4NCL Online

[JustinHorton]

As a bit of a diversion, I wonder who is the strongest player in history who would struggle to provide an answer to the question “best response to 1.d4?” (in a negative context, not as a way of choosing between several good option).

How about Garry Kasparov and Bobby Fischer, both of whom lost faith in their weapon of choice when it came to the crunch

[Roger Lancaster]

It's now clear that switching windows during an online game arouses suspicion, so, for example, you shouldn't check your email during a game.

Well, yes, but problem is that 4NCL procedures mean - obviously without intent - that team captains are more or less bound to do exactly this.

[John Swain]

It's now clear that switching windows during an online game arouses suspicion, so, for example, you shouldn't check your email during a game.

Well, yes, but problem is that 4NCL procedures mean - obviously without intent - that team captains are more or less bound to do exactly this.

That is certainly a problem, I agree.

We don't know how often it's possible to switch windows in a game before it becomes significant for lichess anti-cheating software (and I don't imagine lichess would be keen to divulge this).

[Roger Lancaster]

It's now clear that switching windows during an online game arouses suspicion, so, for example, you shouldn't check your email during a game.

Well, yes, but problem is that 4NCL procedures mean - obviously without intent - that team captains are more or less bound to do exactly this.

That is certainly a problem, I agree.

We don't know how often it's possible to switch windows in a game before it becomes significant for lichess anti-cheating software (and I don't imagine lichess would be keen to divulge this).

A little like driving along a road where, although you suspect there's a speed limit, no-one will tell you what it is?

[Adam Raoof]

Well, yes, but problem is that 4NCL procedures mean - obviously without intent - that team captains are more or less bound to do exactly this.

That is certainly a problem, I agree.

We don't know how often it's possible to switch windows in a game before it becomes significant for lichess anti-cheating software (and I don't imagine lichess would be keen to divulge this).

A little like driving along a road where, although you suspect there's a speed limit, no-one will tell you what it is?

Or where there are lane restrictions, and you get a fine for choosing the wider lane even though the narrow lane looks too tight and has scary bollards?

Oh wait, that's Islington....

[MartinCarpenter]

It's now clear that switching windows during an online game arouses suspicion, so, for example, you shouldn't check your email during a game.

Well, yes, but problem is that 4NCL procedures mean - obviously without intent - that team captains are more or less bound to do exactly this.

Well since - we presume - LiChess aren't hacking webcams, all they can actually check is whether the browser window containing the chess game is actually active at a given time.

So you could dodge it by reading email on a different device, or just having multiple tiled windows open on your main desktop at once with the email window on the side. Honestly its such a weak deterrence to cheating that I'd find it slightly depressing if its effective in that sense.

Random move timings are much harder to avoid in many home settings!

[Nick Burrows]

I have had 4 lichess tabs open and have intermittently checked the other games in my match without (i presume) being flagged yet.

[Richard Bates]

Hi Justin

My perception is also that substantially fewer than 15% of people cheat at blitz which is all I play and not all that often. Part of the reason that users might think that 15% is way too high is that cheats do get weeded out (often fairly quickly) and therefore play fewer games than the average, honest person - so 15% of users equates to a lot less than 15% of games. However, I have no idea whether it's improved or not nor do I know how chess.com or lichess compares with the ICC of 2014-15.

Your second question is harder. At the time there seemed to be very few titled cheats. I'm struggling to think of more than one, an FM who beat a super-GM 9.5-2.5 in 1 minute chess (which I watched live with incredulity) and beat Magnus Carlsen, playing under a new identity with a rating of 2200, in 25 moves with black. And still didn't get banned. As for the rest, the majority had identities unknown to me so I can't vouch for their OTB strength whereas their ICC ratings could be anywhere upwards of 1700, at a time where ICC ratings were about 100 points lower than FIDE (for 15 minute chess). I do recall playing 65 15 minute games prior to restarting OTB and, when my rating reached about 1900 on the ICC, almost one game in three was against an engine-user...which was what dragged me into this whole sorry affair in the first place.

As a general thing, branching somewhat away from the 4ncl topic, it’s funny how people’s experiences are different. I think I can genuinely say, having played probably tens of thousands of games of online blitz chess across platforms since the late 90s (originally ICC and then mainly latterly chess24) that there isn’t a single game EVER where I have suspected my opponent of computer cheating.

