![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8412510%2Feda78a7241a08aff64f3bc9b521d7c51%2FFinal_Sar_Summary.png?generation=1666467906484458&alt=media\")
Welcome to the Big Data Derby! We're so excited to once again expand our Analytics Competition offerings to the Kaggle community to explore unique competition ideas that tell compelling stories.
\nIf this is your first analytics competition, I'd encourage you to see Kaggle Grandmaster John Miller's excellent write-up on tips and tricks for making a great analytics submission, as well as the top solutions of other analytics competitions.
\nThe types of questions the NYTHRA/NYRA is looking to see examined are provided on the Evaluation page. Please note that list is not comprehensive, nor is it meant to be a guide for you to try and specifically answer (or address all the questions).
\nAs a reminder - Submissions must be made public at the competition deadline to be considered eligible for scoring (The NYTHA/NYRA has to be able to actually see your hard work!).
\nBest of luck, and happy modeling!
", "messages": [ { "id": 1895000, "postDate": "2022-08-11T21:03:07.097Z", "content": "Welcome to the Big Data Derby! We're so excited to once again expand our Analytics Competition offerings to the Kaggle community to explore unique competition ideas that tell compelling stories.
\nIf this is your first analytics competition, I'd encourage you to see Kaggle Grandmaster John Miller's excellent write-up on tips and tricks for making a great analytics submission, as well as the top solutions of other analytics competitions.
\nThe types of questions the NYTHRA/NYRA is looking to see examined are provided on the Evaluation page. Please note that list is not comprehensive, nor is it meant to be a guide for you to try and specifically answer (or address all the questions).
\nAs a reminder - Submissions must be made public at the competition deadline to be considered eligible for scoring (The NYTHA/NYRA has to be able to actually see your hard work!).
\nBest of luck, and happy modeling!
", "rawMarkdown": "Welcome to the Big Data Derby! We're so excited to once again expand our Analytics Competition offerings to the Kaggle community to explore unique competition ideas that tell compelling stories. \n\nIf this is your first analytics competition, I'd encourage you to see Kaggle Grandmaster John Miller's [excellent write-up](https://www.kaggle.com/jpmiller/creating-a-good-analytics-report) on tips and tricks for making a great analytics submission, as well as the top solutions of other analytics competitions.\n\nThe types of questions the NYTHRA/NYRA is looking to see examined are provided on the [Evaluation](https://www.kaggle.com/competitions/big-data-derby-2022/overview/evaluation) page. Please note that list is not comprehensive, nor is it meant to be a guide for you to try and specifically answer (or address all the questions). \n\nAs a reminder - Submissions must be made public at the competition deadline to be considered eligible for scoring (The NYTHA/NYRA has to be able to actually see your hard work!).\n\nBest of luck, and happy modeling!", "votes": 41 }, { "id": 1998569, "postDate": "2022-10-21T17:41:00.733Z", "content": "Hello Addison:
\nWorking on this project with a mathematical model that does not depend on distances but only on coordinates, I find that the GPS signal is not every 0.25 seconds but every 0.2825 seconds in all the races.
\nI leave the information here so that the other competitors can take advantage of it, otherwise the speeds and distances will not be correct.
\nI have verified this by watching the races on YouTube and corroborating the times of my model with the televised race.
", "rawMarkdown": "Hello Addison:\n\nWorking on this project with a mathematical model that does not depend on distances but only on coordinates, I find that the GPS signal is not every 0.25 seconds but every 0.2825 seconds in all the races.\n\nI leave the information here so that the other competitors can take advantage of it, otherwise the speeds and distances will not be correct.\n\nI have verified this by watching the races on YouTube and corroborating the times of my model with the televised race.\n", "votes": 5, "replies": [ { "id": 1998699, "postDate": "2022-10-21T19:23:34.827Z", "content": "Hello Gabriel,
\nYou are indeed correct about GPS coordinates. However, I found that it's measured every 0.2759 seconds for all races.
\n
I have checked other races and I think the time is 99% accurate. However, we don't know when the GPS tracker started tracking and at what time it stopped so in regard to that I have manipulated the track_us_index to 0.2759 to get approx time taken by an entity to complete the race.
", "rawMarkdown": "Hello Gabriel,\n\nYou are indeed correct about GPS coordinates. However, I found that it's measured every **0.2759** seconds for all races.\n\n##### Brief Summary \n\n\n\n* The final Time of the above Summary Visualization can be found [here](https://youtu.be/-OId_SqGt_I?t=2221).\n\nI have checked other races and I think the time is **99%** accurate. However, we don't know when the **GPS tracker** started tracking and at what time it stopped so in regard to that I have manipulated the **track_us_index** to **0.2759** to get approx time taken by an entity to complete the race.\n\n", "votes": 2 }, { "id": 1998797, "postDate": "2022-10-21T21:12:15.947Z", "content": "Thanks for the information.
\nPerhaps that difference could be due to the method used for the coordinates: Haversine, Vincenty, Karney or Pythagoras in flat space.
\nBut decidedly 0.25 seconds doesn't work with any method.
Thank you for sharing!
", "rawMarkdown": "Thank you for sharing!" }, { "id": 1999983, "postDate": "2022-10-22T23:41:06.943Z", "content": "Hi @gabrielbeguerie , I'd like to add that a race's timing starts when the first horse crosses the run-up_distance. The 1st horse to cross the run_up_distance will serve as the official starting point (timer) in USA instead of the starting gate. But for the trakus_index, the recording begins when the horses leave the gate…
\nregarding the distances, for me, I think everything is consistent (I merely had a tolerable inaccuracy of around 2 meter to 10 meter on the overall distance covered by each horse (I mean each finishing horse), but that's okay, it doesn't have a significant impact (…))
Thanks so much for the explanation. I'm not American and I don't know the racing customs there.
\nSo if the trakus_index every 0.25 seconds makes sense.
\nIt is a detail that I overlooked and was unaware of.
\nAt the end of my work I gave up taking the final time because I had many time errors that I couldn't find the solution to.
Thanks again.
", "rawMarkdown": "Thanks so much for the explanation. I'm not American and I don't know the racing customs there.\nSo if the trakus_index every 0.25 seconds makes sense.\nIt is a detail that I overlooked and was unaware of.\nAt the end of my work I gave up taking the final time because I had many time errors that I couldn't find the solution to.\n\nThanks again." }, { "id": 2000921, "postDate": "2022-10-23T16:11:02.920Z", "content": "I'm glad I was able to help.
", "rawMarkdown": "I'm glad I was able to help.", "votes": 1 }, { "id": 2001079, "postDate": "2022-10-23T18:50:50.923Z", "content": "There I uploaded a second version with your correction.
\nGreat, now it's perfect.
\nI leave you the link in case you want to see it.
\n\nThanks again.
", "rawMarkdown": "There I uploaded a second version with your correction.\n\nGreat, now it's perfect.\n\nI leave you the link in case you want to see it.\n\n[https://www.kaggle.com/code/gabrielbeguerie/data-derby-2022-analysis-positions-of-the-horses?scriptVersionId=108928119](https://www.kaggle.com/code/gabrielbeguerie/data-derby-2022-analysis-positions-of-the-horses?scriptVersionId=108928119)\n\nThanks again.\n\n", "votes": 2 }, { "id": 2004942, "postDate": "2022-10-26T15:55:50.233Z", "content": "Thank you for such crucial information @abireltaief I really appreciate it. This helps a lot.
", "rawMarkdown": "Thank you for such crucial information @abireltaief I really appreciate it. This helps a lot.", "votes": 1 }, { "id": 2005062, "postDate": "2022-10-26T17:53:55.880Z", "content": "@kushtrivedi14728 Thanks, I'm glad I was able to be of help.
