Borislav Ivanov: a Lilov's add-on

[Don]

Don, I think you and many others start from the wrong assumption that a player can not play a couple of hundred elos above his current rating.

I don't make that assumption. It probably happens in almost every tournament that someone plays 200 points above their rating and yet nobody singles them out for "persecution." I have played 200 points over my rating several times on my way to 1914 from being a 1500 player when I first started playing. And I did it in short time.

However, what this guy has done is not to have a good tournament but to have a phenomenal tournament. It's almost unheard of what he is doing. And then he had MORE of them.

That it is impossible to learn fast. But actually modern engine software makes learning and progressing much easier than ever before.

Ok, so you believe the guy is for real and that he has simply improved at a phenomenal rate. I'm going to play this back to you when he is caught.

training with top engines would mean that you will be able to see some pattern in their approach to the game that would facilitate learning enormously, in sharp distinction to engine software dating back a decade ago, when trying to learn with Fritz would be difficult basically because it would be close to impossible to discern a pattern in its moves, they were like on a random basis. With state-of-the-art engines of today one should be able to progress several times quicker than otherwise, not to mention training with books and a chessboard, where already energetically producing the moves would deprive you of some learning advantages.

Nobody questions that dedicated study combined with computers has some great benefits for improvement. When should be seeing hundreds of these guys in tournaments though if it is as wonderful as you say.

[Taner Altinsoy]

Hi Taner.
The 10-0 routs of Rybka and Houdini were obviously an emotional statement; but you can expect that from a person at whom all are pointing a finger shouting 'You, cheater!'. Imagine for a moment that the accusation comes as a complete surprise to him.

I think his earlier games show some very good positional understanding combined with making lots of mistakes. Getting rid of mistakes, improving tactically could warrant a sudden strength increase.
About playing weak in some tournaments and strong in others - but that is about true for all, look at some recent world-class tournaments, there are very often players significantly underperforming, and players significantly outperforming. And that might be valid not just for a single tournament, but for a whole stretch of events spanning an extended period of time.

The smoke is in and around Taksim Square, will there be fire?

Being a Turkish born in Plovdiv Bulgaria and who still have residence permit there I even sympathized with him and really wanted this miracle to be true. A lonely geek outsmarting chess GM's and such. But unfortunately it is very hard to believe in his story if you add everything up. For me convincing points (if I remember all) are:
- phenomenal playing strength (2600-2700) in many games and then playing at his own lvl (1800-2000) when broadcasting was stopped. You dont play perfect 6 games without a blunder and then blunder 5 times the next game. Yes playing strength fluctuate even at top level but only 50-100 elo not 500 elo.
- Anyone who played and studied chess would know that improving your level becomes multiplicatively more difficult as you get better. Getting that 400-500 elo points should take him at least 3-5 years or forever at his age not a few months.
- We have seen statistical analysis of the probability of his performance and it is pretty close to earth being hit by a meteor tomorrow.
- There are some observational evidence like never spending more than 4-5 secs on a move (which is pretty strange as humans sometimes take some time even in blitz to construct a sound strategy, evaluate different options etc) and being ultra focused on the board and never looking at opponents face (which is again strange as it is almost instinctive to look at your opponents face after you move to get a confirmation, see his reactions or self satisfaction etc.).
- His interview which he stated something along the lines when he has beaten Rybka and Houdini 10-0 each he thought he was ready for the big boys.
- Some other points which I cannot recall right now which to any person who has a background in chess and/or chess computers would be enough to see him as a cheater.

Well Taksim had a lot of "fire" last week or so after years of smoke caused by the policies of the ruling party and the message was clearly taken by the government although our prime minister looks like he does not want to back off. We'll see how it goes but I'm not expecting a Turkish spring type of revolution in Turkey but things changed for good here.

[Lyudmil Tsvetkov]

Hi Taner.
The 10-0 routs of Rybka and Houdini were obviously an emotional statement; but you can expect that from a person at whom all are pointing a finger shouting 'You, cheater!'. Imagine for a moment that the accusation comes as a complete surprise to him.

