{
  "id": 132016,
  "title": "Kenpom Data",
  "url": "/competitions/google-cloud-ncaa-march-madness-2020-division-1-mens-tournament/discussion/132016",
  "author_name": "Zachary Snoen",
  "post_date": "2020-02-23T12:43:12.189000",
  "votes": 2,
  "comment_count": 3,
  "views": 0,
  "content": "<p>As we all have probably realized, Kenpom's current data for previous years includes tournament results from those years. Has anybody found or have a dataset from him that does not include those games? Appreciate any feedback or help on how to achieve this, as I believe most of us are overfitting our data.</p>\n\n<p>If you have any idea how to calculate just how much this affects our scores that would be great too because I am inexperienced in the field and my score should not be as good as it is comparing it to past winners scores.</p>",
  "messages": [
    {
      "id": 754345,
      "postDate": "2020-02-23T12:43:12.190Z",
      "content": "<p>As we all have probably realized, Kenpom's current data for previous years includes tournament results from those years. Has anybody found or have a dataset from him that does not include those games? Appreciate any feedback or help on how to achieve this, as I believe most of us are overfitting our data.</p>\n\n<p>If you have any idea how to calculate just how much this affects our scores that would be great too because I am inexperienced in the field and my score should not be as good as it is comparing it to past winners scores.</p>",
      "rawMarkdown": "As we all have probably realized, Kenpom's current data for previous years includes tournament results from those years. Has anybody found or have a dataset from him that does not include those games? Appreciate any feedback or help on how to achieve this, as I believe most of us are overfitting our data.\n\nIf you have any idea how to calculate just how much this affects our scores that would be great too because I am inexperienced in the field and my score should not be as good as it is comparing it to past winners scores.",
      "votes": 2
    },
    {
      "id": 757658,
      "postDate": "2020-02-27T02:08:02.543Z",
      "content": "<p>Despite your inexperience, your skepticism shows good instincts and judgment.</p>",
      "rawMarkdown": "Despite your inexperience, your skepticism shows good instincts and judgment.",
      "votes": -1
    },
    {
      "id": 754640,
      "postDate": "2020-02-23T21:28:59.283Z",
      "rawMarkdown": "",
      "isDeleted": true,
      "replies": [
        {
          "id": 754649,
          "postDate": "2020-02-23T21:49:19.433Z",
          "content": "<p>I think they are saying that the late February games or early March games are less predictive of late March tournament games than early February games are.  However, late March tournament games are going to be extremely predictive of themselves, and so ratings that already incorporate those late March tournament games are not appropriate to use to predict late March tournament games from the recent past.  Notice the wording where it says \"games <strong>at the end of the regular season</strong>\" do not have a dramatic effect.</p>",
          "rawMarkdown": "I think they are saying that the late February games or early March games are less predictive of late March tournament games than early February games are.  However, late March tournament games are going to be extremely predictive of themselves, and so ratings that already incorporate those late March tournament games are not appropriate to use to predict late March tournament games from the recent past.  Notice the wording where it says \"games **at the end of the regular season**\" do not have a dramatic effect.",
          "votes": 1
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 757658,
      "author_name": "S Charlesworth",
      "author_url": "",
      "post_date": "2020-02-27T02:08:02.543000",
      "content": "<p>Despite your inexperience, your skepticism shows good instincts and judgment.</p>",
      "votes": -1,
      "replies": []
    },
    {
      "id": 754640,
      "author_name": "",
      "author_url": "",
      "post_date": "2020-02-23T21:28:59.283000",
      "content": "",
      "votes": 0,
      "replies": [
        {
          "id": 754649,
          "author_name": "Jeff Sonas",
          "author_url": "",
          "post_date": "2020-02-23T21:49:19.433000",
          "content": "<p>I think they are saying that the late February games or early March games are less predictive of late March tournament games than early February games are.  However, late March tournament games are going to be extremely predictive of themselves, and so ratings that already incorporate those late March tournament games are not appropriate to use to predict late March tournament games from the recent past.  Notice the wording where it says \"games <strong>at the end of the regular season</strong>\" do not have a dramatic effect.</p>",
          "votes": 1,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "754345": "As we all have probably realized, Kenpom's current data for previous years includes tournament results from those years. Has anybody found or have a dataset from him that does not include those games? Appreciate any feedback or help on how to achieve this, as I believe most of us are overfitting our data.\n\nIf you have any idea how to calculate just how much this affects our scores that would be great too because I am inexperienced in the field and my score should not be as good as it is comparing it to past winners scores.",
    "757658": "Despite your inexperience, your skepticism shows good instincts and judgment.",
    "754640": ""
  }
}