We are incredibly excited to announce the winners of the 2022 Big Data Derby.
\nFirst and foremost, we want to thank the 108 teams who took the time to enter a submission. We are incredibly grateful that you considered our data and our mission strong enough for you to participate.
\nThe winners are:
\n1- The team of Brendan Kumagai, Dr. Tyrel Stokes, Liam Welsh, Kimberly Kroetch, Gurashish Bagga: https://www.kaggle.com/code/bkumagai/bayesian-velocity-models-for-horse-race-simulation
\n2- Kyle King: https://www.kaggle.com/code/kingstats101/track-bias
\n3- Timothy and Philip Leung: https://www.kaggle.com/tlyleung/advanced-horse-race-tactics-using-coordinate-data
\n4- Artem Volgin and Ekaterina Melianova https://www.kaggle.com/code/artvolgin/winning-strategies-what-works-better
\nWe received so many good submissions, we want to further acknowledge the following as Honorable Mention winners. The judges were impressed with their work and we want to distribute it more broadly because it was well done and on topic.
\nAdrian Muresan: https://www.kaggle.com/code/louisbunuel/speed-vs-distance-not-aerodynamic-drag
\nAbir Eltaif :https://www.kaggle.com/code/abireltaief/from-freedom-at-the-racetrack-to-data-science
\nLeonie: https://www.kaggle.com/code/iamleonie/monitoring-racing-strategies-for-injury-prevention/
\nProf Groh: https://www.kaggle.com/profgroh/a-novel-metric-to-rank-jockeys-including-drafting
\nMauro: https://www.kaggle.com/mauromauro/horace-horse-racing-conditional-estimator
\nFinally, David MacGillvray: https://www.kaggle.com/code/ntsdav561/build-racecourse-maps-1of6
\nWe’d like to acknowledge this entry in particular for its comprehensive nature, it was very impressive.
We’d like to thank you all again, it was wonderful to see your work and we are hopeful we can leverage this knowledge for better and safer racing in our future.
", "messages": [ { "id": 2057148, "postDate": "2022-12-06T19:51:42.633Z", "content": "We are incredibly excited to announce the winners of the 2022 Big Data Derby.
\nFirst and foremost, we want to thank the 108 teams who took the time to enter a submission. We are incredibly grateful that you considered our data and our mission strong enough for you to participate.
\nThe winners are:
\n1- The team of Brendan Kumagai, Dr. Tyrel Stokes, Liam Welsh, Kimberly Kroetch, Gurashish Bagga: https://www.kaggle.com/code/bkumagai/bayesian-velocity-models-for-horse-race-simulation
\n2- Kyle King: https://www.kaggle.com/code/kingstats101/track-bias
\n3- Timothy and Philip Leung: https://www.kaggle.com/tlyleung/advanced-horse-race-tactics-using-coordinate-data
\n4- Artem Volgin and Ekaterina Melianova https://www.kaggle.com/code/artvolgin/winning-strategies-what-works-better
\nWe received so many good submissions, we want to further acknowledge the following as Honorable Mention winners. The judges were impressed with their work and we want to distribute it more broadly because it was well done and on topic.
\nAdrian Muresan: https://www.kaggle.com/code/louisbunuel/speed-vs-distance-not-aerodynamic-drag
\nAbir Eltaif :https://www.kaggle.com/code/abireltaief/from-freedom-at-the-racetrack-to-data-science
\nLeonie: https://www.kaggle.com/code/iamleonie/monitoring-racing-strategies-for-injury-prevention/
\nProf Groh: https://www.kaggle.com/profgroh/a-novel-metric-to-rank-jockeys-including-drafting
\nMauro: https://www.kaggle.com/mauromauro/horace-horse-racing-conditional-estimator
\nFinally, David MacGillvray: https://www.kaggle.com/code/ntsdav561/build-racecourse-maps-1of6
\nWe’d like to acknowledge this entry in particular for its comprehensive nature, it was very impressive.
We’d like to thank you all again, it was wonderful to see your work and we are hopeful we can leverage this knowledge for better and safer racing in our future.
", "rawMarkdown": "We are incredibly excited to announce the winners of the 2022 Big Data Derby.\n\nFirst and foremost, we want to thank the 108 teams who took the time to enter a submission. We are incredibly grateful that you considered our data and our mission strong enough for you to participate. \n\nThe winners are:\n\n1- The team of Brendan Kumagai, Dr. Tyrel Stokes, Liam Welsh, Kimberly Kroetch, Gurashish Bagga: https://www.kaggle.com/code/bkumagai/bayesian-velocity-models-for-horse-race-simulation\n\n2- Kyle King: https://www.kaggle.com/code/kingstats101/track-bias\n\n3- Timothy and Philip Leung: https://www.kaggle.com/tlyleung/advanced-horse-race-tactics-using-coordinate-data\n\n4- Artem Volgin and Ekaterina Melianova https://www.kaggle.com/code/artvolgin/winning-strategies-what-works-better\n\n\nWe received so many good submissions, we want to further acknowledge the following as Honorable Mention winners. The judges were impressed with their work and we want to distribute it more broadly because it was well done and on topic.\n\nAdrian Muresan: https://www.kaggle.com/code/louisbunuel/speed-vs-distance-not-aerodynamic-drag\n\nAbir Eltaif :https://www.kaggle.com/code/abireltaief/from-freedom-at-the-racetrack-to-data-science\n\nLeonie: https://www.kaggle.com/code/iamleonie/monitoring-racing-strategies-for-injury-prevention/\n\nProf Groh: https://www.kaggle.com/profgroh/a-novel-metric-to-rank-jockeys-including-drafting\n\nMauro: https://www.kaggle.com/mauromauro/horace-horse-racing-conditional-estimator\n\nFinally, David MacGillvray: https://www.kaggle.com/code/ntsdav561/build-racecourse-maps-1of6\nWe’d like to acknowledge this entry in particular for its comprehensive nature, it was very impressive.\n\nWe’d like to thank you all again, it was wonderful to see your work and we are hopeful we can leverage this knowledge for better and safer racing in our future.\n", "votes": 14 }, { "id": 2057276, "postDate": "2022-12-06T23:36:21.257Z", "content": "@nythanyra Thanks for the honorable mention. It's a really small thing: For position one, can I claim that I saved the participants because their notebook did not include any code (or a link to code) at the deadline date and, fortunately, I asked them to do so in a comment on their notebook, So the judges utilized this code \"in the response to my comment,\" to evaluate the entry. Or did you make your decision solely based on the written report without looking at the code?
", "rawMarkdown": "@nythanyra Thanks for the honorable mention. It's a really small thing: For position one, can I claim that I saved the participants because their notebook did not include any code (or a link to code) at the deadline date and, fortunately, I asked them to do so in a comment on their notebook, So the judges utilized this code \"in the response to my comment,\" to evaluate the entry. Or did you make your decision solely based on the written report without looking at the code?", "votes": 4 }, { "id": 2057270, "postDate": "2022-12-06T23:14:34.817Z", "content": "Congratulations to all Winners and Honorable Mentions. it was a great experience to participate and learn in Big Data Derby 2022.
