Hey Codeforces community,
The recent selections for the ICPC India Amritapuri Regionals just dropped, and the current selection algorithm feels increasingly broken for competitive programming as a sport.
In the preliminary round, we solved 6 problems (A-E and H) and finished with an overall rank of ~450. However, because we were College Rank 2 at our institute, we did not get selected for the regionals.
Meanwhile, teams with a global rank around ~3700 are advancing simply because they hold the College Rank 1 spot at their respective universities, often with only 1 or 2 solves.
It is incredibly frustrating to see top-500 teams get boxed out by 1-solve teams purely due to institutional diversity quotas. Guaranteeing spots to smaller colleges might help grassroots participation, but the current math heavily penalizes secondary teams at competitive colleges to the point where absolute merit is completely ignored.
I understand the underlying philosophy: ICPC wants to promote grassroots competitive programming across as many colleges as possible. Guaranteeing a slot to the top team from every college encourages participation and justifies registration fees for smaller institutions. But there has to be a limit to how heavily meritocracy is penalized in the name of ecosystem growth.
Is this really the best way to foster elite competitive programming?
Setting a minimum benchmark like solving at least 'X' number of questions, makes a lot of sense if you want to ensure a baseline level of competency for regional qualifiers
How many team can participate in the official icpc on site in india icpc ?
there is a selection process,
all top 50 teams has no restriction, even if all 50 teams are from the same college then all 50 get selected
then next top 100, each college should not exceed 5 teams
then next top 140, each college should not exceed 2 teams
then the rest, only top institute get qualify
Thats really frustrating.
So, what's the problem? Just transfer to a university with weaker teams.
why would you want to go to a worse university for years just to compete a single competition
If you don't give a damn about ICPC, then these qualifiers shouldn't concern you. If you love CP so much, then such transgression shouldn't be a problem to you. I can tell you, that it's a legit practice for strong Russian teams to switch unis just to have an opportunity to qualify for finals. Besides, most of them just enroll into the second university and make a deal with it: they study only in the primary uni and doesn't bother with the second one, and in return they bring medals for this uni.
This in no way justify how bad the qualification process is
https://codeforces.me/blog/entry/157370
You can't expect regionals in India even after giving your best. The reason is given in the above blog.
Fckkkkk ICPC amritapuriii, My friends team secured 2nd place in their college, having solved 6 problems and achieved a rank within the top 350, yet they still haven't been selected.
What makes this even more frustrating is that the selection criteria were changed recently, whereas they registered based on the criteria announced during the registration process. If the criteria were going to be changed, the organizers should have communicated this before the registration process began, rather than changing them after teams had already registered and competed based on the originally announced criteria.
Extremely poor management and a complete lack of transparency from ICPC Amritapuri. Some random tier 69 colleges are selected. they destroyed the pride of ICPC in india, even school students can solve that 1st question. They completely diluted the prestige and significance of the ICPC Regionalist tag.
+1, not to mention the rampant cheating which plagued the contest (because they left it to the colleges and prayed everyone would give it honestly??), and their post-contest BS plag check.
Was trying to give them the benefit of the doubt, but the conduct of the prelims this year has been exceptionally pathetic.
Well for the vast majority of elite teams, they would have been selected. In terms of meritocracy, this is absolutely not the way to go. Lowkey feels like college apps lol
Bro fym top 80 Team isn't qualifying for Amritapuri but fucking random ass Teams qualify just because they are best in their college
Agree,
There should be some min no of problems that should be solved to get into regionals,
This no should not be fixed, depending upon the contest difficulty it should change.
This 3 questions were very basic, so they should have considered teams who have solved 3 or more problems.
Love to know other's view : )
really frustrating to see the deserving candidates being left out while people get selected due to quotas, just ruins the day >:
On your point of fostering elite competitive programming, the regional sites have reserved slots for top performing teams regardless of college quota like amrita had 50 seats reserved for it, so if your goal isn't primarily just reaching regionals then just place in top 50 and like we gotta draw a line somewhere, high and diverse participation is imp for keeping competition alive.
