Overwatch Wiki & Community Forum for Overwatch & Ohnickel Youtube Fans under watch - Underwatch overwatch,wiki,forum,underwatch nickl

Explore
Jeff Kaplan submitted "Overwatch Players Guide to BlizzCon -- GET HYPE!!!!"
by
BlizzNews
Hey everyone, We just got back from setting up for BlizzCon and we couldn’t be more excited for this weekend! We wanted to give you a quick Overwatch Fan guide to getting the most out of BlizzCon! >> << Opening Ceremonies is always one of the best parts of BlizzCon. It’s our chance to welcome everyone, thank the community for being so awesome, and also show you some of the cool stuff we’ve been working on. If you’re not coming to the Anaheim Convention Center for BlizzCon, you can watch Opening Ceremonies for free right here: If you are joining us in Anaheim this weekend, we’re doing Opening Ceremonies a little different this year. Rather than just having one main stage, like we used to, Opening Ceremonies will now span the entire convention center. But what does this mean for you? Well, you don’t have to rush to get seats in the main hall! In fact, we’ll be presenting all of the cool, new Overwatch stuff in the Overwatch Arena this year. So: That’s where I’ll be—I really hope to see you guys there! >> << Speaking of the Overwatch Arena: It’s amazing this year! The stage is fantastic and you can watch the best players in the world from any angle – a full 360 degrees. Even if you’ve never been interested in watching Overwatch played professionally before, I highly recommend coming by and checking it out. It’s exhilarating. You can find all sorts of information about matches here: And of course, those of you staying at home can be there right alongside the pros all weekend. >> << This is located in Hall A, right next to the Overwatch Arena. Want to play the newest stuff? This is where you want to be! >> << This panel is going to be awesome. This is where the team will go into more detail about new stuff. There might even be some news here that doesn’t happen during Opening Ceremonies. I feel like this panel is a “must watch” if you’re an Overwatch fan. If you happen to be going to BlizzCon, this one will be on the Mythic Stage, which is on the far end of the convention center—opposite end from the Overwatch Arena. I know the team would love seeing you at the panel and really enjoy your support. >> << This is a really fun panel. We’re going to show you a bunch of behind-the-scenes “making of” moments from the inception of Overwatch to our BlizzCon 2014 announce. Ever wanted to see early gameplay footage of the first Overwatch playtest? This panel is for you! >> << Epic Stage Friday 5:15 PM PDT Epic Stage Saturday 4:30 PM PDT Legendary Stage Friday 4:15 PM PDT Legendary Stage Saturday 2:30 PM PDT Mythic Stage Saturday 5:30 PM PDT It’s going to be an awesome BlizzCon, whether you’re joining us at the show or watching from home! Time to get hype!!!!! http://us.battle.net/forums/en/overwatch/topic/20759170996#post-1 - Jeff Kaplan
Jeff Kaplan submitted "Quick Update"
by
BlizzNews
Hey all, Thanks for all of the discussions here on the game. We've been reading as much as we can and appreciate all of the passion. Sorry we haven't posted as much as we usually do lately. We've been working really hard on some pretty awesome stuff for Blizzcon and we're super excited to share it all with you next week! I know some of you have concerns about some issues in the game so I wanted to just take a quick minute and tell you what our current focus is internally. We've been discussing the state of the Mercy rework a lot and listening to all of the feedback. Overall, we feel like the new direction for Mercy is much better for both Mercy and non-Mercy players. But we do feel like she's too dominant right now. In particular, we feel like the resurrect ability is changing the tone of the game to a degree which is simply too powerful. As such, we're experimenting with different toned-down versions of the ability. After all, resurrect is no longer an ultimate and is now an on-cooldown ability so it needs to be less powerful. I know that you're going to want more details but we're nearing a point with internal iteration where we feel comfortable putting something on the PTR soon. We'd rather have you play the changes than theorycraft them. Another topic that we've been focused on a lot lately is Competitive Play. We don't have immediate changes to discuss right now but we're closely studying issues and perceived problems with the system with an eye toward improvement. Along with that, we talk a lot about matchmaking within the competitive system. We have some plans that should improve some issues that players are seeing -- especially at the high end of the skill rating spectrum. We have long and short term plans for Competitive that should hopefully improve the experience over time. We're hoping that we can get more of these changes in at a faster pace rather than waiting for major milestones. Some of the matchmaking work we want to do is strictly done on our servers which means that we don't have to wait for a big patch to go out. Again, I know you want more specific details but game development is a process and we're actively discussing and iterating on ideas and systems right now. The third big focus that's being discussed is toxicity. A lot of what we do here we cannot and will not discuss publicly. The more players understand about how these systems work, the more they will be able to game them. There have been a lot of improvements in this area and the number of actioned players continues to rise. The more we restore your faith in our systems, the more we hope you use them (talking about reporting here specifically). We're also hoping to have player feedback in the game when someone you reported was actioned by the end of the year. The email pilot program has gone well so far and we keep trying to expand that until we get the in-game system up and running. Please keep using the reporting system. I swear to you, it does work and we're really focused on improving it. But we need your help. Obviously, there is a lot more going on besides the three topics I brought up. But I think those three things are some of the issues the community as a whole is most concerned about and therefore I think those should be my team's top priorities. We have a ton of cool, fun stuff coming -- some of which will be announced at Blizzcon. And we're also really proud of some of the work we just completed to make spectating esports better. World Cup should be pretty cool and will teach us a lot for when it comes to improving spectating in the future. We really appreciate all of the passion you've shown for the game. We're lucky to have players like you. All the best, ow team http://us.battle.net/forums/en/overwatch/topic/20759395949#post-1 - Jeff Kaplan
Jeff Kaplan commented on "Do you think Overwatch team simply not have enough people?"