As this is clearly at odds with what others claim is widespread blatant cheating what is the explanation for this? That I somehow fall into some sweet spot of playing mode (almost exclusively pure 5-min) and or rating (doesn’t really make sense*?) that running into cheats is very rare? The “15% of users doesn’t equate to 15% of games” thing? Or just that if you are playing online purely for enjoyment and don’t think too much about your opponent as opposed to the board, the whole thing just passes you by? There are enough mistakes in my own play to explain/rationalise defeats without thinking there is something rotten on the other side of the board. And generally winning far more games than I lose means the whole thing is generally enjoyable and therefore worthwhile.

On the other hand I suppose you do occasionally wonder how some people (including opponents) seem to have obtained quite such high ratings when they don’t often seem that strong. So they must be cheating against everyone else!

(there is of course the possibility in pool based random pairings where people are paired in close relation to strength - either by design eg. the ICC 5-min pool or choice (players seeking out opponents of equivalent strength) that for some the purpose of “cheating” is simply to give their rating an artificial boost to play opponents (without cheating) of a higher calibre than their normal strength would justify. So they cheat, play some stronger players, lose, fall back, rinse repeat. Perhaps even getting genuinely stronger for the experience along the way)

*what is more strange is the accusations of cheating a relatively weak levels when the supposed cheats (if cheating exclusively) must be using computers from circa 1985!

[Michael Farthing]

Richard,

There is the famous story (amended here to make it on topic) that casts light on your thoughts:

A chess player arrived at the gates of heaven and was welcomd by St Peter. "Do you have chess congresses in Heaven?" he enquired. "Oh indeed yes" replied St Peter. "What are they like?" enquired the chessplayer. "Well, I'm not sure how to answer", replied St Peter, "What are they like on Earth?"


"Oh the congesses were superb! The games were hard fought and I used to spend four hours on tenterhooks. My opponents were always friendly and it was great to go through the games afterwards and see how much our thinking had in common. The organisation was wonderful and clearly hours of work went into setting them up so well. And I particularly enjoyed playing juniors - such enthusiasm and challenging, attacking play!"

"I see", said St Peter, "Well I think you'll find it's much the same here"

Later in the day St Peter greeted a second chessplayer and the conversation followed a similar course. The second player's description of Earth congresses was slightly different, however:

"Pretty ropey", said the chessplayer, "There were children not concentrating and distracting everyone - eating food and wriggling in their chairs. There was tons of cheating - people going to the loo all the time and obviously using an engine - and fixing games to manipulate their grades and win prizes in sections they shouldn't be playing in. And the arbiters just played patience on their computers and never did their job properly"

"I see", said St Peter, "Well I think you'll find it's much the same here"

[Richard Bates]

Very good

[Richard Bates]

A perhaps interesting little side story to this (which hopefully I haven't told before) in the world of over the board chess - which either demonstrates by naivety, or (I think more likely) the danger of overinterpreting computer 'evidence' particularly from a small sample (although as I don't know how the 'cheating' algorithms work this may be a an irrelevance).

It goes without saying, I suppose, that I have never suspected any over the board opponent of external assistance either. However several years ago I was playing in the early rounds of an international tournament. My opponent was a fairly "promising" junior/young player, but nothing remarkable in rating (either at the time, in the rest of the tournament, or since) and I was clearly the seeded player. I got a slightly disappointing position with white and never really recovered. At one point I thought I was on the verge of getting back into the game with a few chances against an open king, but it wasn't quite enough and eventually I succumbed. I didn't really feel I played very well and got the result I deserved.

The next day I bumped into the seed on the adjacent board, who said something along the lines of "you were really unlucky yesterday. Your opponent played like a computer!". I don't think this was on the basis of any post game engine analysis, just casual observation of the game as a bystander.

Anyway I didn't think anything more of it, until a few months later I stumbled across a video that somebody had for some reason made of the game (I think at the time). It wasn't a video made to make any allegations or anything, but at the start it had a throwaway line that "there was a very high level of computer correlation in the game". So for fun, I ran the game through a computer and it was true! My opponent's level of correlation was something like 95+%! (not sure of the precise algorithm). However, and here's the rub - it turned out that the level of my own play was very high as well (over 90%)! So having thought I had lost to a weaker player having played badly, it turned out it was probably one of the most computer accurate games I had ever played. And I still lost!

In reality I think, it was just one of those positions/games where by coincidence the most natural moves throughout (where not forced) are also the good ones, and where it also turns out that there is quite a narrow band of 'natural' (and therefore best) moves available to choose from.

And a postscript - I played the same opponent again a few years later. I played what I thought was a promising pawn sac in the opening and felt like I was going to win very quickly. However I just couldn't find a killing blow (obviously overestimated my position), and suddenly found myself grovelling for equality. I thought i'd done enough when he suddenly found a brilliant concept which I completely overlooked. Fortunately I managed to compose myself and hold on for the draw. However, as stated originally - there was nothing in historic rating, performances anything to suggest anything untoward. So if there was something suspicious (as stated I believe 100% not) then in a reversal of the statement in relation to online players - there was no cheating against any other players - except me!

