lior5654's blog

[Suggestion] When should a round be rated for a registered user?
By lior5654, history, 4 years ago, In English

Dear codeforces community. Lately, I have started participating in some division 1 rounds, and I experience a weird issue. Suppose some Div1's problems are >= 2100 in difficulty. Then most purples after reading the problems choose to quit and not do the round. The purples that do choose to do the round, might not even solve the first problem, but the main issue is that a round is only rated for you if you submit. The latest codeforces div1 round (round #707) had about 1300 registered participants, but only ~700 submitted something, and only ~620 solved a problem. So solving problem A with bad timing causes you to be last place, despite doing better than half of the participants.

I suggest that a round should be rated for a user if he views one of the problems of the round as a registered participants. Was this suggested before? I think atcoder had some contests that worked that way but I am not sure.. Any way, the current system causes low div1's to simply quit, causing the people that do submit and succeed to lose massive amounts of rating.

I would appreciate discussing the issue in the blog's comment section.

Thanks, lior5654

  • Vote: I like it
  • +187
  • Vote: I do not like it

Comments Comments (29)
| Write comment?
»
4 years ago, # |
  Vote: I like it +2 Vote: I do not like it

Auto comment: topic has been updated by lior5654 (previous revision, new revision, compare).

→ Reply
»
4 years ago, # |
  Vote: I like it +73 Vote: I do not like it

I think that your suggestion that it counts only when someone has viewed a problem is a good idea. Those who forget or would not be able to make the round would not be adversely affected.

However, there could be cases of people using unrated accounts to view problems instead.

→ Reply
  • »
    »
    4 years ago, # ^ |
      Vote: I like it +74 Vote: I do not like it

    Doesn't matter. There is a change of mentality. When you skip the contest after reading, you're just doing something slightly unfair and making other people drop rating even when they don't deserve it, while if you read problems from an alt, you're straight up cheating.

    → Reply
    • »
      »
      »
      4 years ago, # ^ |
        Vote: I like it -49 Vote: I do not like it

      That's also cheating. A guy is registered, viewed problems and decided not to participate.

      → Reply
  • »
    »
    4 years ago, # ^ |
    ← Rev. 2   Vote: I like it -29 Vote: I do not like it

    Nevermind, skip my comment

    → Reply
»
4 years ago, # |
  Vote: I like it +12 Vote: I do not like it

It should be "Rated if registered". Anything else gives chances to cheat and tactical not-submit-anything behaviour.

→ Reply
  • »
    »
    4 years ago, # ^ |
      Vote: I like it +68 Vote: I do not like it

    I believe it's better to make the participants click a "confirmation button" like 1~2 minutes before the contest starts...

    This sounds fair to me, cuz like during an onsite contest, you have to kinda report at the place a little before the contest starts, so why not in online contests too? 2 minutes isn't a lot of time anyway

    → Reply
    • »
      »
      »
      4 years ago, # ^ |
        Vote: I like it +24 Vote: I do not like it

      Yes, of course there are several posibilities to implement this feature.

      The important point is: The decision if rated or not-rated should be before the contest starts.

      → Reply
»
4 years ago, # |
  Vote: I like it +91 Vote: I do not like it

this was discussed $$$4$$$ times earlier.

→ Reply
»
4 years ago, # |
← Rev. 2   Vote: I like it -24 Vote: I do not like it

What if someone faces some technical issues as soon as the contest starts?

Dont you think it would be unfair for them.

Upd : after seeing blogs which mentions users who are holding their submissions of A and B until they solve C, I now feel that this should definitely be implemented.

→ Reply
  • »
    »
    4 years ago, # ^ |
      Vote: I like it +26 Vote: I do not like it

    Same goes for "What if someone faces some technical issues as soon as he submits an easy problem like Div2A ?" If it's rated for him, it should be rated for anyone entering a contest too.

    → Reply
»
4 years ago, # |
  Vote: I like it +51 Vote: I do not like it

I suggested the same thing before here but got downvoted. I agree with your suggestion. Alternatively a solution would be making div1 A easier, I am sure this would increase participation.

→ Reply
»
4 years ago, # |
  Vote: I like it 0 Vote: I do not like it

Hmm, you can choose to not to log in and see the problems and then decide whether to participate or not.