I think his earlier games show some very good positional understanding combined with making lots of mistakes. Getting rid of mistakes, improving tactically could warrant a sudden strength increase.
About playing weak in some tournaments and strong in others - but that is about true for all, look at some recent world-class tournaments, there are very often players significantly underperforming, and players significantly outperforming. And that might be valid not just for a single tournament, but for a whole stretch of events spanning an extended period of time.

The smoke is in and around Taksim Square, will there be fire?

Well Taksim had a lot of "fire" last week or so after years of smoke caused by the policies of the ruling party and the message was clearly taken by the government although our prime minister looks like he does not want to back off. We'll see how it goes but I'm not expecting a Turkish spring type of revolution in Turkey but things changed for good here.

And in Bulgaria things definitely changed for worse recently, also with a Turkish implication.....

Best, Lyudmil

[Leto]

Don, I think you and many others start from the wrong assumption that a player can not play a couple of hundred elos above his current rating. That it is impossible to learn fast. But actually modern engine software makes learning and progressing much easier than ever before. training with top engines would mean that you will be able to see some pattern in their approach to the game that would facilitate learning enormously, in sharp distinction to engine software dating back a decade ago, when trying to learn with Fritz would be difficult basically because it would be close to impossible to discern a pattern in its moves, they were like on a random basis. With state-of-the-art engines of today one should be able to progress several times quicker than otherwise, not to mention training with books and a chessboard, where already energetically producing the moves would deprive you of some learning advantages.

Big difference between playing a couple hundred elo about current rating (happens all the time) and playing 400+ elo higher than current rating. Combine that with the fact that most of his moves match Houdini 3's top moves and it's quite obvious he is cheating.

It's not matching Stockfish, or Komodo, or Rybka, or even Houdini 1.5 or Houdini 2, it's specifically matching Houdini 3. Even that alone is enough to make it obvious that he's cheating and using Houdini 3.

[Don]

Don, I think you and many others start from the wrong assumption that a player can not play a couple of hundred elos above his current rating. That it is impossible to learn fast. But actually modern engine software makes learning and progressing much easier than ever before. training with top engines would mean that you will be able to see some pattern in their approach to the game that would facilitate learning enormously, in sharp distinction to engine software dating back a decade ago, when trying to learn with Fritz would be difficult basically because it would be close to impossible to discern a pattern in its moves, they were like on a random basis. With state-of-the-art engines of today one should be able to progress several times quicker than otherwise, not to mention training with books and a chessboard, where already energetically producing the moves would deprive you of some learning advantages.

Big difference between playing a couple hundred elo about current rating (happens all the time) and playing 400+ elo higher than current rating. Combine that with the fact that most of his moves match Houdini 3's top moves and it's quite obvious he is cheating.

What is annoying to me is that people take the evidence one point at a time and think that is an effective way to refute it.

For example:

1. Anyone can have a good tournament.
2. Everyone matches Houdini's moves sometimes.
3. Anyone can have a bad tournament.
4. Just because he started playing bad when the broadcast went down does't prove anything.
5. It's possible to play 400 ELO over your head or below your rating.

Each issue, taken one at a time seems like a reasonable argument. So people who want to defend the person to the point of unreasonabless just pick one issue and make that the entire justification of the argument. They might talk about each issue in turn but they won't "combine" the evidence in the end, they will image that they refuted this 10 times. Note that in the above example point 1 and 5 are basically the same point, but if you rephrase and repeat you can make it seem like you refuted it twice. It works best when combined with an anecdote about how you once had a great tournament, etc .... SEE, IT CAN HAPPEN so therefore he is not guilty!!!


To show how ridiculous this is, suppose that you are depending on someone to roll 10 dice and report the results truthfully and you will pay them for each 6 they report rolling and that you must pay them the same amount for each 1 that is rolled. It's a silly scenario to trust someone with a private roll, but bear with me. Suppose they report that they threw 10 dice and all 10 came up 6 and you must pay them. Probably, you would suspect them of lying about it because the odds of that happening are 1 in tens of millions. But then someone will argue, "but rolling a 6 is not that rare, I can see me throwing a dice and rolling a 6."

