{
  "id": 588855,
  "title": "Question on Contest Rules",
  "url": "/competitions/drw-crypto-market-prediction/discussion/588855",
  "author_name": "",
  "post_date": "2025-07-08T17:47:06.965963300Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I've been checking out the newest publicly-available submissions, which are far higher in score than the other submissions and which appear to be inferring the order of the test data and using that to make predictions. Per my understanding, that approach is prohibited by contest policy and will result in disqualification. Am I correct about that?</p>",
  "messages": [
    {
      "id": "3244953",
      "postDate": "07/08/2025 17:47:06",
      "content": "<p>I've been checking out the newest publicly-available submissions, which are far higher in score than the other submissions and which appear to be inferring the order of the test data and using that to make predictions. Per my understanding, that approach is prohibited by contest policy and will result in disqualification. Am I correct about that?</p>",
      "rawMarkdown": "I've been checking out the newest publicly-available submissions, which are far higher in score than the other submissions and which appear to be inferring the order of the test data and using that to make predictions. Per my understanding, that approach is prohibited by contest policy and will result in disqualification. Am I correct about that?",
      "votes": null
    },
    {
      "id": "3245058",
      "postDate": "07/08/2025 20:26:49",
      "content": "<p>That's my understanding, but honestly it's frustrating either way. Very hard to know what truly consistitutes a good score here, how much is through inferring test data, what the consequences will be, if there's really any point spending more time doing it \"properly\" or not.</p>",
      "rawMarkdown": "That's my understanding, but honestly it's frustrating either way. Very hard to know what truly consistitutes a good score here, how much is through inferring test data, what the consequences will be, if there's really any point spending more time doing it \"properly\" or not.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3245058,
      "author_name": "greatestcutie",
      "author_url": "",
      "post_date": "07/08/2025 20:26:49",
      "content": "<p>That's my understanding, but honestly it's frustrating either way. Very hard to know what truly consistitutes a good score here, how much is through inferring test data, what the consequences will be, if there's really any point spending more time doing it \"properly\" or not.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3244953": "I've been checking out the newest publicly-available submissions, which are far higher in score than the other submissions and which appear to be inferring the order of the test data and using that to make predictions. Per my understanding, that approach is prohibited by contest policy and will result in disqualification. Am I correct about that?",
    "3245058": "That's my understanding, but honestly it's frustrating either way. Very hard to know what truly consistitutes a good score here, how much is through inferring test data, what the consequences will be, if there's really any point spending more time doing it \"properly\" or not."
  },
  "source": "meta"
}