Originally posted by marinakatombrecced.
As far as i'm concerned, there are certain open and shut cases that are taking too long to determin. Some players charts are enough to ban them. If a player is losing to 1000 rated players when they first join, while using patzer moves like Fools mate, then suddenly display Grandmaster understanding of the the Kings gambit 3 weeks later and then rise to ...[text shortened]... e investigated for months when this method would reduce their thinking time significantly....
Originally posted by XanthosNZXanthos: I dimly remember that, but it is not the same as an actual amnesty. However there really isn't much point in debating it as the time has passed.
If you remember correctly there was a point before the Game Mods were introduced where everyone was required to affirm their agreement to the ToS statement on engine use. I vaguely remember a screen with two buttons one with a brain on it.
Mainakatonb
As far as i'm concerned, there are certain open and shut cases that are taking too long to ...[text shortened]... ts are enough to ban them. [ reduced ] ... would reduce their thinking time significantly....
[/i]
Mainakatomb: Well yes sometimes it's blatant. But really this is one of those crimes that really does only hurt the perpetrator. I lost to Ironman31 6 times and I lost nothing. He lost his soul in the process. The reason it takes ages to check out whether a player is cheating or not is because the site does not want to get it wrong. This requires deep analysis. You can't ban a player soley because they've started winning - that could just prove to be admirable application. Players get banned under 3b at a rate of 1 or 2 a month, most cheaters will either get more sophisticated or give up.
Just a thought but if someone got so clever at cheating it looks to the games mods that they aren't cheating then to us all it's as if a human were playing. They'd probably have to work at it longer than it takes us to work out our moves anyway, but get a higher rating, lucky them. Just plain bad luck if you don't use an engine but play like one...
Originally posted by DeepThoughtAre you saying that you believe a player can increase their engine match up rate by 40-60% in three weeks? I have played this game long enough to know how hard it is to improve. I could play one game at a time, spending 1 hour on each move, and i still wouldn't beat a top 20 player. My engine match up rate might increase a little if i did this, i'd certainly make less mistakes, but it would be a small increase. NO ONE goes from losing with fools mate repeatedly for a month to hitting a consistent 2000 rating. It is impossible, simple as that!
Xanthos: I dimly remember that, but it is not the same as an actual amnesty. However there really isn't much point in debating it as the time has passed.
Mainakatomb: Well yes sometimes it's blatant. But really this is one of those crimes that really does only hurt the perpetrator. I lost to Ironman31 6 times and [b]I lost nothing. He ...[text shortened]... rating, lucky them. Just plain bad luck if you don't use an engine but play like one...[/b]
Originally posted by marinakatombWhat I'm saying is that there's a difference between suspicion and proof. In Domeninc0789's case it was blatant, in others it's not clear. If you can give me an example, then send it to the games mods. Although, I have to admit to being a patzer when it comes to spotting engine use...
Are you saying that you believe a player can increase their engine match up rate by 40-60% in three weeks? I have played this game long enough to know how hard it is to improve. I could play one game at a time, spending 1 hour on each move, and i still wouldn't beat a top 20 player. My engine match up rate might increase a little if i did this, i'd cert ...[text shortened]... repeatedly for a month to hitting a consistent 2000 rating. It is impossible, simple as that!
btw who gave me a rec - I wanted to stay on 42!
Actually i'm responsible for the banning of about 12 players to date. I've had players removed even before the mods were in place, but most of them were for multiple accounts. I don't hesitate to report people once i feel they're cheating. I'll send you my latest one by PM (wouldn't want to get banned for accusing people in the forum....). You'll notice the first few games are rather poor, but all of a sudden...
In my opinion, there are certain shortfalls and challenges in the game mod process.
Identifying cheaters is not really the biggest challenge. Advanced tools and methodologies have made identifying obvious culprits a great deal easier than when the team first kicked off. There are still some cases which require substantial in-depth analysis, which can be very time consuming, but these are the exception rather than the rule.
The biggest challenge the process faces is co-ordination of game mod members and the prioritization of cases. This is a group of volunteers... so if you can imagine a group of headless chickens running amok, you wont be far from the truth. The hardest (and most tiresome) aspect is getting everyone to focus and work on any one specific case, reach a concensus, and then convince the admins to take action.
On the other hand, who wants a trigger-happy cheat police with players getting banned on the whim of a single member of the team? The burdon of proof is set very high. The litmus test of overwhelming evidence, beyond a reasonable doubt is very much in place. A concensus opinion from game mods and site admins is required in every case.
My opinion, which I am fortunately entitled to, is that the process is much too slow, and that it depends too heavily on subjective analysis. It also depends too heavily on the availability and personal circumstances of each game mod, and individual contributions into the evidence "pool". For example, I've recently come back from a 3 week holiday. Another mod has been substantialy absent from modding for the past 2 months due to real world work pressures. Members cannot always make the personal contribution that is required in every case.
Personally, I would like to see far more reliance on objective statistical analysis. The major benefit is that analysis can be automated, and there is no subjective "is this a match or not" approach. Any evaluation is based on a large sample of games, and the results compared against an appropriate control set. The outcome is a statistic that give the probability of a suspect being wrongly identified as an engine user.
In a recent banning, that statistic was 1 in 10 billion! Stats like that should be a no brainer. However, it still requires input from all four moderators to reach a decision in such a case.
The only problem with the statistical method is that it is not "official". It is a method of analysis employed by just one member of the game mod team, one person's "contribution" to the evidence.
The effect of this: The number of suspects currently determined to be cheating by one or more game mods is 19. One game mod has reached a positive conclusion in 2 cases. Two game mods have reached positive conclusions in 3 cases. The remaining game mod has made a positive conclusion in 17 cases.
Now, eventually many of these cases will attain 4 positives and the players in question will be banned. But it is a slow and tedious process. Adding additional game mods is not necesarily the answer here. In fact, adding more game mods, given the current M.O. would only slow the process down further.
The best I can hope for over time is that an objective/statitistical method will become the standard approach to identifying systematic engine use. It should become a tool, rather than being viewed as the subjective output of one member of the team. Not only will this save a great deal of time, but will allow more game modding effort to be channelled into studying the less obvious cases where engine use may be only partial or selective.
I don't think I'm compromising the team in any way by expressing my views here. For all its shortcoming, what we are currently doing is far better than doing nothing. And the progress and achievements over the past year are tangible. There is no text book on game modding, and when the outcome of our efforts result in the permanent banning of members, it is far, far better to err on the side of caution.
Ultimately, there will be internal debate and a resolution of such issues. While we want to keep discussion of indivdual cases out of the forums, broader discussions on the game mod process is fair game, and should be encouraged. Don't expect us to be too forthcoming on details though!
What we have to balance is the frustration of seeing obvious cheats seemingly getting away with it, versus the right of everyone to be treated fairly and every case to be examined thoroughly.
While I've reached the point where I'm convinced that there is a very reliable technological solution to identifying systematic cheating, I'm not sure everyone is equally convinced.
I'm still optimistic, though, that the effectiveness and efficiency of the Game Mod process will continue to improve.
Originally posted by marinakatombWell getting an openings book could account for some it (no more Qf3/6 in openings) after about 20 or so games) - but I do see what you mean...
Actually i'm responsible for the banning of about 12 players to date. I've had players removed even before the mods were in place, but most of them were for multiple accounts. I don't hesitate to report people once i feel they're cheating. I'll send you my latest one by PM (wouldn't want to get banned for accusing people in the forum....). You'll notice the first few games are rather poor, but all of a sudden...