", "rawMarkdown": "@kushtrivedi14728 Thanks, I'm glad I was able to be of help. " } ] }, { "id": 1925990, "postDate": "2022-09-04T13:28:53.267Z", "content": "Hi Addision,
\nWould you be able to provide the results of the race. Such as, which horse won and rest of the positions for each race please. As well as the starting gate number each horse please. Also, whether the starting gate's largest number always face the viewing stands please?
\nThanks & Best Regards.
\nSchroter Michael
This would also be helpful.
\nThank you for posting!
", "rawMarkdown": "This would also be helpful.\n\nThank you for posting!", "votes": 2 }, { "id": 1929020, "postDate": "2022-09-06T19:08:03.197Z", "content": "Post Position is always from the rail out, so the #1 horse is closest to the rail, #2 horses next out, #3 is third widest and so forth. The highest numbered post position is farthest from the rail.
", "rawMarkdown": "Post Position is always from the rail out, so the #1 horse is closest to the rail, #2 horses next out, #3 is third widest and so forth. The highest numbered post position is farthest from the rail.", "votes": 2 }, { "id": 1929034, "postDate": "2022-09-06T19:15:21.317Z", "content": "There are some notebooks that have figured out finish position, such as : https://www.kaggle.com/code/mlconsult/updated-get-finish-postion-by-horse-coordinates
\nHowever, the timing is only every 1/4 of a second, so in a close race, there could be discrepancies. Because of that, we are going to work to put up an order of finish list that links to the other csv's that were provided.
\nThis all does raise the question of whether winning is the best proxy for performance, but we will let you all work that out philosophically.
", "rawMarkdown": "There are some notebooks that have figured out finish position, such as : https://www.kaggle.com/code/mlconsult/updated-get-finish-postion-by-horse-coordinates\n\nHowever, the timing is only every 1/4 of a second, so in a close race, there could be discrepancies. Because of that, we are going to work to put up an order of finish list that links to the other csv's that were provided.\n\nThis all does raise the question of whether winning is the best proxy for performance, but we will let you all work that out philosophically.", "votes": 2 }, { "id": 1929460, "postDate": "2022-09-07T04:39:10.383Z", "content": "Hi Joe,
\nThanks alot. Would you be able to give the data for the starting gate number as well please.?
\nThanks & Best Regards
\nSchroter Michael
Hello Addison. In the competition description I see this: \" Along with stride data, we can now collect measures for heart rate, EKG, longitudinal movement, dorsal/ventral movement, medial/lateral deviation, total power and total landing vibration. \" . In the data description I only see position data for the horses, I do not see heart rate, longitudinal movement, dorsal/ventral movement, total power (well it can be estimated from the position data) or landing vibration. Is the data for this competition a subset of all the data that the customer collects? Thank you.
", "rawMarkdown": "Hello Addison. In the competition description I see this: \" Along with stride data, we can now collect measures for heart rate, EKG, longitudinal movement, dorsal/ventral movement, medial/lateral deviation, total power and total landing vibration. \" . In the data description I only see position data for the horses, I do not see heart rate, longitudinal movement, dorsal/ventral movement, total power (well it can be estimated from the position data) or landing vibration. Is the data for this competition a subset of all the data that the customer collects? Thank you. ", "votes": 5, "replies": [ { "id": 1899032, "postDate": "2022-08-15T02:57:09.827Z", "content": "Hi Adrian,
\nThe additional data collected/noted by the hosts is not included in this specific competition's dataset, but rather references the data advancements that the NYTHA/NYRA is making in the industry.
", "rawMarkdown": "Hi Adrian,\n\nThe additional data collected/noted by the hosts is not included in this specific competition's dataset, but rather references the data advancements that the NYTHA/NYRA is making in the industry.", "votes": 3 }, { "id": 1923927, "postDate": "2022-09-02T15:58:45.010Z", "content": "It would be nice if some of this data was provided. It would help to better understand the equine safety.
", "rawMarkdown": "It would be nice if some of this data was provided. It would help to better understand the equine safety." } ] }, { "id": 1955313, "postDate": "2022-09-25T23:08:06.773Z", "content": "Hello All,
\nHope you all will like my work.
\nBasically, I have analyzed one Race that happened at Saratoga on July 27, 2019. The actual video of the race can be found here.
\nQuestion: Why did Javier Castellano lose to Joel Rosario when he had a very good chance of winning?
\nAnswer: I would first emphasize that the following is just a sample visualization. Although, I have already made a more sophisticated animation and will release it before the competition's deadline.
![\"\"](\"https://i.postimg.cc/nVv03rmH/jvj.gif\")
In the visualization above, I am comparing and contrasting the race-tracking data of Javier Castellano and Joel Rosario in order to understand what led Joel Rosario to win and what led Javier Castellano to lose. Here, I have already plotted Joel Rosario's race-tracking data just for clarity and comparing purposes.
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8412510%2Fbf413bd197b67134b1e6166e154b24a0%2FSummary.png?generation=1664143904781760&alt=media\")
Typical Time Speed Curve
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8412510%2F8aba1981983c84423c7579bcb340233e%2FSpeed-time-curve.jpeg?generation=1665796320262949&alt=media\")
1 Joel Rosario Speed Time Curve
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8412510%2F0fcc90339d1cdc856a7eeed6ee149bcc%2Fjoel_rosario_sar.png?generation=1666450429793585&alt=media\")
2 Javier Castellano Speed Time Curve
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F8412510%2F423e9a1bde8e2f6a4d584349f9e959a2%2Fjavier_castellano_sar.png?generation=1666450416978191&alt=media\")
Also. I've used Stats and Physics to assess the Optimal Path for each race type and its category, defined each Jockey's Performance metric, created a Machine Learning Model for Drafting Strategy, and much more. Moreover, I would really appreciate comments and feedback on this. Thank you.
\nHi Addison, Joe, I wonder whether the rails are moved during the season or not. I read on some internet resources that sometimes the rails are moved because the ground close to them gets worse in time.
", "rawMarkdown": "Hi Addison, Joe, I wonder whether the rails are moved during the season or not. I read on some internet resources that sometimes the rails are moved because the ground close to them gets worse in time. ", "votes": 1, "replies": [ { "id": 1959032, "postDate": "2022-09-27T20:47:31.973Z", "content": "The rails on the turf courses are moved from time to time to reduce wear and tear on the course. I believe the information is available in the charts on Equibase.com
", "rawMarkdown": "The rails on the turf courses are moved from time to time to reduce wear and tear on the course. I believe the information is available in the charts on Equibase.com", "votes": 4 }, { "id": 1987423, "postDate": "2022-10-14T17:29:52.247Z", "content": "Hi Joe, is the finish line moved from time to time also? I guess no, because it has to be in the same place in front of the grandstands. I ask just to make sure. My analysis shows there is sometimes a mismatch between the \"advertised\" total distance (run-up + distance) and the actual distance ran by the horses all the way to the finish line. Not often, sometimes. The total ran distance is smaller, maybe the run-up distance was included already in the distance column or something like that.
", "rawMarkdown": "Hi Joe, is the finish line moved from time to time also? I guess no, because it has to be in the same place in front of the grandstands. I ask just to make sure. My analysis shows there is sometimes a mismatch between the \"advertised\" total distance (run-up + distance) and the actual distance ran by the horses all the way to the finish line. Not often, sometimes. The total ran distance is smaller, maybe the run-up distance was included already in the distance column or something like that. " } ] }, { "id": 1895950, "postDate": "2022-08-12T13:02:38.947Z", "content": "Hi, I'm a newcomer to this type of competition and I'm meeting it for the first time. Can you tell me how to submit for this type of contest?
\nThanks!
Hey there -
\nIn this competition you are tasked with creating a notebook submission that analyzes the data. You can submit your notebook for this competition per the instructions here.