\nAlso, I am a learner and I would really appreciate it if Winners and Honorable Mentions can help to guide me to enhance my results for this kind of scenario and would really appreciate everyone's opinion; I saw some notebooks that use different approaches and strategy to find winning a place or make profits on the bets and combination of both. However, I couldn't find an efficient or proven way or a statical proven way to overcome the above-mentioned scenario in any of the above-mentioned notebooks as a Data Scientist, ML/ AI Engineer, or Equestrian.
\nAlso, a humble request to @addisonhoward and Kaggle team I just want to give small feedback on the analytics competition that is being hosted or is going to happen in the future. Please have a live leaderboard so other non-winners and non-honorable mentions can see where they are standing or else simply try to come up with some other strategy for this type of competition to have fair standing for everyone (including winners and non-winners). I hope the Kaggle team will think about a live leaderboard for the analytics competition in the future. Also, would appreciate it if the Kaggle support team can solve my issue request (87579). Thank you.
", "rawMarkdown": "Congratulations to all Winners and Honorable Mentions. it was a great experience to participate and learn in Big Data Derby 2022.\n\nAlso, I am a learner and I would really appreciate it if Winners and Honorable Mentions can help to guide me to enhance my results for this kind of [**scenario**](https://www.kaggle.com/competitions/big-data-derby-2022/discussion/370658) and would really appreciate everyone's opinion; I saw some notebooks that use different approaches and strategy to find winning a place or make profits on the bets and combination of both. However, I couldn't find an efficient or proven way or a statical proven way to overcome the above-mentioned scenario in any of the above-mentioned notebooks as a Data Scientist, ML/ AI Engineer, or Equestrian.\n\nAlso, a humble request to @addisonhoward and Kaggle team I just want to give small feedback on the analytics competition that is being hosted or is going to happen in the future. Please have a live leaderboard so other non-winners and non-honorable mentions can see where they are standing or else simply try to come up with some other strategy for this type of competition to have fair standing for everyone (including winners and non-winners). I hope the Kaggle team will think about a live leaderboard for the analytics competition in the future. Also, would appreciate it if the Kaggle support team can solve my issue request (87579). Thank you.\n\n", "votes": 1, "replies": [ { "id": 2057301, "postDate": "2022-12-07T00:43:54.567Z", "content": "Hi Kush,
\nThanks for the feedback! We're currently looking into a method to include a leaderboard in analytics competitions, but as there is no ground truth for competitions like this, there's no way to keep it \"live\" with updating scores - participants have the entire competition to come up with a single submission to be judged after the deadline.
\nI'll check in with support on your ticket!
", "rawMarkdown": "Hi Kush,\n\nThanks for the feedback! We're currently looking into a method to include a leaderboard in analytics competitions, but as there is no ground truth for competitions like this, there's no way to keep it \"live\" with updating scores - participants have the entire competition to come up with a single submission to be judged after the deadline.\n\nI'll check in with support on your ticket!" }, { "id": 2057369, "postDate": "2022-12-07T03:28:55.943Z", "content": "Thank you for your response @addisonhoward I appreciate your time to read my text.
\nI completely understand it's hard to keep it live. However, I do believe the Kaggle team can make it possible as the team is been organizing analytics competitions for the past few years. So, I do believe there is a way to have somewhat live results instead of thinking it has no way to keep it \"live\" with updating scores by this what I mean is :
Few Suggestions to Keep it somewhat live:
\nPro's
\nCon's
\nparticipants have the entire competition to come up with a single submission to be judged at the end we are back to the heated conversation that the Kaggle team is facing currently or was facing before. Also, there are a few more con's but will discuss that later.However, I gotcha you for the above-mentioned con
What if the Kaggle team creates a mandatory rule to submit one sample submission with an abstract of their research and a few visualizations( it should be in their final notebook ) on the 45th day or somewhat after in the middle of the ongoing analytics competition?
\nNext, it should be the participant's choice if they want to share the R or Python executable code but they have to share the sample submission as a markdown notebook/file on the 45th day.
\nAlso, let's think about what if a participant joins late but is a good candidate and is not able to submit a sample submission. What should we do?
\nI do have an idea to make it nearly live for this type of analytics competition and look forward to having a discussion with a respected Kaggle official and still, the Kaggle team believes there is no way to keep it live. I would like to share a small opinion and it might sound a little harsh but they should stop hosting this type of competition where it creates heated conversation, IP theft, or any other such things.
\nLastly, I love Kaggle but I don't like the way Kaggle Analytics Competition is evaluated currently. Thank you and appreciate your time.
", "rawMarkdown": "Thank you for your response @addisonhoward I appreciate your time to read my text.\n\nI completely understand it's hard to keep it live. However, I do believe the Kaggle team can make it possible as the team is been organizing analytics competitions for the past few years. So, I do believe there is a way to have somewhat live results instead of thinking it has **` no way to keep it \"live\" with updating scores `** by this what I mean is :\n\n**Few Suggestions to Keep it somewhat live**:\n\n1. What if the Kaggle team creates a tentative leaderboard (tentative winners and tentative honorable mentions) **for each month**? \n\n- **Pro's**\n - No heated conversation on the analytics competitions.\n - Somewhat fair results than they used to be earlier.\n - By this at least it would be somewhat live than it was before.\n - By having a tentative leaderboard; competitors can know that their work is being recognized and much more pro's like this.\n\n- **Con's**\n - Though the competitors are being recognized tentatively. What if they did not win at the time of the final decision? this is the same situation that @addisonhoward mentioned **` participants have the entire competition to come up with a single submission to be judged `** at the end we are back to the heated conversation that the Kaggle team is facing currently or was facing before. Also, there are a few more con's but will discuss that later.\n\n- However, I gotcha you for the above-mentioned con\n\nWhat if the Kaggle team creates a mandatory rule to submit one sample submission with an abstract of their research and a few visualizations( it should be in their final notebook ) on the 45th day or somewhat after in the middle of the ongoing analytics competition? \n\nNext, it should be the participant's choice if they want to share the R or Python executable code but they have to share the sample submission as a markdown notebook/file on the 45th day. \n\nAlso, let's think about what if a participant joins late but is a good candidate and is not able to submit a sample submission. What should we do? \n\n- Always use a simple brain **(** the way I had used it in my [**notebook**](https://www.kaggle.com/code/kushtrivedi14728/key-to-win-the-bet-on-horses-with-math-model) and I respect other competitors work however I believe there is no way to find a path or winning place according to this [**scenario**](https://www.kaggle.com/competitions/big-data-derby-2022/discussion/370658) **)** we can provide one last opportunity for everyone to submit it on the 60th day or remove the 45th day and have the 60th day as a sample submission day. This is how the Kaggle team can evaluate the score if they want it to be somewhat live.\n\nI do have an idea to make it nearly live for this type of analytics competition and look forward to having a discussion with a respected Kaggle official and still, the Kaggle team believes there is no way to keep it live. I would like to share a small opinion and it might sound a little harsh but they should stop hosting this type of competition where it creates heated conversation, IP theft, or any other such things.\n\nLastly, I love Kaggle but I don't like the way Kaggle Analytics Competition is evaluated currently. Thank you and appreciate your time.", "votes": 3 }, { "id": 2057806, "postDate": "2022-12-07T11:07:19.260Z", "content": "I agree with Kush. It would be nice to have at least some results for each team. No feedback, no evaluation, no comments. This causes depression and frustration. I had a great experience and I thank the creators of the competition, but it feels like my work was not even watched. 🙁
", "rawMarkdown": "I agree with Kush. It would be nice to have at least some results for each team. No feedback, no evaluation, no comments. This causes depression and frustration. I had a great experience and I thank the creators of the competition, but it feels like my work was not even watched. 🙁", "votes": 4 }, { "id": 2057900, "postDate": "2022-12-07T12:41:40.793Z", "content": "@addisonhoward and Kaggle broadly: As a kaggler who participated in this contest (nearly from the beginning till the results were announced, including the \"famous\" deadline day), even though I got an honorable mention(…I expected better (% to the 5 evaluation criteria…)), I wholeheartedly agree with Kush Trivedi's statement @kushtrivedi14728 \"Lastly, I love Kaggle but I don't like the way Kaggle Analytics Competition is evaluated currently\". Without going into great detail, it seems that Kaggle does not take its handling of these analytical competitions very seriously (I'm ready to go into further detail if you request it…). Last but not least, for the sake of scientific integrity, please kindly revise some winners' works (I will try to remain impartial and not pick on anyone), but I have read almost all of the submissions since the deadline and I ran all the code…(anyway, we are all data scientists here, and I am a mathematician, passionate about math since childhood (I master Bayes' law like many I think, and I have already used it in many very advanced applications and international olympiads…). Then, with regard to one winner's work (a really non-in-depth work on the science side: many other submissions were much better and met the 05 evaluation criteria (scoring), cited on the evaluation page…), I don't think we award a prize solely for using a famous TV Analyst and Handicapper's tweets!!! This is for the scientific integrity!
