So can you get a negative rating just by playing the same opponent many times and losing all the games?
Can Karjakin surpass Carlsen ELO rating just drawing with him enough times?

So can you get a negative rating just by playing the same opponent many times and losing all the games?
Well, I I don't think FIDE allows negative ratings. But if they did, then in theory yes, if you play all of the games between rating updates. But it would have to be a lot of games.

The answer to the initial question is of course NO. You cannot surpass somebody by just making draws. You only can have the exact same rating after many many draws.
I shouldn't be surprised.
Zetinot gave a clear, concise and correct answer on the first page, but the thread continues anyway.
The answer to the initial question is of course NO. You cannot surpass somebody by just making draws. You only can have the exact same rating after many many draws.
I shouldn't be surprised.
Zetinot gave a clear, concise and correct answer on the first page, but the thread continues anyway.
It seems, e.g. here, that FIDE doesn't update the ratings each time, in which case it would be possible to surpass someone's rating by simply drawing. This answer was also given on the first page.

SmyslovFan wrote:
zeitnotakrobat wrote:
The answer to the initial question is of course NO. You cannot surpass somebody by just making draws. You only can have the exact same rating after many many draws.
I shouldn't be surprised.
Zetinot gave a clear, concise and correct answer on the first page, but the thread continues anyway.
xxxxxxx
And yet you don't know the answer. The courage of the ignorant

I hope SmyslovFan realizes his mistake and apologizes. The resident troll is better informed and more helpful (and nice) that him.

I hope SmyslovFan realizes his mistake and apologizes. The resident troll is better informed and more helpful (and nice) that him.
And macer wasn't banned for cheating

I don't know for sure, but I think FIDE would have safeguards in place to protect the ratings so that such a scenario would be rectified before ratings are published. I do know they have people to double check the results before they are published.
I think it's not actually possible, even if there were a theoretical scenario that would take advantage of how ratings are calculated in a single rating period.

My answer was based on common sense and the fact that in our league we get rating updates every month. But it might be wrong for some cases that can you can construct.
If you apply a rating update every month Karjakin would gain 22, 10, 4, 2, 2, 0 points if they draw 20 games per month according to the fide ratings calculator. There migth be a change for the last month, but the calculator has not enough digits to show them.
If you assume no rating updates Karjakin can get to a higher rating than Carlsen by 37 draws within one period.
However, there is also a special rule for tournaments that might run longer than 90 days that the interim results must be submitted on a monthly basis. But there is no statement regarding rating updates for such tournaments.
In the end the right answer therefore depends only on the rating update period you assume.
In theory, if he does it enough times in the same month, then yes. Ratings are only updated once a month, and in the meantime, no matter what the live ratings of the two players are, the rating change in the case of a draw (or a win/loss, for that matter) is entirely based on the two players' last official ratings. Currently Carlsen loses 1.1 points for a draw, and Karjakin gains that many points, so if they were to draw 50 games before the official ratings are updated, then yes, Karjakin would surpass Carlsen.
Wow, FIDE is ridiculous. To use live ratings would require what... a single modern computer?