In today's round Codeforces Round 996 (Div. 2), the gap between problem difficulties was hilariously wide, according to clist. The last two problems are rated approximately 3000 and 3500, which clearly means that they are not suitable for participants rated below 2100.
I think this makes the round literally consist of only four problems, where your performance depends mostly on speed. According to Carrot, the performance of the second-place participant was LGM, who solved four problems. There are also participants who solved four problems and achieved a CM-level performance, which is very weird.
I am not complaining only about today's contest, but it serves as a great example of the unbalanced difficulty seen in recent contests. This imbalance particularly affects CPers rated between 1600 — 2100, who do not have the opportunity for rated contests except for div 2 contests, making us have fewer opportunities to achieve a positive delta, as luck starts to play a significant role in determining our rating.
It's honestly not very obvious to me why so few people have solved problem E. I'm not saying this is a simple task, I haven't solved it either (I really don't like this kind of task), but judging by the analysis, this is a common task a little more complicated than d2E, definitely not deserving of the complexity of 3000
Let us neglect the 3000, it is still complicated for a < 2100 participant
Also, note that the only one who solved it in div2 was a grandmaster. XD
yep, I agree it is hard even for some 2100+ guys, but still solvable. Also I would like to have more 2300+ tasks in div2s, rather then 2700+. But challenging tasks create strong programmers)
Wow, only GM solved this? This is really strange)
GPT era
I am still puzzled about the difficulty of D. I could't find out a complete correct sol during the contest with thinking the situation is too complicated. Maybe I am so bad at greedy. How could so much people solve this problem? Could anyone introduce some similar greedy problems?
"who do not have the opportunity for rated contests except for div 2 contests, making us have fewer opportunities to achieve a positive delta".
It's another one of these: https://codeforces.me/blog/entry/129569
Whenever I bring up this topic, the usual claim is that these problems are too standard for higher divisions. But in my opinion, if a problem is too standard to appear in a higher division, it shouldn't appear in ANY division at all and should only be made as a non-contest problem. I don't really get why these problems are considered to be okay if they are in low enough division so that they aren't solvable for almost anyone in the division. To be honest, I feel that the standard for this standardness is in general too high, but that's for another discussion.
Even if we keep the current trend of having unsolvable-for-official-participants problems, they should at least not be counted when we construct a problemset. I think every problem that fits a division should at least have a nontrivial chance to be solvable by the official participants, and these participants shouldn't only be someone whose actual skill level is something like GM but has low rating just because of too few contest participation or something. I think at least 5 problems should be feasible for some good number of CMs, and even for some Experts, considering that Div. 2 was originally meant to be targeted to up to Experts. Otherwise, we will only be able to keep having these huge difficulty gaps between problems, causing huge performance difference between fast $$$X$$$ solves and slow $$$X$$$ solves for every $$$X$$$.
I also think we shouldn't be afraid of having too many AKs, if that 'too many' is something like 50. It's still only a very small proportion of the participants (even within CMs alone), and most of them would only have solved them in the very last 15 minutes. Almost no official participant "wasted" more than 15 minutes.