Hi Kush. I quite enjoyed your contributions to this competition throughout the duration. As to the 45-day suggestion and the live leaderboard, can you share the additional cons you mentioned? It's an interesting concept.
\nA single data point: in the last NFL competition, my colleague and I had quite an out-of-the-box report that caught the eye of the judges. I believe showing the submission earlier would have invited many skilled Kagglers to improve on it and perhaps knock us out of the running.
\nThis time there was no issue as we got crushed ☹️. Painful, but OK, since we reviewed our submission in detail and talked about all the things we got wrong. Next time we hope to do better!
\nPerhaps we can share thoughts offline about the evaluation process. I respect the judges' decisions, and also understand your concerns about transparency.
", "rawMarkdown": "Hi Kush. I quite enjoyed your contributions to this competition throughout the duration. As to the 45-day suggestion and the live leaderboard, can you share the additional cons you mentioned? It's an interesting concept.\n\nA single data point: in the last NFL competition, my colleague and I had quite an out-of-the-box report that caught the eye of the judges. I believe showing the submission earlier would have invited many skilled Kagglers to improve on it and perhaps knock us out of the running.\n\nThis time there was no issue as we got crushed ☹️. Painful, but OK, since we reviewed our submission in detail and talked about all the things we got wrong. Next time we hope to do better!\n\nPerhaps we can share thoughts offline about the evaluation process. I respect the judges' decisions, and also understand your concerns about transparency. " }, { "id": 2058617, "postDate": "2022-12-08T04:37:25.943Z", "content": "Look (Abeer, I'm mainly talking with you here, and sharing with the broader community) … these evaluations are subjective, which means that a judge's score is a function of his or her perception, values and logic. That's it. And ultimately, the hosts will move forward with the submissions that best support their agenda. Is it fair? Depends on your definition. As my grandfather used to say, \"Sometimes you're the bug, sometimes you're the windshield. Try not to be the bug.\"
\nMy teammate and I are looking at all the things we could have done better instead of whining about the outcome. Many thanks to the hosts and Kaggle - it was fun.
", "rawMarkdown": "Look (Abeer, I'm mainly talking with you here, and sharing with the broader community) ... these evaluations are subjective, which means that a judge's score is a function of his or her perception, values and logic. That's it. And ultimately, the hosts will move forward with the submissions that best support their agenda. Is it fair? Depends on your definition. As my grandfather used to say, \"Sometimes you're the bug, sometimes you're the windshield. Try not to be the bug.\" \n\nMy teammate and I are looking at all the things we could have done better instead of whining about the outcome. Many thanks to the hosts and Kaggle - it was fun." }, { "id": 2058716, "postDate": "2022-12-08T06:30:18.720Z", "content": "@jpmiller Anyway, I won't engage in this debate (because I follow the community guidelines of kaggle). Unfortunately, your intervention here serves only to prevent people from speaking up and does nothing to advance the cause of greater openness, justice, and clarity (and Firmness to deal with this kind of competitions…). Please review the other comments…: \"whining about the outcome (???!!!)….\". I'll leave you with a French quote (and good luck): \" Ne vous laissez pas abuser. Souvenez-vous de vous méfier. Et même de l'évidence : elle passe son temps à changer. Ne mettez trop haut ni les gens ni les choses. Ne les mettez pas trop bas. Non, ne les mettez pas trop bas. Montez. Renoncez à la haine : elle fait plus de mal à ceux qui l'éprouvent qu'à ceux qui en sont l'objet. Ne cherchez pas à être sage à tout prix. La folie aussi est une sagesse. Et la sagesse, une folie. Fuyez les préceptes et les donneurs de leçons: Jetez ce livre…\" Jean d'Ormesson.
", "rawMarkdown": "@jpmiller Anyway, I won't engage in this debate (because I follow the community guidelines of kaggle). Unfortunately, your intervention here serves only to prevent people from speaking up and does nothing to advance the cause of greater openness, justice, and clarity (and Firmness to deal with this kind of competitions...). Please review the other comments...: \"whining about the outcome (???!!!)....\". I'll leave you with a French quote (and good luck): \" Ne vous laissez pas abuser. Souvenez-vous de vous méfier. Et même de l'évidence : elle passe son temps à changer. Ne mettez trop haut ni les gens ni les choses. Ne les mettez pas trop bas. Non, ne les mettez pas trop bas. **Montez**. Renoncez à la haine : elle fait plus de mal à ceux qui l'éprouvent qu'à ceux qui en sont l'objet. Ne cherchez pas à être sage à tout prix. La folie aussi est une sagesse. Et la sagesse, une folie. **Fuyez les préceptes et les donneurs de leçons: Jetez ce livre**...\" Jean d'Ormesson. \n", "votes": 3 }, { "id": 2059803, "postDate": "2022-12-09T08:37:54.733Z", "rawMarkdown": "", "isDeleted": true }, { "id": 2061661, "postDate": "2022-12-11T11:22:32.193Z", "content": "Hi @addisonhoward . I read the comments about the analytics evaluation and I believe there is a relatively easy thing to implement. Trouble is, we do not know the background of the people who will evaluate the entries. Are they marketing? Are they sales? Are they technical? Any presentation has to be geared towards the audience and there is no way for us to do this adaptation. Ideally, you may ask the competition initiator to give beforehand the names of the evaluators together with their credentials and jobs. And to give the evaluators full freedom to decide on the winners as a team of evaluators with no control from the customer/initiator, a bit as it happens with the other competitions. With such a procedure in place, a kaggler who considers an analytics competition, may say \"well, the evaluators are so far away from my scientific/educational background, there is no way I can communicate to them clearly, I'll skip this one\". Or alternatively, \"The evaluators are exactly the type of people I work with on a daily basis, I'll jump into this one\". If you cannot give us the names, at least the composition of the evaluators team with minimal info: type of education, a few previous jobs, current job. I think such a minimal and easy to implement change can go a long way into decreasing distrust.