", "rawMarkdown": "Hey there -\n\nIn this competition you are tasked with creating a notebook submission that analyzes the data. You can submit your notebook for this competition per the instructions [here](https://www.kaggle.com/competitions/big-data-derby-2022/overview/submission-instructions).", "votes": 1 }, { "id": 1896136, "postDate": "2022-08-12T16:02:22.203Z", "content": "Thank you for your answer.😄
", "rawMarkdown": "Thank you for your answer.😄" } ] }, { "id": 1906647, "postDate": "2022-08-20T05:02:47.373Z", "content": "Hi , Could you please elaborate on the data columns and what they mean please? Some of us are not familiar with the industry.
\nThanks & Best Regards
\nSchroter Michael
Go to the data tab and it breaks down all the data files.
\nhttps://www.kaggle.com/competitions/big-data-derby-2022/data
", "rawMarkdown": "Go to the data tab and it breaks down all the data files.\n\nhttps://www.kaggle.com/competitions/big-data-derby-2022/data", "votes": 1 } ] }, { "id": 1896699, "postDate": "2022-08-13T04:21:04.913Z", "content": "new to Kaggle. I asked if anyone wants a teammate on the leaderboard, apparently. But I can't find the leaderboard. So, does anyone want a teammate?? I'm a professional mathematician with several years of programming. I understand most of the ML and stats concepts from many coursera courses
", "rawMarkdown": "new to Kaggle. I asked if anyone wants a teammate on the leaderboard, apparently. But I can't find the leaderboard. So, does anyone want a teammate?? I'm a professional mathematician with several years of programming. I understand most of the ML and stats concepts from many coursera courses", "votes": 2, "replies": [ { "id": 1896967, "postDate": "2022-08-13T10:28:04.197Z", "content": "I think this is an individual competition
", "rawMarkdown": "I think this is an individual competition", "replies": [ { "id": 1897190, "postDate": "2022-08-13T14:13:10.597Z", "content": "oh ok thanks
", "rawMarkdown": "oh ok thanks" }, { "id": 1897191, "postDate": "2022-08-13T14:15:20.570Z", "content": "wait actually it says team size limit = 5
", "rawMarkdown": "wait actually it says team size limit = 5" } ] }, { "id": 1899034, "postDate": "2022-08-15T03:00:38.953Z", "content": "Teams are allowed in this competition - please ensure that all team members are added as a collaborator to any notebooks you submit for final judging.
", "rawMarkdown": "Teams are allowed in this competition - please ensure that all team members are added as a collaborator to any notebooks you submit for final judging. " }, { "id": 1907180, "postDate": "2022-08-20T14:59:46.960Z", "content": "HI there, if you still have a spot left let me in. I am currently pursuing masters in data analytics and machine learning and i am familiar with this field from past 3 years
", "rawMarkdown": "HI there, if you still have a spot left let me in. I am currently pursuing masters in data analytics and machine learning and i am familiar with this field from past 3 years" }, { "id": 1907457, "postDate": "2022-08-20T19:49:36.807Z", "content": "I am looking for a teammate too, is there any spot left in your team? I am a Data Science grad student, I have 5years of Analytics experience at Uber and I understand ML concepts and most of my coursework is in Stats and ML. I'd like to solve real-world problems and practice what I learn in school on real data. Let me know if you or anyone else here wants to collaborate.
", "rawMarkdown": "I am looking for a teammate too, is there any spot left in your team? I am a Data Science grad student, I have 5years of Analytics experience at Uber and I understand ML concepts and most of my coursework is in Stats and ML. I'd like to solve real-world problems and practice what I learn in school on real data. Let me know if you or anyone else here wants to collaborate." }, { "id": 1915033, "postDate": "2022-08-26T16:07:36.693Z", "content": "I'm looking for a teammate as well, if you are still looking. I am a data analyst though I've only just started and I'm generally familiar with ML concepts. Hoping to apply what I know and I'd love to collaborate with someone while doing so.
", "rawMarkdown": "I'm looking for a teammate as well, if you are still looking. I am a data analyst though I've only just started and I'm generally familiar with ML concepts. Hoping to apply what I know and I'd love to collaborate with someone while doing so." } ] }, { "id": 2026040, "postDate": "2022-11-11T17:43:08.470Z", "content": "Hi i am wondering if I add something to my Notebook, it will not considered anymore?
", "rawMarkdown": "Hi i am wondering if I add something to my Notebook, it will not considered anymore?" }, { "id": 2024794, "postDate": "2022-11-10T19:21:46.283Z", "content": "How do i convert the post-time column from str to a time datatype?
", "rawMarkdown": "How do i convert the post-time column from str to a time datatype?" }, { "id": 1955246, "postDate": "2022-09-25T21:24:59.783Z", "content": "I have a question to Kaggle Staff and Competition Hosts.
\nIt was only after a month of competition that I decided to look at the rules. I saw a terrible thing for myself.. \"Competitions are open to residents of the United States and worldwide, except that if you are a resident of Crimea, so-called Donetsk People's Republic (DNR) or Luhansk People's Republic (LNR), Cuba, Iran, Syria, North Korea, or are subject to U.S. export controls or sanctions, you may not enter the Competition.\" I'm living in Russia, but there are no personal sanctions on me and I don't support all this horror that is happening now. And do I understand correctly that I have no right to participate?…
", "rawMarkdown": "I have a question to Kaggle Staff and Competition Hosts.\n\nIt was only after a month of competition that I decided to look at the rules. I saw a terrible thing for myself.. \"Competitions are open to residents of the United States and worldwide, except that if you are a resident of Crimea, so-called Donetsk People's Republic (DNR) or Luhansk People's Republic (LNR), Cuba, Iran, Syria, North Korea, or are subject to U.S. export controls or sanctions, you may not enter the Competition.\" I'm living in Russia, but there are no personal sanctions on me and I **don't** support all this horror that is happening now. And do I understand correctly that I have no right to participate?...", "replies": [ { "id": 1956895, "postDate": "2022-09-26T16:39:56.273Z", "content": "Hi Dmitry,
\nWe cannot interpret competition rules for you, but per my understanding there are no sanctions for entering a competition simply for living in Russia, as long as you are not living in Crimea, so-called Donetsk People's Republic (DNR) or Luhansk People's Republic (LNR).
", "rawMarkdown": "Hi Dmitry,\n\nWe cannot interpret competition rules for you, but per my understanding there are no sanctions for entering a competition simply for living in Russia, as long as you are not living in Crimea, so-called Donetsk People's Republic (DNR) or Luhansk People's Republic (LNR).", "votes": 1 }, { "id": 1957100, "postDate": "2022-09-26T18:32:11.597Z", "content": "@addisonhoward Thank you very much for the answer!
", "rawMarkdown": "@addisonhoward Thank you very much for the answer! " } ] }, { "id": 1943600, "postDate": "2022-09-17T16:25:43.510Z", "content": "Hi Addision,
\nBy filtering the data ( ../input/big-data-derby-2022/nyra_2019_complete.csv) , It could be found that some missing trakus_index for a particular horse in a race (For an example, race date - 2019-04-20 / race - 7/ program number - 6 / trakus index between 7 to 216). That kind of missing range might be an issue for the accuracy of the velocity of the horse because the distance between adjacent trakus_index points might not be the actual distance that the horse traveled.
\nI am hoping for your advice on this issue, please.
