Whenever you're running out of time and can't debug proved problem go for a walk because clearly under stress conditions you would not solve problem, yes you can try to show opposite to show that you are truly GM/LGM but as practice shows (by direct pushing you can solve only very restricted class of problems, that's why market economics mostly outperforms plan economics, but even market economics starts from a plan economics.. so most things in Universe are created by indirect processes, Eratosphene created his sieve by sieve, Euler created graph theory by moving through the Keninsberg, Newtone made gravity theory by interaction with an apple, etc, so the thing is about finding out truly your own ideas as rng_58 told).. If you have no ideas how to solve problems, "practice, practice and once again practice" because nothing better was not invented..

Maybe lie down, close your eyes till you feel better.

You're stuck because you aren't training enough, like you only solved 7 1000-rated problems (install cp analysis extension).

As for stressing in contest, just ignore it, you'll be cool with it the more you participate and train.

Upsolve more problems.

Think that if you drop rating in this contest your chances of getting rating next round will go up.

practice more and more, do virtual contests and don't give too much attention for your rating

Maybe writing very simple test-cases and solving them can help find the solution. Or try a completely different approach.

And also, you can try to answer your question yourself. I mean, after you struggle with the problem, see the editorial and ask yourself: "What i could've done to get to the same thought process?" or smth like that

The answer to all questions

I've had sometimes feelings like this, but after a few months of contests I got so used to seeing "Wrong answer on test 2" that I became quite emotionally resistant to it. You can also try doing virtual contests to increase your resistance.

I think when you get a wrong answer you should try to take test cases and try to dry run your code on those test cases and make sure your code is doing what you intended it to do. Also, you should not get frustrated by it because if your rating drops below your potential you will eventually get it back in future contests.

I believe the rating system is put in such a way that the effect of some bad contests can easily be overcome (and vice versa... but you want to attain a stably well performance). Rating is very sensitive to current performance. So if you perform bad (in your context say 800 level performance) in 10 contests. You don't need 10 contests to wash off that effect. You require much less!

Once you start performing stably well, effect of past contests (whether good or bad) decreases, Hence, I make this psychological note that decrease in rating is temporary and calm myself down. Maybe you too can.

These worked for me: Try to enjoy the problem you try to solve (The more you enjoy what you do the more your brain tries to do its best D:).

Also upsolve the contests from A to the problem that you try to get stable on it.

I could only remind myself to stay calm and not bail out.

Usually, I failed the former (if you could see me in either a Codeforces round or an ICPC contest with something ranging from -4 to -10, that would certainly be me going mad).

Yet the latter, I actually pulled some nice last-minute-ACs. Maybe you could actually try to have that feeling. The mentality here is to always remember that nothing is over, that you would have to try until the timer reach zero.

Last contest was an example of this. If I actually bailed out after 2 hours, I would certainly embrace a -200 instead of just a -83 (though I clearly don't like the latter, anything salvageable is good).

edge