[JustinHorton]

I used to play a fair amount of correspondence chess by server in the first few years of the present century, wwhen strong programs were already available to anybody that wanted them, but competitors were barred from using them. I played one game on the White side of the Slav with 5 a4 Bg4, which you don't see very much now (I think it's the only game I've sever played in that line) but which often leads to bizarre positions. I lost very quickly, not much more than a dozen moves, culminating in a winning queen sacrifice, and I wasn't at all pleased about it. Nor was I pleased when my complaint was rejected by the arbiter.

I looked at the game again a few years later and..... I couldn't work out what I'd been bothered about. There was only one really difficult move my opponent had found, and that was thematic enough, the arbiter had been in the right and I'd been in the wrong.

After that, I became a bit more wary of assuming that my first impressions were anything to go by.

[John McKenna]

As so far 100% of the times (approx 7 or 8 times) I have reported someone for suspected cheating lichess has then banned them for engine use.

That sounds suspicious - please say when you get a failed denunciation or when you find someone you suspect gets banned before you report them.

Why? Some cheating is really obvious in a way that everything correlates. To illustrate one example: there was a user with a rapid rating around 2400 on lichess, which I think would correspond to roughly 210 - 230 ECF, who was at the same time asking on the forums what was the best response to 1. d4. From context it was clear he had no idea of standard openings, but in games he was playing a variety of response to 1. d4 at IM+ level.

Certainly there are times when a game has seemed a bit suspicious, but not enough to report, and later on I have noticed that the user was banned.

Thanks, but why - when you've more or less answered what I asked for - did you ask, "Why?"

Cheating online has become similar to the real word virus - test, track and trace, as best you can, but those with the virus are hard to discern at first, if ever. And the authorities are behind the curve, constantly, therefore doing too little too late.

From time to time, however, they'll trumpet a far from shining instance of some heavy-handed deed they've managed to do such as making an example of every nth man.

Edit - we know the answer to what's Alphazero's best v. 1.d4 (not sure it has done the diligence v. 1.b4!?)

[Pete Heaven]

If I can just come back on a couple of Richard's points, briefly for once.

It's very common for players for all strengths not to consider that their opponents are cheating. Personally I envy such people whose lack of a suspicious nature holds up most of the time.

Maybe my original, counter-intuitive point of high-rated players not cheating wasn't particularly well made. By high-rated players I am not referring to players with high online ratings, usually playing under a pseudonym, with anonymity their only friend. I am referring to those who have high ratings in real life and are playing Div 1 of the 4NCL online, where everybody knows your name, as the song goes. As I mentioned previously, the number of titled cheats accused on the ICC when I was involved was almost zero.

[Roger Lancaster]

Post by Roger Lancaster » Wed May 06, 2020 9:42 pm

"It's a common phenomenon that the dissatisfied shout louder than the satisfied. Just to check that there isn't a silent majority here, may I ask whether there is anyone on this forum who has total confidence in the anti-cheating mechanisms of Chess.com and Lichess? If there is, and they are fearful of getting abuse if they say so publicly, they are welcome to PM me and I'll undertake to give them [4NCL management board members excepted] total anonymity. I'll report back on the number of messages, omitting names, in a day or two".

I posted the above at 3.39 yesterday [Tuesday] afternoon. By way of interim report, I've received two PMs. One, from a gentleman who rarely posts on this forum, indicated lack of total confidence in Chess.com or Lichess, mainly due to lack of transparency. The other, from a regular poster here, didn't express a clear view about the two platforms but indicated his total confidence in the integrity of any actions taken as a result of 4NCL's own investigations. [For what it's worth, my question related only to the platforms because I, too, would differentiate between them and 4NCL].

So far, no-one has come forward to express total confidence in the platforms. However, this survey is unscientific in the extreme and, moreover, there's plenty of time for their defenders to show up - and, indeed, I urge them to do so. I'll post again once a little more time has elapsed.

By way of final update, I have received just one further PM. This expressed total confidence in Chess.com while explaining that the person in question had little or no experience of Lichess and was therefore unable to express an opinion. To summarise, Lichess have just issued a statement where they express "full confidence in our systems and methods" so it may be significant that, when members of this forum are asked to affirm "total confidence" [which I would take to have the same meaning as "full confidence"] in Lichess, there seems to be a marked reluctance to do so.