\nBest Regards
\nSameera Sandaruwan
", "rawMarkdown": "Hi Addision,\n\nBy filtering the data *( ../input/big-data-derby-2022/nyra_2019_complete.csv)* , It could be found that some missing **trakus_index** for a particular horse in a race (For an example, race date - 2019-04-20 / race - 7/ program number - 6 / trakus index between 7 to 216). That kind of missing range might be an issue for the accuracy of the velocity of the horse because the distance between adjacent trakus_index points might not be the actual distance that the horse traveled. \n\n\nI am hoping for your advice on this issue, please.\n\nBest Regards\n\nSameera Sandaruwan", "replies": [ { "id": 1949714, "postDate": "2022-09-21T20:42:06.133Z", "content": "Hi Sameera,
\nWe're not seeing the same issue on our end. We see 388 unique trakus indexes on our end - can you try re-downloading the data and re-processing?
", "rawMarkdown": "Hi Sameera,\n\nWe're not seeing the same issue on our end. We see 388 unique trakus indexes on our end - can you try re-downloading the data and re-processing?" }, { "id": 1951027, "postDate": "2022-09-22T18:02:26.817Z", "content": "Hi Addison,
\nThanks and I'll try again to sort it out.
How confident are we really about the Trakus time intervals? I know in the data description it says \"from what we can tell\" it's collected every 0.25 seconds. But from what I'm seeing, that seems consistently short. I posted with some more detail here: https://www.kaggle.com/competitions/big-data-derby-2022/discussion/352553
\nSome of the total recording times as derived from a 0.25s trakus interval are actually less than the winner's time. Unless I'm misunderstanding something.
", "rawMarkdown": "How confident are we really about the Trakus time intervals? I know in the data description it says \"from what we can tell\" it's collected every 0.25 seconds. But from what I'm seeing, that seems consistently short. I posted with some more detail here: https://www.kaggle.com/competitions/big-data-derby-2022/discussion/352553\n\nSome of the total recording times as derived from a 0.25s trakus interval are actually less than the winner's time. Unless I'm misunderstanding something. " }, { "id": 1931505, "postDate": "2022-09-08T18:18:52.647Z", "content": "Please elaborate on the following:
\nodds - Odds to win the race passed as an integer. Divide by 100 to derive the odds to 1. Example - 1280 would be 12.8-1. Odds of betting or some other odds of winning? If this is the betting odds, in your examples, it means that if this horse wins, I will get 12.8 times my money?
trakus_index - The common collection of point of the lat / long of the horse in the race passed as an integer. From what we can tell, it's collected every 0.25 seconds. It is still unclear what does trakus_index column stand for
Not able to decipher data column names
", "rawMarkdown": "Not able to decipher data column names ", "replies": [ { "id": 1915860, "postDate": "2022-08-27T12:39:16.563Z", "content": "Here is a link to the data that explains the columns
\nhttps://www.kaggle.com/competitions/big-data-derby-2022/data
", "rawMarkdown": "Here is a link to the data that explains the columns\n\nhttps://www.kaggle.com/competitions/big-data-derby-2022/data" }, { "id": 1915918, "postDate": "2022-08-27T13:27:52.647Z", "content": "thanks 👍
", "rawMarkdown": " thanks 👍" } ] }, { "id": 1911686, "postDate": "2022-08-24T08:45:03.393Z", "content": "new to kaggle. I wanna ask how can I summit my result of the competition?? Because I can't find the leaderboard.
", "rawMarkdown": "new to kaggle. I wanna ask how can I summit my result of the competition?? Because I can't find the leaderboard." }, { "id": 1910550, "postDate": "2022-08-23T14:02:37.457Z", "content": "Hello,
\nIs the Trakus data reliable?
\nI don't mean to be rude, and perhaps it's entirely my error, but I've looked at one race in particular (the Whitney Stakes, run at Saratoga on Aug 3, 2019 - race 9 - viewable on Youtube), and the Trakus data doesn't seem to accurately reflect what you see actually happen in the race video.
\nIn the Whitney, the horses break in a pretty good line, and the horse in gate 1 (#1) gets away on terms and holds the inside until the first turn.
\nHowever, if you look at the Trakus data for the race, and concentrate say, on the first 10 trakus_index values for each horse, it appears to show horse #2 holding the inside for the first few strides, with horse 1 being 'between horses' (which definitely doesn't happen in the race).
\nI visualized it using this code:-
\nsns.scatterplot(data=tracking_df, x='latitude', y='longitude', hue='program_number')
\n..where tracking_df is a subset of the tracking dataset restricted to the Whitney Stakes, and cut down further to show the first handful of trakus_index values for each horse.
\nAs it seems we are being asked (possibly) to investigate 'drafting', and the tactical lines taken by jockeys in the race, if the path a horse takes on the track (straight out of the gate for instance) isn't being accurately reflected in the locational data, that's going to be an impossible task.
\nI'm happy to be shown I've got something wrong here, but at the moment, I can't see what it is.
\nThanks
\nGerry.
Gerry
\nWe believe the vast majority of the traking data to be accurate. However, there will be occasional instances when a device has failed or produces inaccurate measure. We believe that number to be a very small minority.
Thanks for the reply Joe.
\nI have a couple more questions to ask about the Trakus data - you might not want to answer, or rather, the answer might be \"work it out for yourself\", but I'll ask anyway…
\nAre the transponders synchronised across all horses in a race? - ie. for a particular race, is the location-data for trakus_index=1 (say) taken at the same moment in time for all the horses?
It appears that different tracks (or perhaps just different races) have a different strategy for 'shutting down' the Trakus data - or at least, leaving a last recording for a race. I looked at a Belmont race, and the Trakus data continued recording after the horses crossed the line, and appears to continue as they walk back, whereas in the Whitney (Saratoga), and a race at Aqueduct, the data seemed to shut down as the winner crossed the line (or very near to it). Question - is that the case? - is there a 'track policy' for terminating Trakus-recording for a race?
I need to do Some homework both on the specs and as you surmised whether we are supposed to release or not.
", "rawMarkdown": "I need to do Some homework both on the specs and as you surmised whether we are supposed to release or not. " }, { "id": 1913025, "postDate": "2022-08-25T05:18:44.493Z", "content": "Hi Joe,
\nCould you please give some indication as to which type of information each columns represent please?
\nLike a better elaboration on column data.
\nThanks & Best Regards
\nSchroter Michael
Hey Guys, I am new to Kaggle and Analytics in general . This is probably not for my level since I am still learning all the basics of data analytics but I would like to learn more about the process. So if there is anyone who like to add me to their team I would be really grateful. For me it will be a really good learning opportunity.
\nP.S: I have zero experience and I am learning (beginner level)
Thank you Kaggle for providing the wonderful opportunity for us to learn
", "rawMarkdown": "Thank you Kaggle for providing the wonderful opportunity for us to learn " }, { "id": 1895354, "postDate": "2022-08-12T04:57:34.917Z", "content": "So the ranking will be displayed after the end of the competition?
", "rawMarkdown": "So the ranking will be displayed after the end of the competition?", "replies": [ { "id": 1896114, "postDate": "2022-08-12T15:38:04.433Z", "content": "Hi Rohit,
\nAs the competition is not using a standard evaluation metric (where a score is immediately provided for each submission), the top winners will be announced at the end of the competition as selected by competition hosts.
", "rawMarkdown": "Hi Rohit,\n\nAs the competition is not using a standard evaluation metric (where a score is immediately provided for each submission), the top winners will be announced at the end of the competition as selected by competition hosts. " } ] }, { "id": 1906546, "postDate": "2022-08-20T01:17:27.770Z", "rawMarkdown": "", "isDeleted": true, "replies": [ { "id": 1907403, "postDate": "2022-08-20T18:35:15.657Z", "content": "@pondicherrya @pbarendse
\nif you are interested we can work together to solve this problem.
\nIf yes please let me know as soon as possible.