You can really twist this by reasoning illogically and saying, it is expected to roll at least 1 or 2 sixes and perhaps 3 or 4 would not be that far fetched - let's say 4. Not that unusual to roll 4 of them - so that leaves only 6 out of 10. It's not likely but it's certainly POSSIBLE, therefore this is no evidence at ALL!

It's not matching Stockfish, or Komodo, or Rybka, or even Houdini 1.5 or Houdini 2, it's specifically matching Houdini 3. Even that alone is enough to make it obvious that he's cheating and using Houdini 3.

We have shown time and time again that even computers of equal strength have their own styles and will not match the moves of each other consistently. There is still a very strongly held misconception that if you keep getting stronger you are going to match the moves of a strong program a lot more. The reality is that you might match the moves of a strong program SLIGHTLY more and that is it. This misconception give the fallacious argument that "of course he matches Houdini, he is playing really strongly!!!"

It has been pointed out that great care should be taken when chaining together arguments. One could cherry pick what they choose to use as evidence to combine and especially if you are a prosecutor you might sub-consciously be ignoring some evidence that doesn't support you, in the same way that the defenders seem totally blind to most of the evidence. I don't know if it's possible to get a number of "totally independent observations", but you have to use some degree of reasonableness. In my dice example either side could start making up other things to pile on. Perhaps the suspected cheater has a reputation for honesty? Or you can ask, "why would he lie, he already has a lot of money of his own" and you can add that to the defense evidence and try to make that be the one that stands out. You basically just hammer on the ones you want to emphasis and ridicule the rest of the evidence and pretend you are being objective.

[Don]

Hi Taner.
The 10-0 routs of Rybka and Houdini were obviously an emotional statement;

So now you are calling him a liar with no facts to back this up and simply just saying it's "obvious?" Where are your facts? Did you see him play these 20 games? How do you know he didn't do this?

Sarcasm aside, why is it not ok for us to see what is ridiculously obvious but it's ok for you?

I think this lie (I agree with you on this) speaks to his character and motivation. It's not proof of anything but it fits a very natural pattern. He is willing to cheat and lie to look good.

I am not that willing to discount a single good performance because it has happened to me in chess and other things. But I still don't believe he won 20 games in a row. But here are 2 anecdotes about unusually good performances:

I used to bowl and I am something like a 130 average bowler, basically I suck at it. I visited my younger brother from out of state many years ago who has a little bit of a younger brother inferiority complex (the older brother usually is better at things growing up because he is older.) He challenged me to bowl as he felt he could beat me at this and was a good tournament bowler. There was little question in my mind that he would win and in fact I wanted him to win to put the foolish sibling rivalry complex stuff behind us. We bowled 3 games and every game I bowled was over 220, I was scoring like a pro and my best game that night was 268 which was just ridiculously high for me. I don't think I failed to either strike or spare. My brother bowled much higher than I should have been expected to bowl but it was nothing like I was bowling. It really struck me as a bizarre ironic thing that it worked out that way. Telling him it was a fluke and that he should have won just sounded hollow.

In other experience, I was teaching a chess playing friend to golf for the first time and we went to little par 3 golf course where you could rent clubs in order to get him started. He immediately fell in love with the game. On only our SECOND trip to this little course, he shot a hole in one on the second longest hole! It was only 105 yards, not very long as far as par 3 holes go but it was still pretty impressive and unexpected. We saw it roll up onto the green and then it was hard to see the ball. We spend about 5 minutes looking for it after getting to the green and there was really nowhere for it to go so it seemed strange to us that it wasn't in plain view but we looked very carefully around the green and wondered if some animal carried it away. Psychologically we were not even considering the possibility that it went in the hole but we eventually decided that it would make sense to check there!

... but you can expect that from a person at whom all are pointing a finger shouting 'You, cheater!'. Imagine for a moment that the accusation comes as a complete surprise to him.

