I'm not going to go into this too deeply in case any engine users stumble across this.
Firstly, yes, it is possible to cheat in 1+0 or 2+0 games. It's harder, though, because your hardware has to be up to spec.
Secondly, when you compare thinkabit's games with other very good 1+0 players, you notice certain issues:
http://en.lichess.org/sMQVBy7nhttp://en.lichess.org/Cv8sA9fnhttp://en.lichess.org/mS0wEjDQIf you look at all of these movetimes, they're from 100% consistent (ie, all within a small range of each other) to around 90% consistent. This trait is very common with thinkabit's games, with the lowest consistency I can spot at 82%. Additionally, he rarely makes inaccuracies, mistakes, or blunders. "Oh" but you cry, "He may just be really good - like GM level?!"
Interestingly, we can actually compare him to lichess' resident GM, Milacek. Milacek is a certified, bona fide GM, who has proven himself with ID. His movetimes range from around 70%
http://en.lichess.org/kQGSrIWYto around 30% consistency:
http://en.lichess.org/PDZlrn9l/blackFurthermore, despite being a GM he makes errors, because in 1+0 you very rarely follow the perfect line. His average number of unforced errors seems to be around 5, whereas thinkabits seems to be more like 1 or 2.
So, this would mean that thinkabit either:
a) plays 1+0 better than a certified GM
b) is receiving assistance
I flirted with a) for a while - perhaps he's Nakamura? Undoubtedly the greatest human bullet player currently alive. Yet thinkabit claimed he was Ukrainian living in New York - whilst Nakamura lives in NY, he is American. Secondly, Nakamura, it has been pointed out to me, makes around 3 inaccuracies, 1 mistake, and occasionally even a blunder in his games, even when playing other GMs - that's just the nature of 1+0.
So, you've got a player who's playing 1+0 chess at a much higher level than Milacek, our proven GM, and also Nakamura.
So, suspicious, if you then decide to look at who thinkabit has lost against, you get a couple of interesting trends:
1. He has lost frequently to other certified engine users, and those suspected of using an engine.
2. Many of his losses are due to timing out, even quite early into the game (a sign of engine use)
3. Some losses are against players much worse than him, and thinkabit making a number of errors (unusual for him).
Here's a massive blunder, against a much lower rated opponent, which even me as an 1800 wouldn't do, and wouldn't expect of someone who has been playing at 3000 level:
http://en.lichess.org/q6DSH7wBIn my view, his engine wasn't working/stopped working that game. Plus here:
http://en.lichess.org/sYAuFmnP/blackThen, thinkabits game against DeepShredder set to 2800 level and 10 moves ahead:
http://en.lichess.org/e8JHYg0c here, proves he must have been using engines in 1+0. There is no chance any human would fare as well as he did in, against an engine set at such a high rating in 1+0.
Suspicion aroused, the mods asked thinkabit if he had a title - nothing more. On the request, he replied with "I understand. You will not see me play here again."
On top of that, there's some other techniques that the system has which red flagged him for a while (around 5 months, really), except at the end of the day, it came down to a human judgement call, which on the balance of probability appears to be the correct one.
I'm going to leave this thread open for a little bit before closing it, but this is the last of the matter. It's rarely black and white for the moderation team to decide whether someone is cheating, and in these big cases, a lot of discussion is had, which may go on for weeks until everyone feels confident with the proof that it is likely someone is cheating. So this isn't some conspiracy or anything. It's just human error, mixed with thinkabit using a specialised engine.
So, that's it. No more. Carry on with your lives. The Lichess King is dead, all hail the Lichess King, etc.
(Also, he closed his account himself).