\nBest regards
Thank you for sharing :)
", "rawMarkdown": "Thank you for sharing :) " }, { "id": 1995224, "postDate": "2022-10-19T14:28:50.750Z", "content": "thanks a lot
", "rawMarkdown": "thanks a lot" } ], "comments": [ { "id": 1998569, "author_name": "Gabriel Beguerie", "author_url": "", "post_date": "2022-10-21T17:41:00.733000", "content": "Hello Addison:
\nWorking on this project with a mathematical model that does not depend on distances but only on coordinates, I find that the GPS signal is not every 0.25 seconds but every 0.2825 seconds in all the races.
\nI leave the information here so that the other competitors can take advantage of it, otherwise the speeds and distances will not be correct.
\nI have verified this by watching the races on YouTube and corroborating the times of my model with the televised race.
", "votes": 5, "replies": [ { "id": 1998699, "author_name": "Kush Trivedi", "author_url": "", "post_date": "2022-10-21T19:23:34.827000", "content": "Hello Gabriel,
\nYou are indeed correct about GPS coordinates. However, I found that it's measured every 0.2759 seconds for all races.
\n
I have checked other races and I think the time is 99% accurate. However, we don't know when the GPS tracker started tracking and at what time it stopped so in regard to that I have manipulated the track_us_index to 0.2759 to get approx time taken by an entity to complete the race.
", "votes": 2, "replies": [] }, { "id": 1998797, "author_name": "Gabriel Beguerie", "author_url": "", "post_date": "2022-10-21T21:12:15.947000", "content": "Thanks for the information.
\nPerhaps that difference could be due to the method used for the coordinates: Haversine, Vincenty, Karney or Pythagoras in flat space.
\nBut decidedly 0.25 seconds doesn't work with any method.
Thank you for sharing!
", "votes": 0, "replies": [] }, { "id": 1999983, "author_name": "", "author_url": "", "post_date": "2022-10-22T23:41:06.943000", "content": "Hi @gabrielbeguerie , I'd like to add that a race's timing starts when the first horse crosses the run-up_distance. The 1st horse to cross the run_up_distance will serve as the official starting point (timer) in USA instead of the starting gate. But for the trakus_index, the recording begins when the horses leave the gate…
\nregarding the distances, for me, I think everything is consistent (I merely had a tolerable inaccuracy of around 2 meter to 10 meter on the overall distance covered by each horse (I mean each finishing horse), but that's okay, it doesn't have a significant impact (…))
Thanks so much for the explanation. I'm not American and I don't know the racing customs there.
\nSo if the trakus_index every 0.25 seconds makes sense.
\nIt is a detail that I overlooked and was unaware of.
\nAt the end of my work I gave up taking the final time because I had many time errors that I couldn't find the solution to.
Thanks again.
", "votes": 0, "replies": [] }, { "id": 2000921, "author_name": "", "author_url": "", "post_date": "2022-10-23T16:11:02.920000", "content": "I'm glad I was able to help.
", "votes": 1, "replies": [] }, { "id": 2001079, "author_name": "Gabriel Beguerie", "author_url": "", "post_date": "2022-10-23T18:50:50.923000", "content": "There I uploaded a second version with your correction.
\nGreat, now it's perfect.
\nI leave you the link in case you want to see it.
\n\nThanks again.
", "votes": 2, "replies": [] }, { "id": 2004942, "author_name": "Kush Trivedi", "author_url": "", "post_date": "2022-10-26T15:55:50.233000", "content": "Thank you for such crucial information @abireltaief I really appreciate it. This helps a lot.
", "votes": 1, "replies": [] }, { "id": 2005062, "author_name": "", "author_url": "", "post_date": "2022-10-26T17:53:55.880000", "content": "@kushtrivedi14728 Thanks, I'm glad I was able to be of help.
", "votes": 0, "replies": [] } ] }, { "id": 1925990, "author_name": "Schroter", "author_url": "", "post_date": "2022-09-04T13:28:53.267000", "content": "Hi Addision,
\nWould you be able to provide the results of the race. Such as, which horse won and rest of the positions for each race please. As well as the starting gate number each horse please. Also, whether the starting gate's largest number always face the viewing stands please?
\nThanks & Best Regards.
\nSchroter Michael
This would also be helpful.
\nThank you for posting!
", "votes": 2, "replies": [] }, { "id": 1929020, "author_name": "Joe Appelbaum", "author_url": "", "post_date": "2022-09-06T19:08:03.197000", "content": "Post Position is always from the rail out, so the #1 horse is closest to the rail, #2 horses next out, #3 is third widest and so forth. The highest numbered post position is farthest from the rail.
", "votes": 2, "replies": [] }, { "id": 1929034, "author_name": "Joe Appelbaum", "author_url": "", "post_date": "2022-09-06T19:15:21.317000", "content": "There are some notebooks that have figured out finish position, such as : https://www.kaggle.com/code/mlconsult/updated-get-finish-postion-by-horse-coordinates
\nHowever, the timing is only every 1/4 of a second, so in a close race, there could be discrepancies. Because of that, we are going to work to put up an order of finish list that links to the other csv's that were provided.
\nThis all does raise the question of whether winning is the best proxy for performance, but we will let you all work that out philosophically.
", "votes": 2, "replies": [] }, { "id": 1929460, "author_name": "Schroter", "author_url": "", "post_date": "2022-09-07T04:39:10.383000", "content": "Hi Joe,
\nThanks alot. Would you be able to give the data for the starting gate number as well please.?
\nThanks & Best Regards
\nSchroter Michael
Hello Addison. In the competition description I see this: \" Along with stride data, we can now collect measures for heart rate, EKG, longitudinal movement, dorsal/ventral movement, medial/lateral deviation, total power and total landing vibration. \" . In the data description I only see position data for the horses, I do not see heart rate, longitudinal movement, dorsal/ventral movement, total power (well it can be estimated from the position data) or landing vibration. Is the data for this competition a subset of all the data that the customer collects? Thank you.
", "votes": 5, "replies": [ { "id": 1899032, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-08-15T02:57:09.827000", "content": "Hi Adrian,
\nThe additional data collected/noted by the hosts is not included in this specific competition's dataset, but rather references the data advancements that the NYTHA/NYRA is making in the industry.
", "votes": 3, "replies": [] }, { "id": 1923927, "author_name": "Daniel Fernandez", "author_url": "", "post_date": "2022-09-02T15:58:45.010000", "content": "It would be nice if some of this data was provided. It would help to better understand the equine safety.
", "votes": 0, "replies": [] } ] }, { "id": 1955313, "author_name": "Kush Trivedi", "author_url": "", "post_date": "2022-09-25T23:08:06.773000", "content": "Hello All,
\nHope you all will like my work.
\nBasically, I have analyzed one Race that happened at Saratoga on July 27, 2019. The actual video of the race can be found here.
\nQuestion: Why did Javier Castellano lose to Joel Rosario when he had a very good chance of winning?
\nAnswer: I would first emphasize that the following is just a sample visualization. Although, I have already made a more sophisticated animation and will release it before the competition's deadline.
In the visualization above, I am comparing and contrasting the race-tracking data of Javier Castellano and Joel Rosario in order to understand what led Joel Rosario to win and what led Javier Castellano to lose. Here, I have already plotted Joel Rosario's race-tracking data just for clarity and comparing purposes.
\n
Typical Time Speed Curve
\n
1 Joel Rosario Speed Time Curve
\n
2 Javier Castellano Speed Time Curve
\n
Also. I've used Stats and Physics to assess the Optimal Path for each race type and its category, defined each Jockey's Performance metric, created a Machine Learning Model for Drafting Strategy, and much more. Moreover, I would really appreciate comments and feedback on this. Thank you.
\nHi Addison, Joe, I wonder whether the rails are moved during the season or not. I read on some internet resources that sometimes the rails are moved because the ground close to them gets worse in time.
