My peak rating was above 1500, but recently I've been constantly losing rating, and now I'm around 1200.
The frustrating part is that I can independently solve some 1900+ problems, yet I often struggle with problems rated 1200 or even lower.
I spend a lot of time thinking, analyzing different cases, and trying to derive a solution, only to discover that my very first observation was wrong. Then I read the editorial, and the key idea seems ridiculously simple.
I'm comfortable with problems where I can gradually derive a solution, but I often get completely stuck on problems that require one clever observation or greedy trick.
I've solved hundreds of problems, yet my contest performance seems to be getting worse.
Am I overthinking? Am I practicing the wrong way? Or am I missing something fundamental?
I don't think I'm that stupid, but lately it's really starting to feel that way.
Has anyone experienced this? How did you overcome it?
100%
same i too find 1600 and above problems more easier faster to solve per-day. I struggle with 1400-1500 more ig.
maybe because problem till 1400 relies on sheet logic and observation, but 1600+ problem mainly rely on data structures which you can discover quickly if u have done enough practice.
I feel like simply understanding how difficult a problem should be is very important and not talked about that much. You shouldn't expect to use DP on a Div 2 A or B, for example. You shouldn't expect a simple greedy algorithm to work on Div 2 D unless it involves binary search or trees of some kind. Point is, think more about the meta. Also, if you're finding that your first observations are wrong, then focus on trying to prove or understand why your first observation works rather than just assuming it works. You can also start off with a small, potentially false assumption about how the problem works, but only put large amount of effort into it once you have an idea of why said assumption is true. Like lets take a game between Alice and Bob with an array of even n elements, where Bob wants to maximize the range (max-min) and Alice wants to minimize it. Each turn, a person removes a chosen element, until 2 elements are left. Well, since Alice wants to minimize the range, she will remove the extreme large and small elements. There's no good reason for her to waste time on elements closer to the middle. Bob, on the other hand, wants to make sure that the 2 elements left will at least have some distance, so he will start removing elements near the median. Similarly, there's no good reason for him to spend time on elements closer to the min or max, as Alice will likely end up removing those elements in her quest to minimize the range. Therefore, the optimal solution (to my knowledge) is that Alice removes elements that greedily minimize the range, and Bob always chooses the median of the remaining numbers. Logically, neither of them can benefit from deviating from said strategy.