I mean the round needs to be rated if you're registered. No matter you viewed the problems or not. You have option for registering 5 minutes ago. It's easy to decide to go for a contest 5 minutes ago. So, just register before the contest if you wanna participate.

→ Reply
»
4 years ago, # |
← Rev. 2   Vote: I like it -45 Vote: I do not like it

I think the round should be rated for everyone. There is a unregister button, if you are busy you can unregister and the round won't be rated for you anymore. otherwise, there is no need for registration... + if someone loses a rating because they were busy, it is their own problem..

→ Reply
  • »
    »
    4 years ago, # ^ |
      Vote: I like it -36 Vote: I do not like it

    Wow such a great idiotic idea!

    → Reply
    • »
      »
      »
      4 years ago, # ^ |
        Vote: I like it +25 Vote: I do not like it

      Would you mind explaining why you think this idea is idiotic, instead of just being toxic and insulting?

      → Reply
»
4 years ago, # |
← Rev. 5   Vote: I like it +69 Vote: I do not like it
Image

This was discussed at length recently for Atcoder, but it's a bit of a different situation because it can apply to ABCD or something, whereas on Codeforces you're still incentivized to submit A as soon as possible. So this only applies for contests with very difficult A, and your suggestion may have some merit.

I personally think it's more "user-friendly" to just allow whatever and just try to make A easy enough that this doesn't happen often. I compete & solve problems purely for recreation, so I prefer not having to jump through hoops (and sometimes if I can't do a contest because of work or other obligations, I like to read the problems and follow my friends' standings, or chat with them after the contest having already thought about the problems).

→ Reply
»
4 years ago, # |
← Rev. 2   Vote: I like it +59 Vote: I do not like it

By the way, it's very misleading to say that half the participants couldn't solve A. In every round, there are many registrants who don't submit anything (and probably just because they decided not to participate -- like they had a change of plans).

The last Div1A (Codeforces Round 706 (Div. 1)) was 1200 difficulty, had 1631 registrants, and 1104 submissions. GR13-A (Codeforces Global Round 13) was 800 difficulty, had 19477 registrants, and 10458 submissions.

→ Reply
  • »
    »
    4 years ago, # ^ |
      Vote: I like it +13 Vote: I do not like it

    Comparing the ratio of registrants to actual people who submitted something, today’s div1 round had a much lower participation rate (700/1300 to 1100/1600). I claim that this ratio is highly correlated with the actual difficulty of div1A. Also I claim that the estimated difficulty from div1As stipulated by codeforces is very wrong since codeforces only considers people who participated in the contest to calculate such number. But it does not consider people who skipped the round because they could not solve div1A.

    → Reply
    • »
      »
      »
      4 years ago, # ^ |
        Vote: I like it +47 Vote: I do not like it

      I don't remember if I registered or not but I can say that I've given up on participating on lots of contests that happened at such a time because I was basically still sleeping.

      → Reply
    • »
      »
      »
      4 years ago, # ^ |
        Vote: I like it +32 Vote: I do not like it

      Round also had unusual timing.
      Few stats regarding unusual timing div 1 rounds.
      Round 707 — 700/1300 = 0.538
      Round 691 — 733/1282 = 0.572
      Round 687 — 854/1356 = 0.630
      Round 680 — 896/1313 = 0.682
      Round 679 — 757/1311 = 0.577

      Those 40-50 people who just decided to skip this round wont even increase your delta by double digits.

      → Reply
»
4 years ago, # |
← Rev. 2   Vote: I like it +33 Vote: I do not like it

STOP RANTING. JUST BECAUSE YOU GOT -VE DELTA.

Spoiler
Spoiler
Spoiler
→ Reply
  • »
    »
    4 years ago, # ^ |
    ← Rev. 2   Vote: I like it 0 Vote: I do not like it

    Clicking on the "Read Problems" button making the round rated for oneself requires a higher level of commitment from contestants which sometimes they aren't ready to provide.

    What's the difference between committing yourself in the last 10 minutes before a round starts (by confirming your registration) vs. 5 minutes after it starts when you hit submit button on problem A?

    → Reply
»
4 years ago, # |
  Vote: I like it -12 Vote: I do not like it

Eh? It was downvoted yesterday. Please downvote now

→ Reply
»
4 years ago, # |
  Vote: I like it +5 Vote: I do not like it

I think that codeforces round should have a easy first problem which draws people to solve that problem, so people end up participating.

→ Reply