", "votes": 1, "replies": [ { "id": 1959032, "author_name": "Joe Appelbaum", "author_url": "", "post_date": "2022-09-27T20:47:31.973000", "content": "The rails on the turf courses are moved from time to time to reduce wear and tear on the course. I believe the information is available in the charts on Equibase.com
", "votes": 4, "replies": [] }, { "id": 1987423, "author_name": "Adrian Muresan", "author_url": "", "post_date": "2022-10-14T17:29:52.247000", "content": "Hi Joe, is the finish line moved from time to time also? I guess no, because it has to be in the same place in front of the grandstands. I ask just to make sure. My analysis shows there is sometimes a mismatch between the \"advertised\" total distance (run-up + distance) and the actual distance ran by the horses all the way to the finish line. Not often, sometimes. The total ran distance is smaller, maybe the run-up distance was included already in the distance column or something like that.
", "votes": 0, "replies": [] } ] }, { "id": 1895950, "author_name": "", "author_url": "", "post_date": "2022-08-12T13:02:38.947000", "content": "Hi, I'm a newcomer to this type of competition and I'm meeting it for the first time. Can you tell me how to submit for this type of contest?
\nThanks!
Hey there -
\nIn this competition you are tasked with creating a notebook submission that analyzes the data. You can submit your notebook for this competition per the instructions here.
", "votes": 1, "replies": [] }, { "id": 1896136, "author_name": "", "author_url": "", "post_date": "2022-08-12T16:02:22.203000", "content": "Thank you for your answer.😄
", "votes": 0, "replies": [] } ] }, { "id": 1906647, "author_name": "Schroter", "author_url": "", "post_date": "2022-08-20T05:02:47.373000", "content": "Hi , Could you please elaborate on the data columns and what they mean please? Some of us are not familiar with the industry.
\nThanks & Best Regards
\nSchroter Michael
Go to the data tab and it breaks down all the data files.
\nhttps://www.kaggle.com/competitions/big-data-derby-2022/data
", "votes": 1, "replies": [] } ] }, { "id": 1896699, "author_name": "peterb", "author_url": "", "post_date": "2022-08-13T04:21:04.913000", "content": "new to Kaggle. I asked if anyone wants a teammate on the leaderboard, apparently. But I can't find the leaderboard. So, does anyone want a teammate?? I'm a professional mathematician with several years of programming. I understand most of the ML and stats concepts from many coursera courses
", "votes": 2, "replies": [ { "id": 1896967, "author_name": "Foo Chee Chuan", "author_url": "", "post_date": "2022-08-13T10:28:04.197000", "content": "I think this is an individual competition
", "votes": 0, "replies": [ { "id": 1897190, "author_name": "peterb", "author_url": "", "post_date": "2022-08-13T14:13:10.597000", "content": "oh ok thanks
", "votes": 0, "replies": [] }, { "id": 1897191, "author_name": "peterb", "author_url": "", "post_date": "2022-08-13T14:15:20.570000", "content": "wait actually it says team size limit = 5
", "votes": 0, "replies": [] } ] }, { "id": 1899034, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-08-15T03:00:38.953000", "content": "Teams are allowed in this competition - please ensure that all team members are added as a collaborator to any notebooks you submit for final judging.
", "votes": 0, "replies": [] }, { "id": 1907180, "author_name": "Maloof Bashir", "author_url": "", "post_date": "2022-08-20T14:59:46.960000", "content": "HI there, if you still have a spot left let me in. I am currently pursuing masters in data analytics and machine learning and i am familiar with this field from past 3 years
", "votes": 0, "replies": [] }, { "id": 1907457, "author_name": "pondicherrya", "author_url": "", "post_date": "2022-08-20T19:49:36.807000", "content": "I am looking for a teammate too, is there any spot left in your team? I am a Data Science grad student, I have 5years of Analytics experience at Uber and I understand ML concepts and most of my coursework is in Stats and ML. I'd like to solve real-world problems and practice what I learn in school on real data. Let me know if you or anyone else here wants to collaborate.
", "votes": 0, "replies": [] }, { "id": 1915033, "author_name": "Shreyansh Gupta", "author_url": "", "post_date": "2022-08-26T16:07:36.693000", "content": "I'm looking for a teammate as well, if you are still looking. I am a data analyst though I've only just started and I'm generally familiar with ML concepts. Hoping to apply what I know and I'd love to collaborate with someone while doing so.
", "votes": 0, "replies": [] } ] }, { "id": 2026040, "author_name": "naghmeh Shahverdi", "author_url": "", "post_date": "2022-11-11T17:43:08.470000", "content": "Hi i am wondering if I add something to my Notebook, it will not considered anymore?
", "votes": 0, "replies": [] }, { "id": 2024794, "author_name": "Pelummi", "author_url": "", "post_date": "2022-11-10T19:21:46.283000", "content": "How do i convert the post-time column from str to a time datatype?
", "votes": 0, "replies": [] }, { "id": 1955246, "author_name": "Dmitry Uarov", "author_url": "", "post_date": "2022-09-25T21:24:59.783000", "content": "I have a question to Kaggle Staff and Competition Hosts.
\nIt was only after a month of competition that I decided to look at the rules. I saw a terrible thing for myself.. \"Competitions are open to residents of the United States and worldwide, except that if you are a resident of Crimea, so-called Donetsk People's Republic (DNR) or Luhansk People's Republic (LNR), Cuba, Iran, Syria, North Korea, or are subject to U.S. export controls or sanctions, you may not enter the Competition.\" I'm living in Russia, but there are no personal sanctions on me and I don't support all this horror that is happening now. And do I understand correctly that I have no right to participate?…
", "votes": 0, "replies": [ { "id": 1956895, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-09-26T16:39:56.273000", "content": "Hi Dmitry,
\nWe cannot interpret competition rules for you, but per my understanding there are no sanctions for entering a competition simply for living in Russia, as long as you are not living in Crimea, so-called Donetsk People's Republic (DNR) or Luhansk People's Republic (LNR).
", "votes": 1, "replies": [] }, { "id": 1957100, "author_name": "Dmitry Uarov", "author_url": "", "post_date": "2022-09-26T18:32:11.597000", "content": "@addisonhoward Thank you very much for the answer!
", "votes": 0, "replies": [] } ] }, { "id": 1943600, "author_name": "sameera sandaruwan460", "author_url": "", "post_date": "2022-09-17T16:25:43.510000", "content": "Hi Addision,
\nBy filtering the data ( ../input/big-data-derby-2022/nyra_2019_complete.csv) , It could be found that some missing trakus_index for a particular horse in a race (For an example, race date - 2019-04-20 / race - 7/ program number - 6 / trakus index between 7 to 216). That kind of missing range might be an issue for the accuracy of the velocity of the horse because the distance between adjacent trakus_index points might not be the actual distance that the horse traveled.
\nI am hoping for your advice on this issue, please.
\nBest Regards
\nSameera Sandaruwan
", "votes": 0, "replies": [ { "id": 1949714, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-09-21T20:42:06.133000", "content": "Hi Sameera,
\nWe're not seeing the same issue on our end. We see 388 unique trakus indexes on our end - can you try re-downloading the data and re-processing?
", "votes": 0, "replies": [] }, { "id": 1951027, "author_name": "sameera sandaruwan460", "author_url": "", "post_date": "2022-09-22T18:02:26.817000", "content": "Hi Addison,
\nThanks and I'll try again to sort it out.
How confident are we really about the Trakus time intervals? I know in the data description it says \"from what we can tell\" it's collected every 0.25 seconds. But from what I'm seeing, that seems consistently short. I posted with some more detail here: https://www.kaggle.com/competitions/big-data-derby-2022/discussion/352553
\nSome of the total recording times as derived from a 0.25s trakus interval are actually less than the winner's time. Unless I'm misunderstanding something.
