Lundorff Posted March 27, 2018 Share Posted March 27, 2018 https://i.imgur.com/P2kYoFk.jpg It just seems strange to me, that someone can go so high at the start and then flat-line, but perhaps I am just reading it wrong? Link to comment Share on other sites More sharing options...
KendraP Posted March 27, 2018 Share Posted March 27, 2018 (edited) The ranked season started before 13 MAR, so obviously there is some nonlinearity in the horizontal axis - a shift or leap in value. It appears the graph ia forced to start at zero (or shows the entire rating from the start of the season in that first horizontal grid step.) Alternately, i would assume such things would happen if a person stopped doing ranked, because then their rating would no longer change. Since there is some variation i assume the former (grid step nonlinearity). As an example of the latter is me: i stopped doing ranked after getting my mats. https://m.imgur.com/a/7OnGH I stopped doing ranked after getting my mats, ergo my rating has not changed since 13 mar. It looks like a steep rise due to the entirety of the season before 13 mar being crammed into that first horizontal grid step. Thus what you see is due to scale nonlinearity and possible non-participation. Edited March 27, 2018 by KendraP Link to comment Share on other sites More sharing options...
yellow_ Posted March 27, 2018 Share Posted March 27, 2018 (edited) If you play at about a 1600 level this is exactly what you'd expect it to look like. It doesn't show the whole season so you can't actually see how quickly or slowly they got to that rating before seemingly getting stuck there. Unless they got it in the last couple weeks, anyone with rating will show that same spike at the start of the graph regardless of how rapidly or slowly they accumulated it. Edited March 27, 2018 by yellow_ Link to comment Share on other sites More sharing options...
DavidAtkinson Posted March 27, 2018 Share Posted March 27, 2018 https://i.imgur.com/P2kYoFk.jpg It just seems strange to me, that someone can go so high at the start and then flat-line, but perhaps I am just reading it wrong? The graph shows that the player had a percentage of wins of 80%-70% maybe... Told you in game to stop obsessing and analyzing ranked. That won't help you. Enjoy it even if you lose 200 elo in 2 days. It's a game, not a gambling machine. Link to comment Share on other sites More sharing options...
Rafiknoll Posted March 27, 2018 Share Posted March 27, 2018 It's a game, not a gambling machine. It most certainly is. The currency you spend is time and patience. The currency you might gain is elo (which is effectively pointless as a measurement as long as some people gain it unfairly). I do agree stopping to gamble is the wise option here Link to comment Share on other sites More sharing options...
Lundorff Posted March 27, 2018 Author Share Posted March 27, 2018 The graph shows that the player had a percentage of wins of 80%-70% maybe... Told you in game to stop obsessing and analyzing ranked. That won't help you. Enjoy it even if you lose 200 elo in 2 days. It's a game, not a gambling machine. Oh I am done with it in any serious manner. I absolutely do not believe it can done on any dps class - especially when you lose 17 ELO and only gain 7-11 in most matches. With the random exception of someone who happens to win their first 20-30 matches, I absolutely do not believe it's possible to get any meaningful tracking on legit climbing. The queue is simply too random, and the price of failure is too steep. Link to comment Share on other sites More sharing options...
LordTurin Posted March 27, 2018 Share Posted March 27, 2018 I read this as someone finished their 10th match where that big jump is (so it shot up from 0), then only played matches on the days that had little moves. Link to comment Share on other sites More sharing options...
KendraP Posted March 27, 2018 Share Posted March 27, 2018 (edited) I read this as someone finished their 10th match where that big jump is (so it shot up from 0), then only played matches on the days that had little moves. The primary reason is the nonlinearity in the horizontal grid distribution. Season 9 started in August of last year, so August to March are compressed to fit in that entire grid step. I assume everyone starts the season at rabk zero, therefore this timeline compression is, in essence, a leap discontinuity in the horizontal time distribution. Owing to the obviously nonlinear scale, the data points that appear in this grid step are fundamentally different in appearance to those who lie in the later grid steps. You are right for the later time steps - his rating only changes when he plays, and obviously a few matches a day does not make as much an impact to the result as the cumulative total from all previous months. Edited March 27, 2018 by KendraP Link to comment Share on other sites More sharing options...
Alec_Fortescue Posted March 27, 2018 Share Posted March 27, 2018 what date did they fix backfills? Link to comment Share on other sites More sharing options...
-Bob- Posted March 27, 2018 Share Posted March 27, 2018 What does any of this have to do with Jay-Z and Beyonce? ugghh Link to comment Share on other sites More sharing options...
Hoppinswtor Posted March 29, 2018 Share Posted March 29, 2018 I just assume people don't earn their rating these days Link to comment Share on other sites More sharing options...
Recommended Posts