", "rawMarkdown": "Hi @addisonhoward . I read the comments about the analytics evaluation and I believe there is a relatively easy thing to implement. Trouble is, we do not know the background of the people who will evaluate the entries. Are they marketing? Are they sales? Are they technical? Any presentation has to be geared towards the audience and there is no way for us to do this adaptation. Ideally, you may ask the competition initiator to give beforehand the names of the evaluators together with their credentials and jobs. And to give the evaluators full freedom to decide on the winners as a team of evaluators with no control from the customer/initiator, a bit as it happens with the other competitions. With such a procedure in place, a kaggler who considers an analytics competition, may say \"well, the evaluators are so far away from my scientific/educational background, there is no way I can communicate to them clearly, I'll skip this one\". Or alternatively, \"The evaluators are exactly the type of people I work with on a daily basis, I'll jump into this one\". If you cannot give us the names, at least the composition of the evaluators team with minimal info: type of education, a few previous jobs, current job. I think such a minimal and easy to implement change can go a long way into decreasing distrust. ", "votes": 2 }, { "id": 2061703, "postDate": "2022-12-11T12:20:47.493Z", "content": "What annoys me is not the lack of a leaderboard or information about the people who checked the work, but the absence of any comment on the work at all. Even if they wrote me one sentence - \"your work is bullshit\", it would already be better than silence and abstract praise of everyone for their works. It would not be difficult to write 1 comment for the remaining 98 teams. My teacher at the university asked me about the objectivity of the evaluation a month ago, to which I replied that for me the main thing was to get feedback on the work done… yeeeah.. I got it - silence. In general, I will never waste time on such competitions again.
", "rawMarkdown": "What annoys me is not the lack of a leaderboard or information about the people who checked the work, but the absence of any comment on the work at all. Even if they wrote me one sentence - \"your work is bullshit\", it would already be better than silence and abstract praise of everyone for their works. It would not be difficult to write 1 comment for the remaining 98 teams. My teacher at the university asked me about the objectivity of the evaluation a month ago, to which I replied that for me the main thing was to get feedback on the work done... yeeeah.. I got it - silence. In general, I will never waste time on such competitions again.", "votes": 2, "replies": [ { "id": 2103713, "postDate": "2023-01-17T10:31:12.110Z", "content": "I fully agree with your comment.
\nI didn't expect to win any recognition but at least an email saying: \"We are reviewing your work\" would have been nice.
\nI have the feeling that they didn't even see my notebook, if I don't log in to Kaggle today to see the results or find out about the winners, the awards and this discussion. And that leaves a bitter taste in having participated and wasting so much time to not receive a single email or comment.
Well, there is still hope to get feedback
", "rawMarkdown": "Well, there is still hope to get [feedback](https://www.kaggle.com/competitions/big-data-derby-2022/discussion/372073)" }, { "id": 2107544, "postDate": "2023-01-19T21:07:27.323Z", "content": "@dmitryuarov did you hear back from them? I asked them for their feedback. They sent an email saying that they will provide in 7 to 10 days, but it has been a couple of weeks now and I still I haven't received any feedback.
", "rawMarkdown": "@dmitryuarov did you hear back from them? I asked them for their feedback. They sent an email saying that they will provide in 7 to 10 days, but it has been a couple of weeks now and I still I haven't received any feedback." }, { "id": 2108351, "postDate": "2023-01-20T13:11:03.943Z", "content": "@naghmehsh I wrote to them on January 9 and they didn't answer me at all )))0)0
", "rawMarkdown": "@naghmehsh I wrote to them on January 9 and they didn't answer me at all )))0)0", "votes": 1 }, { "id": 2108850, "postDate": "2023-01-20T21:49:47.670Z", "content": "Hey there - I've escalated this with the host and they've assured me that they will reply soon and aren't just ignoring requests :) It just may take a little longer than initially indicated.
", "rawMarkdown": "Hey there - I've escalated this with the host and they've assured me that they will reply soon and aren't just ignoring requests :) It just may take a little longer than initially indicated.", "votes": 1 }, { "id": 2146534, "postDate": "2023-02-16T00:45:46.033Z", "content": "@addisonhoward may I ask why competition’s result hasn’t been shown in my profile?
", "rawMarkdown": "@addisonhoward may I ask why competition’s result hasn’t been shown in my profile? " }, { "id": 2146920, "postDate": "2023-02-16T08:45:06.713Z", "content": "Kaggle does not show analytics competition results on profile. Look no further than this very post why it can be challenging to implement it. But it is not impossible and they should seriously consider it.
", "rawMarkdown": "Kaggle does not show analytics competition results on profile. Look no further than this very post why it can be challenging to implement it. But it is not impossible and they should seriously consider it." } ] }, { "id": 2063028, "postDate": "2022-12-12T15:26:47.040Z", "rawMarkdown": "", "isDeleted": true }, { "id": 2063210, "postDate": "2022-12-12T18:07:19.110Z", "rawMarkdown": "", "isDeleted": true }, { "id": 2063882, "postDate": "2022-12-13T11:55:17.440Z", "content": "It is useless to write anything here. Kaggle and NYRA are silent. Everyone is satisfied with everything :) But I still adhere to the position that I don't care who won, I needed feedback and I didn't get it. But by the way, the longer the creators of the competition are silent, the more they drown themselves. The position of not responding even to comments here kills)) I thought refereeing of such level exists only in the CIS countries, but it turns out the whole world is the same.
", "rawMarkdown": "It is useless to write anything here. Kaggle and NYRA are silent. Everyone is satisfied with everything :) But I still adhere to the position that I don't care who won, I needed feedback and I didn't get it. But by the way, the longer the creators of the competition are silent, the more they drown themselves. The position of not responding even to comments here kills)) I thought refereeing of such level exists only in the CIS countries, but it turns out the whole world is the same.", "votes": 2, "replies": [ { "id": 2103719, "postDate": "2023-01-17T10:33:53.037Z", "content": "I fully agree with your comment.
", "rawMarkdown": "I fully agree with your comment.\n" }, { "id": 2109106, "postDate": "2023-01-21T06:13:28.183Z", "content": "I agree with you @dmitryuarov. One thing I would say in short words that can answer each and everyone's comment HOUSE ALWAYS WINS and I can prove that as all these all days I was storing the results on a daily basis and the same thing happens that happened in Cigar Mile. Moreover, I found a loophole in that and have made a profit on bets. However, I hope the Kaggle team won't host analytics competitions in the future until they make a nearly live scoring evaluation as their silence to the discussion says a lot.