", "votes": 0, "replies": [] }, { "id": 1931505, "author_name": "etalab", "author_url": "", "post_date": "2022-09-08T18:18:52.647000", "content": "Please elaborate on the following:
\nodds - Odds to win the race passed as an integer. Divide by 100 to derive the odds to 1. Example - 1280 would be 12.8-1. Odds of betting or some other odds of winning? If this is the betting odds, in your examples, it means that if this horse wins, I will get 12.8 times my money?
trakus_index - The common collection of point of the lat / long of the horse in the race passed as an integer. From what we can tell, it's collected every 0.25 seconds. It is still unclear what does trakus_index column stand for
Not able to decipher data column names
", "votes": 0, "replies": [ { "id": 1915860, "author_name": "Ken Miller", "author_url": "", "post_date": "2022-08-27T12:39:16.563000", "content": "Here is a link to the data that explains the columns
\nhttps://www.kaggle.com/competitions/big-data-derby-2022/data
", "votes": 0, "replies": [] }, { "id": 1915918, "author_name": "Maloof Bashir", "author_url": "", "post_date": "2022-08-27T13:27:52.647000", "content": "thanks 👍
", "votes": 0, "replies": [] } ] }, { "id": 1911686, "author_name": "y1nghA0", "author_url": "", "post_date": "2022-08-24T08:45:03.393000", "content": "new to kaggle. I wanna ask how can I summit my result of the competition?? Because I can't find the leaderboard.
", "votes": 0, "replies": [] }, { "id": 1910550, "author_name": "Gerald Deeprose", "author_url": "", "post_date": "2022-08-23T14:02:37.457000", "content": "Hello,
\nIs the Trakus data reliable?
\nI don't mean to be rude, and perhaps it's entirely my error, but I've looked at one race in particular (the Whitney Stakes, run at Saratoga on Aug 3, 2019 - race 9 - viewable on Youtube), and the Trakus data doesn't seem to accurately reflect what you see actually happen in the race video.
\nIn the Whitney, the horses break in a pretty good line, and the horse in gate 1 (#1) gets away on terms and holds the inside until the first turn.
\nHowever, if you look at the Trakus data for the race, and concentrate say, on the first 10 trakus_index values for each horse, it appears to show horse #2 holding the inside for the first few strides, with horse 1 being 'between horses' (which definitely doesn't happen in the race).
\nI visualized it using this code:-
\nsns.scatterplot(data=tracking_df, x='latitude', y='longitude', hue='program_number')
\n..where tracking_df is a subset of the tracking dataset restricted to the Whitney Stakes, and cut down further to show the first handful of trakus_index values for each horse.
\nAs it seems we are being asked (possibly) to investigate 'drafting', and the tactical lines taken by jockeys in the race, if the path a horse takes on the track (straight out of the gate for instance) isn't being accurately reflected in the locational data, that's going to be an impossible task.
\nI'm happy to be shown I've got something wrong here, but at the moment, I can't see what it is.
\nThanks
\nGerry.
Gerry
\nWe believe the vast majority of the traking data to be accurate. However, there will be occasional instances when a device has failed or produces inaccurate measure. We believe that number to be a very small minority.
Thanks for the reply Joe.
\nI have a couple more questions to ask about the Trakus data - you might not want to answer, or rather, the answer might be \"work it out for yourself\", but I'll ask anyway…
\nAre the transponders synchronised across all horses in a race? - ie. for a particular race, is the location-data for trakus_index=1 (say) taken at the same moment in time for all the horses?
It appears that different tracks (or perhaps just different races) have a different strategy for 'shutting down' the Trakus data - or at least, leaving a last recording for a race. I looked at a Belmont race, and the Trakus data continued recording after the horses crossed the line, and appears to continue as they walk back, whereas in the Whitney (Saratoga), and a race at Aqueduct, the data seemed to shut down as the winner crossed the line (or very near to it). Question - is that the case? - is there a 'track policy' for terminating Trakus-recording for a race?
I need to do Some homework both on the specs and as you surmised whether we are supposed to release or not.
", "votes": 0, "replies": [] }, { "id": 1913025, "author_name": "Schroter", "author_url": "", "post_date": "2022-08-25T05:18:44.493000", "content": "Hi Joe,
\nCould you please give some indication as to which type of information each columns represent please?
\nLike a better elaboration on column data.
\nThanks & Best Regards
\nSchroter Michael
Hey Guys, I am new to Kaggle and Analytics in general . This is probably not for my level since I am still learning all the basics of data analytics but I would like to learn more about the process. So if there is anyone who like to add me to their team I would be really grateful. For me it will be a really good learning opportunity.
\nP.S: I have zero experience and I am learning (beginner level)
Thank you Kaggle for providing the wonderful opportunity for us to learn
", "votes": 0, "replies": [] }, { "id": 1895354, "author_name": "Rohit Shirudkar", "author_url": "", "post_date": "2022-08-12T04:57:34.917000", "content": "So the ranking will be displayed after the end of the competition?
", "votes": 0, "replies": [ { "id": 1896114, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-08-12T15:38:04.433000", "content": "Hi Rohit,
\nAs the competition is not using a standard evaluation metric (where a score is immediately provided for each submission), the top winners will be announced at the end of the competition as selected by competition hosts.
", "votes": 0, "replies": [] } ] }, { "id": 1906546, "author_name": "", "author_url": "", "post_date": "2022-08-20T01:17:27.770000", "content": "", "votes": 0, "replies": [ { "id": 1907403, "author_name": "Mohamed LABIDI", "author_url": "", "post_date": "2022-08-20T18:35:15.657000", "content": "@pondicherrya @pbarendse
\nif you are interested we can work together to solve this problem.
\nIf yes please let me know as soon as possible.
