Tribbiani's blog

By Tribbiani, 9 years ago, In English
  1. How do you choose teammates for any team contests like ICPC regional? What is your topmost priority when choosing a teammate?

  2. During a team contest what is your strategy? Do you and your teammates divide problems among yourselves and solve individually or you try to solve one problem at a time but together?

  3. Some suggestions about team training, like what type of problems do u solve when you are practising as a team?

Lots of experienced coders are here. It will be helpful for many of us who're new if u give some suggestions.

PS: I've participated in two local team contests and those memories are not very pleasant.

  • Vote: I like it
  • +22
  • Vote: I do not like it

| Write comment?
»
9 years ago, # |
Rev. 2   Vote: I like it +6 Vote: I do not like it

while we are waiting for the rating change why not take a min of ur valuable time and comment some suggestions here :) :p

»
9 years ago, # |
  Vote: I like it +5 Vote: I do not like it
  1. If you have friends in competitive programming, try to form a team.

  2. You should train a lot and make your own style of teamwork.

  3. Solve contests of your level.

  • »
    »
    9 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    thnx :p

  • »
    »
    9 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    How to determine your level?

    • »
      »
      »
      9 years ago, # ^ |
        Vote: I like it 0 Vote: I do not like it

      If the winner solve two times more problems than you, maybe this contest is too hard for you.

      • »
        »
        »
        »
        9 years ago, # ^ |
          Vote: I like it +20 Vote: I do not like it

        I've just checked several recent World Finals... This criterion brings bad news for a most of teams there:)

        I heard a different criteria, even more loyal — contest is too easy if you are solving all problems, and contest is too hard if you can solve less than a half of problems.

        Solving harder problemsets is a good idea when you have access to all solutions/editorials and aim at upsolving.

        • »
          »
          »
          »
          »
          9 years ago, # ^ |
            Vote: I like it 0 Vote: I do not like it

          Why your criteria is more loyal? It equals mine only if winner solves all the problems.

          • »
            »
            »
            »
            »
            »
            9 years ago, # ^ |
              Vote: I like it +6 Vote: I do not like it

            It cuts a lot of harder problemsets — so it is more loyal to a team, because it leaves more pleasant sets.

            If you are solving some killing problemset from Petrozavodsk, and you are able to solve only 3 problems out of 10 — the fact that winners solved 5 doesn't make this training more enjoyable :)

»
9 years ago, # |
  Vote: I like it +16 Vote: I do not like it

Here are some nice tips regarding team strategies: http://contest-wiki.csc.kth.se/index.php/Team_strategy

  • »
    »
    9 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    this is awesome actually. i didn't know about this wiki.

»
9 years ago, # |
  Vote: I like it +9 Vote: I do not like it

Draw a small picture, icon of every problem on a page, and use it to keep track of which problems are solved, tricky to implement, hard, etc. Also, log the time you spend working on anything. Just looking at this page / table will make some management tasks simple: which problem should go next, has anyone been stuck too long, is someone working on a problem that's too hard when there is a simpler one?

Typically there is a queue of problems ready to be coded. Remember that once you have one solution idea, you can often modify it, which will make it a lot quicker to implement with less probability of mistakes. So there is still quite a lot of useful effort you can put into a problem between when it's "solved" and someone starts coding. If not, you can either get lucky, or you can get three fully-coded WA solutions that are too long and complicated to figure out why they might be wrong (as happened to us on different years).

Eventually you will probably have to solve a hard problem. One where you try everything you can think of and then get stuck after an hour. I'm still not entirely sure which is best in this case:

• Let someone else solve it independently and get stuck. Then compare notes and keep solving it as a pair. This has the advantage of not leading down the same false trail.

• Or, start solving as a pair from the start, and maybe you'll cover a lot more ground. This has the advantage that you don't all get to a psychological state that this problem is not for mere mortals (while in reality it probably takes a single insight).

In either case, often the insight comes when you explain your ideas to someone else (apparently this doesn't have to be a person), because you will find yourself having to justify assumptions as you talk through them, and one of those will turn out to be faulty.

»
9 years ago, # |
  Vote: I like it 0 Vote: I do not like it

My Advise to you, choose weak teammates but discipline ones, rather than choosing good but not discipline ones. Also Don't make my mistake and be with a team where you don't like each others, in my last ACM content i lost because in the contest everyone was working alone, and not hearing others ideas...

  • »
    »
    9 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    i had bitter experiences too :( in last national contest my teammate told me that he can solve a problem. then after some time we let him code his solution. we thought that he was certain that his solution is correct. but he blocked the computer for more than one hour and finally after telling him 3-4 times he told me his ideas and i found that his idea was wrong. what made me angry was that he didn't even checked the sample test case to see if his solution would pass.

    now the question will be why i chose him as a teammate in first place. actually i'm new in college and don't have any close friends yet. he bragged about his capablity which i now think is a bullshit -_-

»
9 years ago, # |
  Vote: I like it +10 Vote: I do not like it

Well,

1) Just take the nerdy ones

2) My strategy: do not distract guys, who are solving the problems. But I am usefull too! I usually bring some food with me and tell FUNNY jokes. I don't solve sh*t, just hanging around, having fun

3) Team training: get bunch of nerds and tell them if they not gonna win a contest, you will beat sh*t out of them. Works 100%

You are welcome, guys!

  • »
    »
    9 years ago, # ^ |
      Vote: I like it 0 Vote: I do not like it

    Then why would the nerdy guys take u as teammate... Nerdy doesn't necessarily mean stupid.

    And after all I'm trying to improve myself through team training. Not willing to take advantage of others.