefully and we're not sure that handling it in public is the best solution.

Also please leave any discussions about random numbers out of this. We will be ignoring them as they are not relevant to this and just confuse the discussion. In fact I'm going to ask that any discussion of it just be deleted. With the number of games in progress and the number of calculations done per turn, almost any possible outcome is going to happen. 1 in a million chances happen every day, multiple times a day. You'll even end up with one person getting two 1 in a million chances and maybe even in a row. When you have millions of people playing games this is what happens and there is no RNG that can avoid it, unless it cheats!