{ "id": 343687, "title": "Various data problems", "url": "/competitions/big-data-derby-2022/discussion/343687", "author_name": "", "post_date": "2022-08-12T06:30:40.206065100Z", "votes": 10, "comment_count": 3, "views": 0, "content": "

Data problems

\n

1 The program_number feature has an incorrect data type.It also has spaces.This is easily seen with the value_count() command.
\ndata['program_number'].value_count()

\n
10     118871\n9       94753\n11      72582\n10      67964\n1A      56753\n12      45569\n
\n

This is how the data looks like '6 '
\n2 The track_condition also has extra gaps in the data.
\nIf you find any more oddities in the data, write:)

\n

Interesting things in the data

\n

1 Almost every track has its own joker.They rarely intersect.
\n2 Categorical data correlate well with each other.I checked with the help of statistics:
\nstats.f_oneway

", "messages": [ { "id": "1895451", "postDate": "08/12/2022 06:30:40", "content": "

Data problems

\n

1 The program_number feature has an incorrect data type.It also has spaces.This is easily seen with the value_count() command.
\ndata['program_number'].value_count()

\n
10     118871\n9       94753\n11      72582\n10      67964\n1A      56753\n12      45569\n
\n

This is how the data looks like '6 '
\n2 The track_condition also has extra gaps in the data.
\nIf you find any more oddities in the data, write:)

\n

Interesting things in the data

\n

1 Almost every track has its own joker.They rarely intersect.
\n2 Categorical data correlate well with each other.I checked with the help of statistics:
\nstats.f_oneway

", "rawMarkdown": "### **Data problems**\n\n\n1 The program_number feature has an incorrect data type.It also has spaces.This is easily seen with the value_count() command.\n\n\n\ndata['program_number'].value_count() \n```\n\n10 118871\n9 94753\n11 72582\n10 67964\n1A 56753\n12 45569\n```\n\nThis is how the data looks like `'6 '`\n\n2 The track_condition also has extra gaps in the data.\nIf you find any more oddities in the data, write:)\n\n\n### **Interesting things in the data**\n\n\n\n1 Almost every track has its own joker.They rarely intersect.\n\n2 Categorical data correlate well with each other.I checked with the help of statistics:\n\n`stats.f_oneway`", "votes": null }, { "id": "1896022", "postDate": "08/12/2022 14:16:02", "content": "

“program_number” is a 3 character string

\n

The tracks shouldn’t intersect, each track is its own location geographically.

", "rawMarkdown": "“program_number” is a 3 character string \n\nThe tracks shouldn’t intersect, each track is its own location geographically.", "votes": null }, { "id": "1896659", "postDate": "08/13/2022 03:01:29", "content": "

\"1A\" is a coupled entry with \"1\"--they are considered a single betting interest because of shared ownership (if both a \"1A\" and \"1\" are racing they'll have the same odds). Often one of the entries will be subsequently scratched and not race.

\n

From an analytical perspective, coupled entries are more likely to have different race styles--one will try and go to the lead while the other will hang back to make a late charge, etc. They will rarely, if ever, duel with each other early.

", "rawMarkdown": "\"1A\" is a coupled entry with \"1\"--they are considered a single betting interest because of shared ownership (if both a \"1A\" and \"1\" are racing they'll have the same odds). Often one of the entries will be subsequently scratched and not race.\n\nFrom an analytical perspective, coupled entries are more likely to have different race styles--one will try and go to the lead while the other will hang back to make a late charge, etc. They will rarely, if ever, duel with each other early.", "votes": null }, { "id": "2011832", "postDate": "10/31/2022 20:28:57", "content": "

Thank you so much for mentioning the extra gaps in track_condition

", "rawMarkdown": "Thank you so much for mentioning the extra gaps in track_condition", "votes": null } ], "comments": [ { "id": 1896022, "author_name": "nythanyra", "author_url": "", "post_date": "08/12/2022 14:16:02", "content": "

“program_number” is a 3 character string

\n

The tracks shouldn’t intersect, each track is its own location geographically.

", "votes": null, "replies": [] }, { "id": 1896659, "author_name": "btimoney", "author_url": "", "post_date": "08/13/2022 03:01:29", "content": "

\"1A\" is a coupled entry with \"1\"--they are considered a single betting interest because of shared ownership (if both a \"1A\" and \"1\" are racing they'll have the same odds). Often one of the entries will be subsequently scratched and not race.

\n

From an analytical perspective, coupled entries are more likely to have different race styles--one will try and go to the lead while the other will hang back to make a late charge, etc. They will rarely, if ever, duel with each other early.

", "votes": null, "replies": [] }, { "id": 2011832, "author_name": "meganrms", "author_url": "", "post_date": "10/31/2022 20:28:57", "content": "

Thank you so much for mentioning the extra gaps in track_condition

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1895451": "### **Data problems**\n\n\n1 The program_number feature has an incorrect data type.It also has spaces.This is easily seen with the value_count() command.\n\n\n\ndata['program_number'].value_count() \n```\n\n10 118871\n9 94753\n11 72582\n10 67964\n1A 56753\n12 45569\n```\n\nThis is how the data looks like `'6 '`\n\n2 The track_condition also has extra gaps in the data.\nIf you find any more oddities in the data, write:)\n\n\n### **Interesting things in the data**\n\n\n\n1 Almost every track has its own joker.They rarely intersect.\n\n2 Categorical data correlate well with each other.I checked with the help of statistics:\n\n`stats.f_oneway`", "1896022": "“program_number” is a 3 character string \n\nThe tracks shouldn’t intersect, each track is its own location geographically.", "1896659": "\"1A\" is a coupled entry with \"1\"--they are considered a single betting interest because of shared ownership (if both a \"1A\" and \"1\" are racing they'll have the same odds). Often one of the entries will be subsequently scratched and not race.\n\nFrom an analytical perspective, coupled entries are more likely to have different race styles--one will try and go to the lead while the other will hang back to make a late charge, etc. They will rarely, if ever, duel with each other early.", "2011832": "Thank you so much for mentioning the extra gaps in track_condition" }, "source": "meta" }