Talk:Mario Kart Tour item probability distributions

Matching the actual in-game values
I can appreciate the effort of creating tables of the probabilities, but the data here doesn’t match the in-game data, even featuring fractional numbers while the probability is coded as integers. I’m sorry, but as it stands, this page is unacceptable and if it won’t be fixed using the actual in-game values I’ll be forced to wipe it as we can’t simply keep information not matching the in-game data as if it was in-game data.--Mister Wu (talk) 00:14, July 15, 2020 (EDT)
 * These are datamined values, so the values are actually in-game. However, due to gliders and something in the data mentioned as AP (the meaning of it is not clear yet), there are additional factors that influence the real probabilities. I don't know why your document contains some different values, but maybe you can just fix the ones that are different.Tomoto17 (talk) 03:22, July 15, 2020 (EDT)
 * I don’t have any doubt about your good faith, but the thing is that you were given an interpretation of the datamined values, based on the dataminers’ knowledge at the time. There are four item tables, not two, they also have way more entries than those, finally the Frenzy percentages are those based on the interpretation also seen in Reddit. If you give me some time I’ll dump the actual tables, next time though I suggest to be given directly the files from which you can rip what we need. Second-hand is very good for sprite et al., just not for this type of data due to how more complex it is structured, making it easy to unwillingly introduce mistakes due to interpretations.—Mister Wu (talk) 05:58, July 15, 2020 (EDT)
 * I updated the tables with the actual in-game data, the general rules will need to be updated to report said data as well (but this is a longer process, as the data is contained in two files per each item). If you want, since you did a nice job here with reporting the data you were given you can work on a simplified formatting for the main Mario Kart Tour page copying the two main tables (presumed probability after the first item box and Frenzy probability) and removing the additional probability columns where they are set to 0 as well as the currently unused items, we'll keep the full data here anyway.--Mister Wu (talk) 11:52, July 15, 2020 (EDT)
 * Thank you for updating the tables. Maybe it is just me, but I think that the readability of the tables would be a lot better if the P and A columns are separately displayed next to each other instead of between each other. The two frenzy tables look identical, does the second one have any added value, therefore? Tomoto17 (talk) 02:41, July 16, 2020 (EDT)
 * I would like to make a simplified table for the main page, but which of the three "following item boxes" tables do you consider to be the main one? Tomoto17 (talk) 02:50, July 16, 2020 (EDT)
 * This was the layout of the tables, in which the additional probability values are between the main probability values, feel free to explore different layouts as long as the values are these ones. Also the redundancy is within the code, so we just display it as it is here, while the main page of course avoids such redundancy, summarizing that the COM Frenzy table is identical. As far as non-Frenzy probability is concerned, I think the second, third of the first group and second group tables are identical? If so, we can use one of these, as they also feature the additional probability likely due to gliders.—Mister Wu (talk) 09:43, July 16, 2020 (EDT)