Draw by repetition

Why do I have to claim a draw by repetition? Don't you think it should be atomic? 

Can anyone help in OTB do I have to claim a draw?

I mean automatic. Sorry about that.

In US Chess Federation games, even if time has run out, they can continue playing if they don't notice one's clock has run out.  People can't interfere even to the extent of no time left.  Sometimes a player will forget to press his clock button after a move and the other player will assume the posture of deep thought, hoping his opponent won't notice his clock is still running - and no one is allowed to tell the victim!  As a former USCF Tournament Director that rule always made me angry, but it's arguable that the players ONLY must play the game and it includes clocks!

USCF rule 13C1: Only players may call flag (time running out). Only the players in a game may call attention to the fall of a flag (5G); it is considered to have fallen only when either player points this out. A director must never initiate a time-forfeit claim.

.Spectators, including players of other games, who point out the fall of a flag in any manner, may be disciplined by the tournament director to the point of expulsion from the playing room, loss of their own games, or expulsion from the tournament. The recipient of such assistance may also be penalized

USCF Rule 14C2. How to claim draw by repetition. If a move is required to complete the third occurrence of the position, the player claiming the draw under 14C should write this move on the scoresheet but not play the move on the board, stop both clocks (5I), and state the claim. If no move is required to complete the repetition, the player should stop both clocks without moving and state the claim.
In both cases, if the opponent agrees, the game is drawn. If the opponent does not agree, the claimant may make the claim to a director. If a director denies the claim, the claimant is still obligated to play any announced or recorded move. The director awards the opponent two extra minutes.

The site tries to mimic OTB, so you have to notice when a valid draw occurs and make the claim.

Thank you, that makes sense. Does this also apply to the 50 move rule?

Yes.

YEP

Either side may claim a draw after three repetitions.  However, if neither side claims the draw, it is not enforced until five repetitions.  This is the same rule as FIDE uses in OTB events.

The fifty-move rule is similar: either side may claim a draw after fifty moves with no progress, but if neither side claims the draw, it is not enforced until 75 moves with no progress.  Again, this is the same as the FIDE rule in OTB events.

Thank you all very much very informative.

So much unnecessary complication. They should just make it so that if one side doesn't feel like playing anymore it's a draw. Then it's a win-win for everybody... or at least a draw-draw.

Perhaps this is slightly off topic, but while reading this post I was thinking... 

The idea behind the different types of "draws" may very well be pointing at something much deeper than the standardized rules of some organization in present time (USCF,FIDE,etc). 

Is not chess a draw to begin with, unless another player makes a mistake?  Is this not the same case in so many "agreed drawn" late middle games and end games? 

Perhaps the offering and accepting of draws in the endgame being commonplace is a reflection of the inevitability of the draw from that simple position given the understanding of the both players.  Still the endgames are very sharp and every elite has made his share of endgame mistakes (just review one of their games by engines). 

If a draw was based upon theory alone and a draw was forced in such a position, wouldn't chess be force drawn on move one.

Just a thought, what do you think?

(let's not debate the understanding of the inevitability of a draw given the great difference needed to force a win)

It is true (almost certainly) that the starting position is a draw if both sides play perfect.  It's also irrelevant, as nobody can play perfectly for the entire game - not even engines, as it's simply not possible to analyse every possible line out to its conclusion.

Draws are agreed not because they can be forced by perfect play, but because they're easy to force, or because neither side can press for a win without taking a significant risk of losing.

I agree with most you said MGleason and thanks for the response.   However, it may be important to point out that it is not a question of the game being perfect from start to finish, but rather a question of the game to be without game losing moves from start to finish which results in a draw by force.  This can be done quite easily and is done consistently at expert level and above.  Strong players play solid moves and wait for the opponent to make a mistake which is great enough to act upon for a forced win, (the game losing mistake) .  Without a game losing mistake, we know the game will end in a draw.  Every chess game with a winner and loser contained a game losing mistake.

Another way to look at it is if two players knew beyond a shadow of a doubt that neither would make a mistake, they might consider drawing before move 1, but they would probably not play since the draw was inevitable and look for a game with complexities beyond their grasp of understanding, where mistake making may even be likely.

I doubt an expert could consistently manage a draw against Alpha Zero.

@MGleason, Alpha Zero still requires the opponent to make a mistake to win!

Yes, but how often can a mere human - at any level - manage to play an entire game without a single mistake?

It is very common, one easy way is to just trade down pieces to a "drawn" position.  I play many games with average centipawn losses ranging from .07-.29, (that is my actual average range).  Many other players (around 2000) like a few in my chess club regularly play a (.07-.15).  As you can imagine, with such low centipawn loss averages, there are definitely no game losing mistakes (blunders) or (smaller) mistakes.  Many times I will analyze our games and they might contain 2 inaccuracies over the course of 70 moves.  Players start playing such quality games from time to time around 1750-1900.  For experts 2000 and above, playing such games (without mistakes) is common, and becomes very regular around 2300.  To become 2400 or above one must play games more remarkable than just "no mistakes".

Take my case for instance, where I am only rated 1775 USCF, yet look at over 100 games I have posted on my blog  (and here's even more).  Nearly all of these games most of all contain "no mistakes" and many contain "no inaccuracies".  All are versus worthy opponents.  Not one move was made with the help of an engine.  (The idea is not to toot my horn, but to show an example and let you know the truth)

In summary, many many players (around 30,000 online) play games with no game losing mistakes more often than not and it is even common to play games without smaller (none losing) mistakes, and for the stronger (human) players even no inaccuracies, yet relatively low rated as in my case.

It may be hard to understand or believe, but I will shed some light on how to play without mistakes for you and any readers.  When one understands what matters most in the position, he will no longer make mistakes, but instead he will make remarkable moves!  How does one understand what matters most? 

Just ask someone who you think might know.

Most players have enough mistakes and inaccuracies that they won't stand a chance against most engines, assuming there isn't some particular line where a particular engine plays the exact same way each time.

If you're hardly making mistakes or inaccuracies, then your rating would be pretty good, over 2200 for sure, both OTB and online.