@dmitryuarov I recently read comments on a different Kaggle analytics competition where participants expressed complaints in nearly identical ways (but over there, really, they say things very directly, sometimes shockingly…) I reasoned, \"That can't be true\" in my head: There is certainly a misunderstanding. But as time goes on, particularly now that I am a participant in an analytics competition and a part of this experience, there are things like that and others (…) that we discover without wanting to discover, which is extremely frustrating for me. Really, I would have preferred not to find out about them…
", "rawMarkdown": "@dmitryuarov I recently read comments on a different Kaggle analytics competition where participants expressed complaints in nearly identical ways (but over there, really, they say things very directly, sometimes shockingly...) I reasoned, \"That can't be true\" in my head: There is certainly a misunderstanding. But as time goes on, particularly now that I am a participant in an analytics competition and a part of this experience, there are things like that and others (...) that we discover without wanting to discover, which is extremely frustrating for me. Really, I would have preferred not to find out about them...", "votes": 2 }, { "id": 2064047, "postDate": "2022-12-13T13:43:11.253Z", "content": "I think mentioning how many points the judges gave for each evaluation criterion or the overall score of the notebook or writing a single-line comment would be great feedback for the participants of the competition.
", "rawMarkdown": "I think mentioning how many points the judges gave for each evaluation criterion or the overall score of the notebook or writing a single-line comment would be great feedback for the participants of the competition.", "votes": 1 }, { "id": 2064055, "postDate": "2022-12-13T13:48:27.417Z", "content": "Of course, Prathamesh, but apparently there are no and there were no such evaluation points and everything was evaluated according to criterion in someone's heads 🙌
", "rawMarkdown": "Of course, Prathamesh, but apparently there are no and there were no such evaluation points and everything was evaluated according to criterion in someone's heads 🙌", "replies": [ { "id": 2103729, "postDate": "2023-01-17T10:40:16.163Z", "content": "As I told you in my previous comment:
\nI forgot about the result of this competition, I am a rookie and I did not expect to win.
\nToday I remembered this competition (since I NEVER received a miserable email about the result of this competition or as you say \"Your work is crap\") and I entered Kaggle to see how this ended.
Many of us are left with this \"bitter taste\" of feeling that everything done in this competition has been worthless, not even to receive an email saying \"Hey, you didn't win.\"
", "rawMarkdown": "As I told you in my previous comment:\n\nI forgot about the result of this competition, I am a rookie and I did not expect to win.\nToday I remembered this competition (since I NEVER received a miserable email about the result of this competition or as you say \"Your work is crap\") and I entered Kaggle to see how this ended.\n\nMany of us are left with this \"bitter taste\" of feeling that everything done in this competition has been worthless, not even to receive an email saying \"Hey, you didn't win.\"" } ] } ] }, { "id": 2057287, "postDate": "2022-12-06T23:54:47.447Z", "content": "Congrats to all winners! 🎉
\nI participated in the analysis competition for the first time, and I think it was very difficult.
\nI'm looking at the results of the winners now and I think they're very insightful and creative. I think it was an opportunity to learn a lot. Thank you.
Along with the comments below, I also have some questions.
\nWhy are most of the winners Novice?
\nWhy aren't some analyses including code?
\nWhere are the evaluation results of all submission results?
\nI didn't know at first, but the analysis competition seems to be going the opposite way to kaggle's policy. (no results, closed, opaque…) I don't want to be a conspiracy theorist, but I think there's only two reasons why they can't share the grading results.
\nEither there is no result or the score is not fair.
If you don't know what the problem is, imagine the following.
\nAnswering the 1st part of your question @jaewook704.
\nIn my observations of the platform there are so many experts (not according to kaggle's definition) and accomplished data scientists [sometimes] with domain expertise who come once in a while and enter kaggle competitions. These are the real deal, if they saw value in kaggle tiers and rankings they could easily have that. So they aren't really novices.
\nFor example, the winners listed at 4th place in this competition, this is their 4th analytics competition win as far as I know. If competition medals were awarded for analytics competitions they would have been competition masters. Analytical competition winners and participants lack visibility in the platform. So Kush's @kushtrivedi14728 suggestion makes a lot of sense.
", "rawMarkdown": "Answering the 1st part of your question @jaewook704.\n\nIn my observations of the platform there are so many experts (not according to kaggle's definition) and accomplished data scientists [sometimes] with domain expertise who come once in a while and enter kaggle competitions. These are the real deal, if they saw *value* in kaggle tiers and rankings they could easily have that. So they aren't really *novices*. \n\nFor example, the winners listed at 4th place in this competition, this is their 4th analytics competition win as far as I know. If competition medals were awarded for analytics competitions they would have been **competition masters**. Analytical competition winners and participants lack visibility in the platform. So Kush's @kushtrivedi14728 suggestion makes a lot of sense. \n", "votes": 1 } ] }, { "id": 2066171, "postDate": "2022-12-15T13:07:29.773Z", "content": "@desalegngeb Thank you for your reply. I really agree with you. Yes, Novice is not a problem.
\nBut I think the activity history of the 4th place team you mentioned and the rest of the team is quite different.
\nThis is what I'm questioning.
\nWere they really shy data experts? Or were they data expert who only targeted this competition?
I really do not know. It could be a bit of both. If I am hard pressed for an answer I will be inclined towards the second.
", "rawMarkdown": "I really do not know. It could be a bit of both. If I am hard pressed for an answer I will be inclined towards the second.", "votes": 1 } ] } ] }, { "id": 2057586, "postDate": "2022-12-07T08:25:15.377Z", "content": "Thank you for giving us the opportunity to work on an original dataset such as this one. I intend to write a scientific paper based on my notebook. Are you, Joe, or someone on your team interested in being a co-author, and put a bit in more detail on how the data is collected (proprietary info left out)? Or should I just go ahead by myself and mention you and NYRA, as the data originators in the acknowledgements?
", "rawMarkdown": "Thank you for giving us the opportunity to work on an original dataset such as this one. I intend to write a scientific paper based on my notebook. Are you, Joe, or someone on your team interested in being a co-author, and put a bit in more detail on how the data is collected (proprietary info left out)? Or should I just go ahead by myself and mention you and NYRA, as the data originators in the acknowledgements? " }, { "id": 2058332, "postDate": "2022-12-07T20:20:49.490Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 2057276, "author_name": "", "author_url": "", "post_date": "2022-12-06T23:36:21.257000", "content": "@nythanyra Thanks for the honorable mention. It's a really small thing: For position one, can I claim that I saved the participants because their notebook did not include any code (or a link to code) at the deadline date and, fortunately, I asked them to do so in a comment on their notebook, So the judges utilized this code \"in the response to my comment,\" to evaluate the entry. Or did you make your decision solely based on the written report without looking at the code?
", "votes": 4, "replies": [] }, { "id": 2057270, "author_name": "Kush Trivedi", "author_url": "", "post_date": "2022-12-06T23:14:34.817000", "content": "Congratulations to all Winners and Honorable Mentions. it was a great experience to participate and learn in Big Data Derby 2022.