I agree that diversity is important and it's good that the top 50 merit rule is there. But telling a team that solved 6 problems and ranked 450 to
just get top 50while a team that solved 1 problem and ranked 3000+ gets a free pass shows the exact double standard I am talking about. The standard for merit should not just apply to the top 1% of the country. There are many strong, dedicated teams in the 51-500 range. Dismissing their hard work by sayingjust place higherwhile drastically lowering the bar for others actively demotivates the middle to high tier of competitive programmers.I agree that it kinda is discouraging when a 1 solve team gets qualified over 6 solves team, but there isnt any easy solution for this problem and creating some heuristic say a function of rank and solves just makes things confusing but like we are trying to balance two extremes:
allowing only one team per-college
allowing all top ranked teams regardless of college quota
This will more or less lead to ugly situation for some class of teams like yours, if not you(2nd ranked in your college) some x rank in your college will be better than other qualified teams and still not qualify.
Then there is extreme examples like IIT Roorkee which has 9 teams in top 100 and 18 in top 200, those will be on worst end of such rules and its not so easy to just come up with rules that satisfies all parties
Isn't there a very simple solution : Set a minimum threshold for no. of problems solved to be considered for regionals. I think for this year, a reasonable such threshold could be 4 problems. They could think of how to formalize this into a rule based on some statistical property of the solve distribution.
I think this was already done for many years. I do not understand why they decided to change it this year, when they already made a massive mess from the abysmal (lack of) proctoring. All this shows is the organizers are not in touch with their contestants, and are more interested in lining their own pockets in the name of "diversity".
While I do agree that diversity is important , the fact is that all the teams with 1-2 solves would just not be doing good at the onsite regionals. They would be up against the best teams of the country and while it is a good opportunity for them to gain experience, the skill differential would render it potentially useless. If the team solving 1 problem is getting through above a team solving 6 that just goes out to show an issue with the selection process itself.
The aim is not to make them perform well, but to provide exposure so that a culture can be fostered in their colleges, allowing us to see great competitive programmers in a few years. The number of seats used to be much lower a few years ago. The increase in seats was for exposure, while making sure that no team that could have qualified for the finals misses a regional spot. X = 5 is pretty safe for that (initially, it used to be at most 2) because if you are not in the top 100 or a top-5 college team, then you probably didn't perform well enough to make it to finals this year.
Even I think there is a decent buffer so that an Asia West team doesn't miss out. Before 2020, the top 1 or 2 teams from smaller regionals and the top 2 or 3 from larger regionals were directly picked for the world finals. Asia West is also a similar effort to provide a stepping stone for a lot of colleges that may never qualify for the finals directly, but with a consistent culture of going to Asia West, they can make it to the finals in a few years.
I somewhat reluctantly agree with you here. While I haven't personally been in that position, it's hard not to sympathize with those who have. At the same time, I understand the argument for increasing exposure and CP culture across more colleges. Ultimately, I don't think there's a win-win decision here. Either way, one side is bound to feel that the system has been unfair to them.
If you think the quota system is unfair, be grateful you are in a privileged position. If you think it is fair, be grateful the system supports those who need it.
In my last prelims, we were ranked 13th with just two minutes remaining. Under the rules, we weren't going to regionals as each site granted direct entry to the top 10, plus the top team per college, and another team from our university held 11th place. We desperately needed a something. With less than 10 seconds left, we submitted a solution and got an AC verdict post-contest, pushing us to 7th place and eventually ended up at the world finals.
Is ICPC harder when you're at a college with top-tier competitive programming culture? Absolutely. But would you reach the same level of skill without that fierce internal competition? Probably not.
At the end, competitions are zero-sum game so every laugh and cheer will have some sad and hopeless faces at the other side. But the pursuit and preparation are positive sum game so focus on that. All of these folks wouldn't have grown this much if there is no competition to look forward.
Fair enough , I guess it's just easier to rant than accept the actual facts.
So, it turns out you werent competing against teams from other universities, but rather against a team from your own college — that looks absolutely awful. A certain number of teams should qualify, starting from the top-ranked ones, a college team shouldnt get to advance simply because it happens to be the only team from that college