I think his earlier games show some very good positional understanding combined with making lots of mistakes. Getting rid of mistakes, improving tactically could warrant a sudden strength increase.
About playing weak in some tournaments and strong in others - but that is about true for all, look at some recent world-class tournaments, there are very often players significantly underperforming, and players significantly outperforming. And that might be valid not just for a single tournament, but for a whole stretch of events spanning an extended period of time.

The smoke is in and around Taksim Square, will there be fire?

[GenoM]

And in Bulgaria things definitely changed for worse recently, also with a Turkish implication.....

Best, Lyudmil

for worse? you're a clairvoyant aren`t you?

[Dr. Axel Schumacher]

... But here are 2 anecdotes about unusually good performances:

I used to bowl and I am something like a 130 average bowler, basically I suck at it. I visited my younger brother from out of state many years ago who has a little bit of a younger brother inferiority complex (the older brother usually is better at things growing up because he is older.) He challenged me to bowl as he felt he could beat me at this and was a good tournament bowler. There was little question in my mind that he would win and in fact I wanted him to win to put the foolish sibling rivalry complex stuff behind us. We bowled 3 games and every game I bowled was over 220, I was scoring like a pro and my best game that night was 268 which was just ridiculously high for me. I don't think I failed to either strike or spare. My brother bowled much higher than I should have been expected to bowl but it was nothing like I was bowling. It really struck me as a bizarre ironic thing that it worked out that way. Telling him it was a fluke and that he should have won just sounded hollow.

In other experience, I was teaching a chess playing friend to golf for the first time and we went to little par 3 golf course where you could rent clubs in order to get him started. He immediately fell in love with the game. On only our SECOND trip to this little course, he shot a hole in one on the second longest hole! It was only 105 yards, not very long as far as par 3 holes go but it was still pretty impressive and unexpected. We saw it roll up onto the green and then it was hard to see the ball. We spend about 5 minutes looking for it after getting to the green and there was really nowhere for it to go so it seemed strange to us that it wasn't in plain view but we looked very carefully around the green and wondered if some animal carried it away. Psychologically we were not even considering the possibility that it went in the hole but we eventually decided that it would make sense to check there!

I have an even better story. This happened to me about 20 years ago; I do not blame anybody if nobody believes it. But it is true.
Once I played darts and the first of 3 darts hit the bulls eye. So far so good. However, the second arrow not only hit at the same spot in the middle, it got stuck at the back of the first dart. Imagine that it must hit within 1/100th of a millimeter in the correct angle to manage to get stuck there. Yet, my guess is that this happens now and then in the darts world; maybe every million arrow or so. However, what happens next is completely against all imaginable odds. The third dart hit again the back of the second dart and got stuck there as well. All three darts hanging in a row from the board. What is the chance?

Black swans are everywhere, you just have to notice them.

Robin Hood regards,

Axel

P.S. I think Ivanov is cheating as well, but you have to give him credit; he is doing a good job

[GenoM]

I do not know why everybody is attacking the guy while there is no evidence whatsoever that he is cheating apart from some inconclusive computer-related speculations.

There is something called "common sense" and there is something called "expertise"

Borislav is living on his own, his parents are divorced and half of the family lives abroad. I think he has a tough time of it just because of this reason. Reading some Bulgarian chess fora, you would be surprised at the viciousness of the attacks against him, although he tries to keep his ground by retorting accordingly.

Oh I`m almost touched. Poor boy, this Ivanov! By the way, I agree with your opinion about attacks against him in bulgarian chess fora -- they are way too much beyond the "bon ton".

He is not a programmer, and I am convinced that he is more or less technically ignorant, not to mention trying to make use of complex schemes beyond the grasp of everybody around for personal benefit.
He has been thoroughly checked on more than one occasions and nothing has been found. It might have been humiliating to pass through this experience. Exceptional results at blitz (the Kyustendil tournament) would also suggest that he is not making use of any engine aid. It is simply impossible to be suspected of cheating while making a couple of moves in a couple of seconds, sometimes immediately after the opponent's move. There is simply no possibility to cheat there.

"Thoroughly"? I`m not so sure. Or may be we have different views about the real meaning of this word "thoroughly".