\nAlso, I am a learner and I would really appreciate it if Winners and Honorable Mentions can help to guide me to enhance my results for this kind of scenario and would really appreciate everyone's opinion; I saw some notebooks that use different approaches and strategy to find winning a place or make profits on the bets and combination of both. However, I couldn't find an efficient or proven way or a statical proven way to overcome the above-mentioned scenario in any of the above-mentioned notebooks as a Data Scientist, ML/ AI Engineer, or Equestrian.
\nAlso, a humble request to @addisonhoward and Kaggle team I just want to give small feedback on the analytics competition that is being hosted or is going to happen in the future. Please have a live leaderboard so other non-winners and non-honorable mentions can see where they are standing or else simply try to come up with some other strategy for this type of competition to have fair standing for everyone (including winners and non-winners). I hope the Kaggle team will think about a live leaderboard for the analytics competition in the future. Also, would appreciate it if the Kaggle support team can solve my issue request (87579). Thank you.
", "votes": 1, "replies": [ { "id": 2057301, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-12-07T00:43:54.567000", "content": "Hi Kush,
\nThanks for the feedback! We're currently looking into a method to include a leaderboard in analytics competitions, but as there is no ground truth for competitions like this, there's no way to keep it \"live\" with updating scores - participants have the entire competition to come up with a single submission to be judged after the deadline.
\nI'll check in with support on your ticket!
", "votes": 0, "replies": [] }, { "id": 2057369, "author_name": "Kush Trivedi", "author_url": "", "post_date": "2022-12-07T03:28:55.943000", "content": "Thank you for your response @addisonhoward I appreciate your time to read my text.
\nI completely understand it's hard to keep it live. However, I do believe the Kaggle team can make it possible as the team is been organizing analytics competitions for the past few years. So, I do believe there is a way to have somewhat live results instead of thinking it has no way to keep it \"live\" with updating scores by this what I mean is :
Few Suggestions to Keep it somewhat live:
\nPro's
\nCon's
\nparticipants have the entire competition to come up with a single submission to be judged at the end we are back to the heated conversation that the Kaggle team is facing currently or was facing before. Also, there are a few more con's but will discuss that later.However, I gotcha you for the above-mentioned con
What if the Kaggle team creates a mandatory rule to submit one sample submission with an abstract of their research and a few visualizations( it should be in their final notebook ) on the 45th day or somewhat after in the middle of the ongoing analytics competition?
\nNext, it should be the participant's choice if they want to share the R or Python executable code but they have to share the sample submission as a markdown notebook/file on the 45th day.
\nAlso, let's think about what if a participant joins late but is a good candidate and is not able to submit a sample submission. What should we do?
\nI do have an idea to make it nearly live for this type of analytics competition and look forward to having a discussion with a respected Kaggle official and still, the Kaggle team believes there is no way to keep it live. I would like to share a small opinion and it might sound a little harsh but they should stop hosting this type of competition where it creates heated conversation, IP theft, or any other such things.
\nLastly, I love Kaggle but I don't like the way Kaggle Analytics Competition is evaluated currently. Thank you and appreciate your time.
", "votes": 3, "replies": [] }, { "id": 2057806, "author_name": "Dmitry Uarov", "author_url": "", "post_date": "2022-12-07T11:07:19.260000", "content": "I agree with Kush. It would be nice to have at least some results for each team. No feedback, no evaluation, no comments. This causes depression and frustration. I had a great experience and I thank the creators of the competition, but it feels like my work was not even watched. 🙁
", "votes": 4, "replies": [] }, { "id": 2057900, "author_name": "", "author_url": "", "post_date": "2022-12-07T12:41:40.793000", "content": "@addisonhoward and Kaggle broadly: As a kaggler who participated in this contest (nearly from the beginning till the results were announced, including the \"famous\" deadline day), even though I got an honorable mention(…I expected better (% to the 5 evaluation criteria…)), I wholeheartedly agree with Kush Trivedi's statement @kushtrivedi14728 \"Lastly, I love Kaggle but I don't like the way Kaggle Analytics Competition is evaluated currently\". Without going into great detail, it seems that Kaggle does not take its handling of these analytical competitions very seriously (I'm ready to go into further detail if you request it…). Last but not least, for the sake of scientific integrity, please kindly revise some winners' works (I will try to remain impartial and not pick on anyone), but I have read almost all of the submissions since the deadline and I ran all the code…(anyway, we are all data scientists here, and I am a mathematician, passionate about math since childhood (I master Bayes' law like many I think, and I have already used it in many very advanced applications and international olympiads…). Then, with regard to one winner's work (a really non-in-depth work on the science side: many other submissions were much better and met the 05 evaluation criteria (scoring), cited on the evaluation page…), I don't think we award a prize solely for using a famous TV Analyst and Handicapper's tweets!!! This is for the scientific integrity!
Hi Kush. I quite enjoyed your contributions to this competition throughout the duration. As to the 45-day suggestion and the live leaderboard, can you share the additional cons you mentioned? It's an interesting concept.
\nA single data point: in the last NFL competition, my colleague and I had quite an out-of-the-box report that caught the eye of the judges. I believe showing the submission earlier would have invited many skilled Kagglers to improve on it and perhaps knock us out of the running.
\nThis time there was no issue as we got crushed ☹️. Painful, but OK, since we reviewed our submission in detail and talked about all the things we got wrong. Next time we hope to do better!
\nPerhaps we can share thoughts offline about the evaluation process. I respect the judges' decisions, and also understand your concerns about transparency.
", "votes": 0, "replies": [] }, { "id": 2058617, "author_name": "JohnM", "author_url": "", "post_date": "2022-12-08T04:37:25.943000", "content": "Look (Abeer, I'm mainly talking with you here, and sharing with the broader community) … these evaluations are subjective, which means that a judge's score is a function of his or her perception, values and logic. That's it. And ultimately, the hosts will move forward with the submissions that best support their agenda. Is it fair? Depends on your definition. As my grandfather used to say, \"Sometimes you're the bug, sometimes you're the windshield. Try not to be the bug.\"
\nMy teammate and I are looking at all the things we could have done better instead of whining about the outcome. Many thanks to the hosts and Kaggle - it was fun.
", "votes": 0, "replies": [] }, { "id": 2058716, "author_name": "", "author_url": "", "post_date": "2022-12-08T06:30:18.720000", "content": "@jpmiller Anyway, I won't engage in this debate (because I follow the community guidelines of kaggle). Unfortunately, your intervention here serves only to prevent people from speaking up and does nothing to advance the cause of greater openness, justice, and clarity (and Firmness to deal with this kind of competitions…). Please review the other comments…: \"whining about the outcome (???!!!)….\". I'll leave you with a French quote (and good luck): \" Ne vous laissez pas abuser. Souvenez-vous de vous méfier. Et même de l'évidence : elle passe son temps à changer. Ne mettez trop haut ni les gens ni les choses. Ne les mettez pas trop bas. Non, ne les mettez pas trop bas. Montez. Renoncez à la haine : elle fait plus de mal à ceux qui l'éprouvent qu'à ceux qui en sont l'objet. Ne cherchez pas à être sage à tout prix. La folie aussi est une sagesse. Et la sagesse, une folie. Fuyez les préceptes et les donneurs de leçons: Jetez ce livre…\" Jean d'Ormesson.
