After QM games we often will get a little notification that says something random like "Great job! You died 3 times as Tyrande, better then the last 30 day average of 2.6." Has it ever been attempted to take all the different stats for Tyrande like deaths, healing, siege damage, etc and comparing them against other Tyrande on the same map for the last 30 days and reward ranked points based on how the average, or maybe similar ranked, player is performing with the same hero on the same map. If you are within +/- 5% of the average on all stats then you get the standard 200 points. If your within +/- 5% on all stats but over performed by 10% in the healing category you get 250 points for a win, or 150 for a loss. Similarly if you are underperforming based on the same hero on the same map you lose more for a loss and gain less for a win.
I know I have seen that looking at just the stats is not a great indicator of performance, but comparing them to the same player on the same map may yield different results. Taking Tyrande as an example again, if she decided to take a damage build then you should see her healing numbers drop by let's say 10% compared to average. However you should also see her hero damage jump 10% to counter the decreased healing and displacing the low healing to still net her the standard 200 points. If she goes damage build and still maintains the same healing numbers as an average Tyrande while performing above average hero damage numbers then she would be rewarded with 250.
IMO, the main reason ranked seems off to me is because players for the most part net the same amount of points gained or lossed as their teammates. One of the seemingly un noticed side effects from this system is from Blizzard actually misplacing heroes when they first join storm league. Yes, there are ranked correcting systems in place to help alleviate the players that have wrongly been placed in X rank when they should be in Y rank. The problem however is how long these systems take to notice the misplaced hero, and how long it takes for that hero to be rightfully placed where they should be once they are identified. This is in large part because the hero can occasionally be carried or carry a team.
Let's take a Bronze skilled player that got placed Gold for example. Naturally you should lose every game and drop back into Bronze rather quickly. However with the carry potential this game has your winrate likely still sits in the 35-40% range. The current system first has to identify your misplacement, then it has to start deducting more points to readjust you. You likely won't drop from Gold all the way to Bronze because as you're deranking you may get carried and the system thinks you have found your new accurate rank. The process has to restart one or two more times until you are finally placed in a proper rank. Alternatively, using a players average performance compared to the same hero on the same map and deducting points based on that comparative performance will allow the system to identify the misplaced hero faster, even in the case of a win, and at the same time ALREADY be taking 50 point per game steps to find the correct rank before the big 500 point game deductions take place. As a huge added benefit, heroes that go afk, will be identified each game easily by let's say 50 % lower the average stats. When identified the afk could get slapped with a leavers penalty and the remaining 9 players could be given 200 points for the wasted time. To make sure party's don't use this to boost their rank, you would also give the party members a less severe penalty of say 200.
I'd love to get some feedback on it. The system would obviously not be perfect, but it doesn't need to be. A different system would just need to perform better then the current one.
Source: Original link
© Post "Have the devs ever tried using their large data collection to help balance ranked?" for game Heroes of the Storm.
Top 10 Most Anticipated Video Games of 2020
2020 will have something to satisfy classic and modern gamers alike. To be eligible for the list, the game must be confirmed for 2020, or there should be good reason to expect its release in that year. Therefore, upcoming games with a mere announcement and no discernible release date will not be included.
Top 15 NEW Games of 2020 [FIRST HALF]
2020 has a ton to look forward to...in the video gaming world. Here are fifteen games we're looking forward to in the first half of 2020.