\nBest regards
Thank you for sharing :)
", "votes": 0, "replies": [] }, { "id": 1995224, "author_name": "Martin Hernandez", "author_url": "", "post_date": "2022-10-19T14:28:50.750000", "content": "thanks a lot
", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "1895000": "Welcome to the Big Data Derby! We're so excited to once again expand our Analytics Competition offerings to the Kaggle community to explore unique competition ideas that tell compelling stories. \n\nIf this is your first analytics competition, I'd encourage you to see Kaggle Grandmaster John Miller's [excellent write-up](https://www.kaggle.com/jpmiller/creating-a-good-analytics-report) on tips and tricks for making a great analytics submission, as well as the top solutions of other analytics competitions.\n\nThe types of questions the NYTHRA/NYRA is looking to see examined are provided on the [Evaluation](https://www.kaggle.com/competitions/big-data-derby-2022/overview/evaluation) page. Please note that list is not comprehensive, nor is it meant to be a guide for you to try and specifically answer (or address all the questions). \n\nAs a reminder - Submissions must be made public at the competition deadline to be considered eligible for scoring (The NYTHA/NYRA has to be able to actually see your hard work!).\n\nBest of luck, and happy modeling!", "1998569": "Hello Addison:\n\nWorking on this project with a mathematical model that does not depend on distances but only on coordinates, I find that the GPS signal is not every 0.25 seconds but every 0.2825 seconds in all the races.\n\nI leave the information here so that the other competitors can take advantage of it, otherwise the speeds and distances will not be correct.\n\nI have verified this by watching the races on YouTube and corroborating the times of my model with the televised race.\n", "1925990": "Hi Addision,\n\nWould you be able to provide the results of the race. Such as, which horse won and rest of the positions for each race please. As well as the starting gate number each horse please. Also, whether the starting gate's largest number always face the viewing stands please?\nThanks & Best Regards.\nSchroter Michael", "1898723": "Hello Addison. In the competition description I see this: \" Along with stride data, we can now collect measures for heart rate, EKG, longitudinal movement, dorsal/ventral movement, medial/lateral deviation, total power and total landing vibration. \" . In the data description I only see position data for the horses, I do not see heart rate, longitudinal movement, dorsal/ventral movement, total power (well it can be estimated from the position data) or landing vibration. Is the data for this competition a subset of all the data that the customer collects? Thank you. ", "1955313": "Hello All,\n\nHope you all will like my work.\n\nBasically, I have analyzed one Race that happened at Saratoga on July 27, 2019. The actual video of the race can be found [here](https://youtu.be/-OId_SqGt_I?t=2141). \n\n**Question:** Why did **Javier Castellano lose** to **Joel Rosario** when he had a very good chance of winning?\n**Answer:** I would first emphasize that the following is just a sample visualization. Although, I have already made a more sophisticated animation and will release it before the competition's deadline. \n\n\n\nIn the visualization above, I am **comparing** and **contrasting** the **race-tracking data** of Javier Castellano and Joel Rosario in order to understand what led Joel Rosario to win and what led Javier Castellano to lose. Here, I have already plotted Joel Rosario's race-tracking data just for clarity and comparing purposes.\n\n\n\n##### Little In-Depth Insight\n\n**Typical Time Speed Curve**\n\n\n\n1 **Joel Rosario Speed Time Curve**\n\n\n\n2 **Javier Castellano Speed Time Curve**\n\n\n\n- If we compare the **above visualizations** with the **typical speed time curve**, we can see both **Joel Rosario** and **Javier Castellano** were running **head to head** at given time intervals. **However,** Javier Castellano had the **lead at 10 -15 sec intervals** **(** could be seen in **2nd visualization** **)** **but** it was a **smart move** by **Joel Rosario** by **dropping little speed just before** the **braking point** and then by took a **wider turn** and accelerated **(** it could be seen with a **blue dot** in **1st visualization** at **20-25 sec interval** **)** by charging the horse and **overtakes** Javier Castellano **between 25-35 sec intervals** where he **apex late** and **blocked** the **path for Javier Castellano** **(** it could be seen in **2nd visualization** at **30-35 sec intervals** where there was a **sudden drop in speed** **)**. Then both of them forced the horse to run till the finish line.\n\nAlso. I've used **Stats** and **Physics** to assess the **Optimal Path** for each race type and its category, defined each **Jockey's Performance** metric, created a **Machine Learning Model for Drafting Strategy**, and **much more**. Moreover, I would really appreciate comments and feedback on this. Thank you.\n\n#### Visit this [Notebook](https://www.kaggle.com/code/kushtrivedi14728/advance-eda-with-ml-big-data-derby) to see my work.", "1958304": "Hi Addison, Joe, I wonder whether the rails are moved during the season or not. I read on some internet resources that sometimes the rails are moved because the ground close to them gets worse in time. ", "1895950": "Hi, I'm a newcomer to this type of competition and I'm meeting it for the first time. Can you tell me how to submit for this type of contest? \nThanks!", "1906647": "Hi , Could you please elaborate on the data columns and what they mean please? Some of us are not familiar with the industry.\nThanks & Best Regards\nSchroter Michael", "1896699": "new to Kaggle. I asked if anyone wants a teammate on the leaderboard, apparently. But I can't find the leaderboard. So, does anyone want a teammate?? I'm a professional mathematician with several years of programming. I understand most of the ML and stats concepts from many coursera courses", "2026040": "Hi i am wondering if I add something to my Notebook, it will not considered anymore?", "2024794": "How do i convert the post-time column from str to a time datatype?", "1955246": "I have a question to Kaggle Staff and Competition Hosts.\n\nIt was only after a month of competition that I decided to look at the rules. I saw a terrible thing for myself.. \"Competitions are open to residents of the United States and worldwide, except that if you are a resident of Crimea, so-called Donetsk People's Republic (DNR) or Luhansk People's Republic (LNR), Cuba, Iran, Syria, North Korea, or are subject to U.S. export controls or sanctions, you may not enter the Competition.\" I'm living in Russia, but there are no personal sanctions on me and I **don't** support all this horror that is happening now. And do I understand correctly that I have no right to participate?...", "1943600": "Hi Addision,\n\nBy filtering the data *( ../input/big-data-derby-2022/nyra_2019_complete.csv)* , It could be found that some missing **trakus_index** for a particular horse in a race (For an example, race date - 2019-04-20 / race - 7/ program number - 6 / trakus index between 7 to 216). That kind of missing range might be an issue for the accuracy of the velocity of the horse because the distance between adjacent trakus_index points might not be the actual distance that the horse traveled. \n\n\nI am hoping for your advice on this issue, please.\n\nBest Regards\n\nSameera Sandaruwan", "1942632": "How confident are we really about the Trakus time intervals? I know in the data description it says \"from what we can tell\" it's collected every 0.25 seconds. But from what I'm seeing, that seems consistently short. I posted with some more detail here: https://www.kaggle.com/competitions/big-data-derby-2022/discussion/352553\n\nSome of the total recording times as derived from a 0.25s trakus interval are actually less than the winner's time. Unless I'm misunderstanding something. ", "1931505": "Please elaborate on the following: \n\n1. `odds - Odds to win the race passed as an integer. Divide by 100 to derive the odds to 1. Example - 1280 would be 12.8-1.` Odds of betting or some other odds of winning? If this is the betting odds, in your examples, it means that if this horse wins, I will get 12.8 times my money? \n\n\n2. `trakus_index - The common collection of point of the lat / long of the horse in the race passed as an integer. From what we can tell, it's collected every 0.25 seconds.` It is still unclear what does trakus_index column stand for\n", "1913681": "Not able to decipher data column names ", "1911686": "new to kaggle. I wanna ask how can I summit my result of the competition?? Because I can't find the leaderboard.", "1910550": "Hello,\n\nIs the Trakus data reliable?\n\nI don't mean to be rude, and perhaps it's entirely my error, but I've looked at one race in particular (the Whitney Stakes, run at Saratoga on Aug 3, 2019 - race 9 - viewable on Youtube), and the Trakus data doesn't seem to accurately reflect what you see actually happen in the race video.\n\nIn the Whitney, the horses break in a pretty good line, and the horse in gate 1 (#1) gets away on terms and holds the inside until the first turn.\n\nHowever, if you look at the Trakus data for the race, and concentrate say, on the first 10 trakus_index values for each horse, it appears to show horse #2 holding the inside for the first few strides, with horse 1 being 'between horses' (which definitely doesn't happen in the race).\n\nI visualized it using this code:-\n\nsns.scatterplot(data=tracking_df, x='latitude', y='longitude', hue='program_number')\n\n..where tracking_df is a subset of the tracking dataset restricted to the Whitney Stakes, and cut down further to show the first handful of trakus_index values for each horse.\n\nAs it seems we are being asked (possibly) to investigate 'drafting', and the tactical lines taken by jockeys in the race, if the path a horse takes on the track (straight out of the gate for instance) isn't being accurately reflected in the locational data, that's going to be an impossible task.\n\nI'm happy to be shown I've got something wrong here, but at the moment, I can't see what it is.\nThanks\nGerry.", "1909966": "Hey Guys, I am new to Kaggle and Analytics in general . This is probably not for my level since I am still learning all the basics of data analytics but I would like to learn more about the process. So if there is anyone who like to add me to their team I would be really grateful. For me it will be a really good learning opportunity.\nP.S: I have zero experience and I am learning (beginner level)", "1896628": "Thank you Kaggle for providing the wonderful opportunity for us to learn ", "1895354": "So the ranking will be displayed after the end of the competition?", "1906546": "", "2023554": "Thank you for sharing :) ", "1995224": "thanks a lot" } }