", "votes": 3, "replies": [] }, { "id": 2059803, "author_name": "", "author_url": "", "post_date": "2022-12-09T08:37:54.733000", "content": "", "votes": 0, "replies": [] }, { "id": 2061661, "author_name": "Adrian Muresan", "author_url": "", "post_date": "2022-12-11T11:22:32.193000", "content": "Hi @addisonhoward . I read the comments about the analytics evaluation and I believe there is a relatively easy thing to implement. Trouble is, we do not know the background of the people who will evaluate the entries. Are they marketing? Are they sales? Are they technical? Any presentation has to be geared towards the audience and there is no way for us to do this adaptation. Ideally, you may ask the competition initiator to give beforehand the names of the evaluators together with their credentials and jobs. And to give the evaluators full freedom to decide on the winners as a team of evaluators with no control from the customer/initiator, a bit as it happens with the other competitions. With such a procedure in place, a kaggler who considers an analytics competition, may say \"well, the evaluators are so far away from my scientific/educational background, there is no way I can communicate to them clearly, I'll skip this one\". Or alternatively, \"The evaluators are exactly the type of people I work with on a daily basis, I'll jump into this one\". If you cannot give us the names, at least the composition of the evaluators team with minimal info: type of education, a few previous jobs, current job. I think such a minimal and easy to implement change can go a long way into decreasing distrust.
", "votes": 2, "replies": [] }, { "id": 2061703, "author_name": "Dmitry Uarov", "author_url": "", "post_date": "2022-12-11T12:20:47.493000", "content": "What annoys me is not the lack of a leaderboard or information about the people who checked the work, but the absence of any comment on the work at all. Even if they wrote me one sentence - \"your work is bullshit\", it would already be better than silence and abstract praise of everyone for their works. It would not be difficult to write 1 comment for the remaining 98 teams. My teacher at the university asked me about the objectivity of the evaluation a month ago, to which I replied that for me the main thing was to get feedback on the work done… yeeeah.. I got it - silence. In general, I will never waste time on such competitions again.
", "votes": 2, "replies": [ { "id": 2103713, "author_name": "Gabriel Beguerie", "author_url": "", "post_date": "2023-01-17T10:31:12.110000", "content": "I fully agree with your comment.
\nI didn't expect to win any recognition but at least an email saying: \"We are reviewing your work\" would have been nice.
\nI have the feeling that they didn't even see my notebook, if I don't log in to Kaggle today to see the results or find out about the winners, the awards and this discussion. And that leaves a bitter taste in having participated and wasting so much time to not receive a single email or comment.
Well, there is still hope to get feedback
", "votes": 0, "replies": [] }, { "id": 2107544, "author_name": "naghmeh Shahverdi", "author_url": "", "post_date": "2023-01-19T21:07:27.323000", "content": "@dmitryuarov did you hear back from them? I asked them for their feedback. They sent an email saying that they will provide in 7 to 10 days, but it has been a couple of weeks now and I still I haven't received any feedback.
", "votes": 0, "replies": [] }, { "id": 2108351, "author_name": "Dmitry Uarov", "author_url": "", "post_date": "2023-01-20T13:11:03.943000", "content": "@naghmehsh I wrote to them on January 9 and they didn't answer me at all )))0)0
", "votes": 1, "replies": [] }, { "id": 2108850, "author_name": "Addison Howard", "author_url": "", "post_date": "2023-01-20T21:49:47.670000", "content": "Hey there - I've escalated this with the host and they've assured me that they will reply soon and aren't just ignoring requests :) It just may take a little longer than initially indicated.
", "votes": 1, "replies": [] }, { "id": 2146534, "author_name": "naghmeh Shahverdi", "author_url": "", "post_date": "2023-02-16T00:45:46.033000", "content": "@addisonhoward may I ask why competition’s result hasn’t been shown in my profile?
", "votes": 0, "replies": [] }, { "id": 2146920, "author_name": "A Legacy Grandmaster!", "author_url": "", "post_date": "2023-02-16T08:45:06.713000", "content": "Kaggle does not show analytics competition results on profile. Look no further than this very post why it can be challenging to implement it. But it is not impossible and they should seriously consider it.
", "votes": 0, "replies": [] } ] }, { "id": 2063028, "author_name": "", "author_url": "", "post_date": "2022-12-12T15:26:47.040000", "content": "", "votes": 0, "replies": [] }, { "id": 2063210, "author_name": "", "author_url": "", "post_date": "2022-12-12T18:07:19.110000", "content": "", "votes": 0, "replies": [] }, { "id": 2063882, "author_name": "Dmitry Uarov", "author_url": "", "post_date": "2022-12-13T11:55:17.440000", "content": "It is useless to write anything here. Kaggle and NYRA are silent. Everyone is satisfied with everything :) But I still adhere to the position that I don't care who won, I needed feedback and I didn't get it. But by the way, the longer the creators of the competition are silent, the more they drown themselves. The position of not responding even to comments here kills)) I thought refereeing of such level exists only in the CIS countries, but it turns out the whole world is the same.
", "votes": 2, "replies": [ { "id": 2103719, "author_name": "Gabriel Beguerie", "author_url": "", "post_date": "2023-01-17T10:33:53.037000", "content": "I fully agree with your comment.
", "votes": 0, "replies": [] }, { "id": 2109106, "author_name": "Kush Trivedi", "author_url": "", "post_date": "2023-01-21T06:13:28.183000", "content": "I agree with you @dmitryuarov. One thing I would say in short words that can answer each and everyone's comment HOUSE ALWAYS WINS and I can prove that as all these all days I was storing the results on a daily basis and the same thing happens that happened in Cigar Mile. Moreover, I found a loophole in that and have made a profit on bets. However, I hope the Kaggle team won't host analytics competitions in the future until they make a nearly live scoring evaluation as their silence to the discussion says a lot.
@dmitryuarov I recently read comments on a different Kaggle analytics competition where participants expressed complaints in nearly identical ways (but over there, really, they say things very directly, sometimes shockingly…) I reasoned, \"That can't be true\" in my head: There is certainly a misunderstanding. But as time goes on, particularly now that I am a participant in an analytics competition and a part of this experience, there are things like that and others (…) that we discover without wanting to discover, which is extremely frustrating for me. Really, I would have preferred not to find out about them…
", "votes": 2, "replies": [] }, { "id": 2064047, "author_name": "Prathamesh Sawant", "author_url": "", "post_date": "2022-12-13T13:43:11.253000", "content": "I think mentioning how many points the judges gave for each evaluation criterion or the overall score of the notebook or writing a single-line comment would be great feedback for the participants of the competition.
", "votes": 1, "replies": [] }, { "id": 2064055, "author_name": "Dmitry Uarov", "author_url": "", "post_date": "2022-12-13T13:48:27.417000", "content": "Of course, Prathamesh, but apparently there are no and there were no such evaluation points and everything was evaluated according to criterion in someone's heads 🙌
", "votes": 0, "replies": [ { "id": 2103729, "author_name": "Gabriel Beguerie", "author_url": "", "post_date": "2023-01-17T10:40:16.163000", "content": "As I told you in my previous comment:
\nI forgot about the result of this competition, I am a rookie and I did not expect to win.
