{
  "id": 5899,
  "title": "how is timeline decided?",
  "url": "/competitions/dogs-vs-cats/discussion/5899",
  "author_name": "",
  "post_date": "2013-09-27T14:10:40.410Z",
  "votes": null,
  "comment_count": 2,
  "views": 1400,
  "content": "<p>I've always wondered how Kaggle decides the timeline of a competition it is hosting. For example, in this particular competition, Jeff achieved a score of 0.96 just within two days. This score is very competitive and IMO, the winning score will be around 0.98. When these kind of scores can be achieved within a short period of time, then what's the use of running a particular competition for a period of 4 months? Won't the results be same even if the duration is halved?&nbsp;</p>",
  "messages": [
    {
      "id": "31660",
      "postDate": "09/27/2013 14:10:40",
      "content": "<p>I've always wondered how Kaggle decides the timeline of a competition it is hosting. For example, in this particular competition, Jeff achieved a score of 0.96 just within two days. This score is very competitive and IMO, the winning score will be around 0.98. When these kind of scores can be achieved within a short period of time, then what's the use of running a particular competition for a period of 4 months? Won't the results be same even if the duration is halved?&nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "31661",
      "postDate": "09/27/2013 14:23:58",
      "content": "<p>Hey Abhishek! We try to balance a few factors:</p>\n<ul>\n<li>If it's too long people start to fit noise and compete on non-meaningful decimal places</li>\n<li>If it's too short people don't see it, or fit it in their schedules. &nbsp;Participation grows linearly with competition duration.</li>\n<li>Once we pick a deadline, we try really hard not to shorten a live competition.</li>\n<li>We err on the side of a little too long over a little too short just to have a margin of safety.</li>\n</ul>\n<p>This competition is for fun and benchmarking, so we picked a longer time to let folks experiment with it, and we're not as concerned if the scores plateau. &nbsp;That said, I thought it would take much much longer to get scores above 0.8!&nbsp;</p>\n<p>There will definitely be a second test set for this competition. Scores this good deserve a reliable estimate of performance on a true holdout sample.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "31682",
      "postDate": "09/28/2013 03:00:57",
      "content": "",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 31661,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "09/27/2013 14:23:58",
      "content": "<p>Hey Abhishek! We try to balance a few factors:</p>\n<ul>\n<li>If it's too long people start to fit noise and compete on non-meaningful decimal places</li>\n<li>If it's too short people don't see it, or fit it in their schedules. &nbsp;Participation grows linearly with competition duration.</li>\n<li>Once we pick a deadline, we try really hard not to shorten a live competition.</li>\n<li>We err on the side of a little too long over a little too short just to have a margin of safety.</li>\n</ul>\n<p>This competition is for fun and benchmarking, so we picked a longer time to let folks experiment with it, and we're not as concerned if the scores plateau. &nbsp;That said, I thought it would take much much longer to get scores above 0.8!&nbsp;</p>\n<p>There will definitely be a second test set for this competition. Scores this good deserve a reliable estimate of performance on a true holdout sample.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 31682,
      "author_name": "tritonsd",
      "author_url": "",
      "post_date": "09/28/2013 03:00:57",
      "content": "",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "31660": "",
    "31661": "",
    "31682": ""
  },
  "source": "meta"
}