I do not buy into those 'first lines' of Houdini. Everyone knows that first lines would depend on hardware used, time control, etc. So if Lilov claims Borislav is playing some 90% Houdini's first choices, than Lilov must be a clairvoyant to have knowledge of the hardware and settings of the machine used for cheating. I would assume it is easier to be a 'genius' than a clairvoyant. Should we now accuse Lilov of 'clairvoyancy'?

The Bulgarian chess federation banned Borislav for 4 months (exactly for the period when there are plenty of tourneys on the Black Sea Coast) on no reasonable grounds at all. They simply did not have the right to do that, as you know that everyone is innocent until proven otherwise. It is interesting to know why did not they investigate alleged cases of purchased GM and other titles throughout the years. You have the games, everything would look very suspicicous, but they would just close their eyes. It is interesting that some of the players that bought their titles (by securing points under the table in vital games) are now very vociferant against Borislav.

Bulgarian chess federation banned B.I. for his "unethical behavior" and public statements like "chess players are buttheads". That was official reason for banning him. Unofficial was that summer tournaments would suffer from many chess players boycotting them (it's just my guess based on some facts).

I have looked into the games of Borko and I would say that even before 2012 he has had a pretty nice level of play and positional understanding, scoring many wins way above what his elo status would suggest at the time. If you do this once, you can certainly do it many more times. Starting from Zadar onwards, I would say that what could arouse doubt is not the probable high-percentage of so called 'Houdini first choices', because some of those moves are really very natural and human, but the 5 to 10 percent of engine-like moves in certain games.

What would you say about his performance at Tringov Memorial this year? A steady growth? Mmm, no. So what is it?

Is 10% a high percentage to claim that he is cheating, or maybe there is a way after all for a normal human to play a couple of strictly-computer moves in a game? I would rather say that it is quite possible to play such moves, if the person has been training extensively with engines and has some natural penchant for the engines' style. I do not see why not. Borislav claims to have started training with computer assistance only recently (to be more precise, during 2012) and maybe that has boosted his strength significantly.

The guy himself says that he is neither a cheater, nor a genius, but has some natural talent for deep concentration and keeping calm throughout the games. I think concentrating properly is really very important and that could explain taking of decisions very close to computer choices.

Personally, I am convinced, that it is possible for humans to play on a much higher level than they would think they could play under normal conditions, if they have the opportunity to concentrate properly. Normal tournament conditions (high level of noise, distracting practices of opponents, unfair competition, stress) are not ideal for concentrating at a level to start playing machine-like moves, but what if a person has a natural talent for concentration?

Borislav has invited a couple of Bulgarian GMs to match their strengths with all necessary checks being carried out, but they have all refused. I would not say a cheater would make such a step. He has proposed to Kiril Georgiev to play in a 'bunker' with no possibility for electronic interference. He is eager to play in other tournaments, and I think no cheater would esily do that too. A couple of days ago he has had a thorough check carried on him in the Military Medical Academy in Sofia, and the medics said that he is clear what concerns possible subcutaneous devices, so that he has nothing implanted on him. His clothes and belongings have also been regularly checked.

Going back to those 'first line' assumptions, I do not see why a really clever cheater would not use Discocheck or Texel for aid instead of Houdini, with sufficient incoming boost of strength, as there would hardly be anyone suggesting the guy has used Discocheck (at least at the current status of Discocheck). Those first line scenarios are absolutely paranoic.

I think humans in general have difficulty imagining that they could play at a level similar to top engines, and therefore most GMs and strong players simply ignore the problem that there is an artificial intelligence far exceeding their normal GM capabilites, making their performances mostly irrelevant. And the problem is of course, lack of sufficient computer training and lack of in-depth concentration. Maybe if you find a way to concentrate at a much deeper level you would not readily join the camp of supporters of cheating accusations when they are totally devoid of ground.