\nToday I remembered this competition (since I NEVER received a miserable email about the result of this competition or as you say \"Your work is crap\") and I entered Kaggle to see how this ended.
Many of us are left with this \"bitter taste\" of feeling that everything done in this competition has been worthless, not even to receive an email saying \"Hey, you didn't win.\"
", "votes": 0, "replies": [] } ] } ] }, { "id": 2057287, "author_name": "Jaewook Kim", "author_url": "", "post_date": "2022-12-06T23:54:47.447000", "content": "Congrats to all winners! 🎉
\nI participated in the analysis competition for the first time, and I think it was very difficult.
\nI'm looking at the results of the winners now and I think they're very insightful and creative. I think it was an opportunity to learn a lot. Thank you.
Along with the comments below, I also have some questions.
\nWhy are most of the winners Novice?
\nWhy aren't some analyses including code?
\nWhere are the evaluation results of all submission results?
\nI didn't know at first, but the analysis competition seems to be going the opposite way to kaggle's policy. (no results, closed, opaque…) I don't want to be a conspiracy theorist, but I think there's only two reasons why they can't share the grading results.
\nEither there is no result or the score is not fair.
If you don't know what the problem is, imagine the following.
\nAnswering the 1st part of your question @jaewook704.
\nIn my observations of the platform there are so many experts (not according to kaggle's definition) and accomplished data scientists [sometimes] with domain expertise who come once in a while and enter kaggle competitions. These are the real deal, if they saw value in kaggle tiers and rankings they could easily have that. So they aren't really novices.
\nFor example, the winners listed at 4th place in this competition, this is their 4th analytics competition win as far as I know. If competition medals were awarded for analytics competitions they would have been competition masters. Analytical competition winners and participants lack visibility in the platform. So Kush's @kushtrivedi14728 suggestion makes a lot of sense.
", "votes": 1, "replies": [] } ] }, { "id": 2066171, "author_name": "Jaewook Kim", "author_url": "", "post_date": "2022-12-15T13:07:29.773000", "content": "@desalegngeb Thank you for your reply. I really agree with you. Yes, Novice is not a problem.
\nBut I think the activity history of the 4th place team you mentioned and the rest of the team is quite different.
\nThis is what I'm questioning.
\nWere they really shy data experts? Or were they data expert who only targeted this competition?
I really do not know. It could be a bit of both. If I am hard pressed for an answer I will be inclined towards the second.
", "votes": 1, "replies": [] } ] } ] }, { "id": 2057586, "author_name": "Adrian Muresan", "author_url": "", "post_date": "2022-12-07T08:25:15.377000", "content": "Thank you for giving us the opportunity to work on an original dataset such as this one. I intend to write a scientific paper based on my notebook. Are you, Joe, or someone on your team interested in being a co-author, and put a bit in more detail on how the data is collected (proprietary info left out)? Or should I just go ahead by myself and mention you and NYRA, as the data originators in the acknowledgements?
", "votes": 0, "replies": [] }, { "id": 2058332, "author_name": "", "author_url": "", "post_date": "2022-12-07T20:20:49.490000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "2057148": "We are incredibly excited to announce the winners of the 2022 Big Data Derby.\n\nFirst and foremost, we want to thank the 108 teams who took the time to enter a submission. We are incredibly grateful that you considered our data and our mission strong enough for you to participate. \n\nThe winners are:\n\n1- The team of Brendan Kumagai, Dr. Tyrel Stokes, Liam Welsh, Kimberly Kroetch, Gurashish Bagga: https://www.kaggle.com/code/bkumagai/bayesian-velocity-models-for-horse-race-simulation\n\n2- Kyle King: https://www.kaggle.com/code/kingstats101/track-bias\n\n3- Timothy and Philip Leung: https://www.kaggle.com/tlyleung/advanced-horse-race-tactics-using-coordinate-data\n\n4- Artem Volgin and Ekaterina Melianova https://www.kaggle.com/code/artvolgin/winning-strategies-what-works-better\n\n\nWe received so many good submissions, we want to further acknowledge the following as Honorable Mention winners. The judges were impressed with their work and we want to distribute it more broadly because it was well done and on topic.\n\nAdrian Muresan: https://www.kaggle.com/code/louisbunuel/speed-vs-distance-not-aerodynamic-drag\n\nAbir Eltaif :https://www.kaggle.com/code/abireltaief/from-freedom-at-the-racetrack-to-data-science\n\nLeonie: https://www.kaggle.com/code/iamleonie/monitoring-racing-strategies-for-injury-prevention/\n\nProf Groh: https://www.kaggle.com/profgroh/a-novel-metric-to-rank-jockeys-including-drafting\n\nMauro: https://www.kaggle.com/mauromauro/horace-horse-racing-conditional-estimator\n\nFinally, David MacGillvray: https://www.kaggle.com/code/ntsdav561/build-racecourse-maps-1of6\nWe’d like to acknowledge this entry in particular for its comprehensive nature, it was very impressive.\n\nWe’d like to thank you all again, it was wonderful to see your work and we are hopeful we can leverage this knowledge for better and safer racing in our future.\n", "2057276": "@nythanyra Thanks for the honorable mention. It's a really small thing: For position one, can I claim that I saved the participants because their notebook did not include any code (or a link to code) at the deadline date and, fortunately, I asked them to do so in a comment on their notebook, So the judges utilized this code \"in the response to my comment,\" to evaluate the entry. Or did you make your decision solely based on the written report without looking at the code?", "2057270": "Congratulations to all Winners and Honorable Mentions. it was a great experience to participate and learn in Big Data Derby 2022.\n\nAlso, I am a learner and I would really appreciate it if Winners and Honorable Mentions can help to guide me to enhance my results for this kind of [**scenario**](https://www.kaggle.com/competitions/big-data-derby-2022/discussion/370658) and would really appreciate everyone's opinion; I saw some notebooks that use different approaches and strategy to find winning a place or make profits on the bets and combination of both. However, I couldn't find an efficient or proven way or a statical proven way to overcome the above-mentioned scenario in any of the above-mentioned notebooks as a Data Scientist, ML/ AI Engineer, or Equestrian.\n\nAlso, a humble request to @addisonhoward and Kaggle team I just want to give small feedback on the analytics competition that is being hosted or is going to happen in the future. Please have a live leaderboard so other non-winners and non-honorable mentions can see where they are standing or else simply try to come up with some other strategy for this type of competition to have fair standing for everyone (including winners and non-winners). I hope the Kaggle team will think about a live leaderboard for the analytics competition in the future. Also, would appreciate it if the Kaggle support team can solve my issue request (87579). Thank you.\n\n", "2057287": "Congrats to all winners! 🎉\nI participated in the analysis competition for the first time, and I think it was very difficult.\nI'm looking at the results of the winners now and I think they're very insightful and creative. I think it was an opportunity to learn a lot. Thank you.", "2057586": "Thank you for giving us the opportunity to work on an original dataset such as this one. I intend to write a scientific paper based on my notebook. Are you, Joe, or someone on your team interested in being a co-author, and put a bit in more detail on how the data is collected (proprietary info left out)? Or should I just go ahead by myself and mention you and NYRA, as the data originators in the acknowledgements? ", "2058332": "" } }