Best, Lyudmil

Honestly, I never thought that in the computer chess forum there will be anyone who doubts B.I. is cheating. But it's never to late to learn new things, obviously

[Don]

... But here are 2 anecdotes about unusually good performances:

I used to bowl and I am something like a 130 average bowler, basically I suck at it. I visited my younger brother from out of state many years ago who has a little bit of a younger brother inferiority complex (the older brother usually is better at things growing up because he is older.) He challenged me to bowl as he felt he could beat me at this and was a good tournament bowler. There was little question in my mind that he would win and in fact I wanted him to win to put the foolish sibling rivalry complex stuff behind us. We bowled 3 games and every game I bowled was over 220, I was scoring like a pro and my best game that night was 268 which was just ridiculously high for me. I don't think I failed to either strike or spare. My brother bowled much higher than I should have been expected to bowl but it was nothing like I was bowling. It really struck me as a bizarre ironic thing that it worked out that way. Telling him it was a fluke and that he should have won just sounded hollow.

In other experience, I was teaching a chess playing friend to golf for the first time and we went to little par 3 golf course where you could rent clubs in order to get him started. He immediately fell in love with the game. On only our SECOND trip to this little course, he shot a hole in one on the second longest hole! It was only 105 yards, not very long as far as par 3 holes go but it was still pretty impressive and unexpected. We saw it roll up onto the green and then it was hard to see the ball. We spend about 5 minutes looking for it after getting to the green and there was really nowhere for it to go so it seemed strange to us that it wasn't in plain view but we looked very carefully around the green and wondered if some animal carried it away. Psychologically we were not even considering the possibility that it went in the hole but we eventually decided that it would make sense to check there!

I have an even better story. This happened to me about 20 years ago; I do not blame anybody if nobody believes it. But it is true.
Once I played darts and the first of 3 darts hit the bulls eye. So far so good. However, the second arrow not only hit at the same spot in the middle, it got stuck at the back of the first dart. Imagine that it must hit within 1/100th of a millimeter in the correct angle to manage to get stuck there. Yet, my guess is that this happens now and then in the darts world; maybe every million arrow or so. However, what happens next is completely against all imaginable odds. The third dart hit again the back of the second dart and got stuck there as well. All three darts hanging in a row from the board. What is the chance?

I don't believe you Ok ... maybe

Black swans are everywhere, you just have to notice them.

Robin Hood regards,

Axel

P.S. I think Ivanov is cheating as well, but you have to give him credit; he is doing a good job

Actually, I think he is doing a terrible job at cheating. He managed to immediately raise suspicion due primarily to his own stupidity.

The comment about beating Rybka and Houdini 20 games in a row (10 games each) and now he is ready for the big boys is really lame because it casts all sorts of doubt about his credibility.

If I were highly motivated to cheat at chess and had the perfect system to pass information back and forth without getting caught, I could do it FAR better than this amateur. Here is how I would proceed. Let's assume that I am about 2200 strength to being with:

1. I would use a different program for every game.

2. I would use much weaker programs most of the time. I would avoid the top 2 or 3 programs completely.

3. I would choose the moves myself and just let the machines verify my own analysis.

4. I would not immediately start winning big tournaments.

5. I would show a history of gradual improvement. Rapid, but not too rapid.

6. I would strategically make weak moves. I would make an occasional blunder but one that seemed to make sense, missing a relatively simple tactic for instance.

7. I would use my own common sense to avoid "computer-like" moves.

8. I would play the same one or two openings every time. I would mimick most strong players in this regard by specializing in just one or two.

9. I would NEVER berate the other players or the organizers. I would try to come across as being 100% cooperative even if secretly I was not cooperating.

I learned as a kid who couldn't control his reckless driving habits that when you get pulled over you don't get in the cops face. I saw this go badly with my friends. Nine times out of ten the cop let me go with a warning when I when I showed the respect that is due the officer and admitted my own foolishness.

This guy is not just winning tournaments but he is playing completely different opening systems in every game. He is doing virtually everything he can do to get caught and doing it in a very amateurish way. The best way to get away with something is to avoid raising any suspicions in the first place.

The only thing he is doing well is not getting caught with the smoking gun in his hand but much of this is due to the fact that the tournament organizer have their hands tied behind their backs.