by
BlizzNews
When we launched Overwatch in May of 2016 our team was comprised of 70 developers. When I say developers, I am specifically talking about artists, engineers, producers, sound/music designers and game designers. We're now at just north of 100 developers. We do have a few open roles (you can check our job site for these) but we mostly feel pretty good about the size of the team in regards to the things we'd like to accomplish. One of the things we value as a team is the concept of doing more with less. Even though I wouldn't say we're necessarily a "small" team as modern development standards go, we're certainly not an overly large team either. We view ourselves as artists and craftspeople and we like to avoid that feeling of being "cogs in a large machine". We like the fact that we own Overwatch together as a unified team with a shared, collective vision. There is not one person -- or select people -- who dictates what we do. We're not peons of some visionary -- just implementing his or her vision -- and we're certainly not beholden to any corporate "overlords" telling us what we can or cannot do. We are 100+ game developers who absolutely love playing and making Overwatch. The day we announced/launched/put OW into Beta, the game belonged to a large community. And we truly do see ourselves as the custodians -- not overlords -- of the game and community. With that said, something that no one can ever take from us is that the 100 or so of us on the team created something together that we truly believe in and stand behind. We have poured our hearts and souls into a game that we believe is special. We've staked our names and our careers on it. We try -- every waking hour -- to show how much love and magic we believe Overwatch can represent. Since the day we have launched, we've only increased our efforts and dedication. Overwatch is a 24/7, 365 days a year affair for us. Overwatch doesn't stop because it's 5 o'clock on a Friday evening. Overwatch doesn't stop because it's our kids' birthday. Do we do everything right? Certainly not. The 100 members of the OW team are *far* more critical than most of you of the game on a good day (although we're a little more professional about how we express our criticism). You will never meet one of us who says, "the game is perfect". We have no shortage of ideas of how to make the game better and we are lucky enough to show up to work every day and try to make those things happen. We've built the team around the people who we consider to be the top developers in the entire industry. And because of that, we view them as amazing, creative individual contributors and not just "mindless implementers" who can be easily multiplied to increase output. While most players have a few "causes" that they focus on, we - the team - care about *literally* every experience that has anything to do with OW. We care about where a button on the screen goes, what sound it makes, what color it is. We care about what the email says to a player who has just been banned. We care about constantly trying to improve ping times worldwide and server up time. We care about what Mei says when she is sad. We care about the integrity of Top 500 play, the tournament rules for esports, how big the OW logo is on a backpack. We care about all 26 heroes equally. Each of us wants to be involved and contribute to all of these things plus 10,000 other OW-related things that come up each day. We are the 100 people who created the experience and whether it is good or bad, that's on us. We know we're only as good as our last patch. Our passion and love is for making and playing Overwatch. Things like posting on the forums -- for many of us -- is not a normal or core part of our job. From Day 1, we felt like we wanted to be a development team that communicated more openly with our player base. We tried to make as many posts on these forums as possible. Our intention -- more than anything -- was show a presence and let you know that we're listening. We're not naive enough to think that we can sufficiently address each player's concerns here. But we do want you to know that we -- the OW development team -- are here with you, listening. We try to do other things to let you know that we're part of this community as well. We make Developer Update videos so we can talk directly to you and explain what we're doing and why we're making decisions. We do not hide behind online handles or layers of Community Managers and PR Spokespeople. Developers speak to you directly, using our real names. And if you'll allow me to speak openly for a moment -- it's scary. Overall, the community is awesome to us. But there are some pretty mean people out there. All of our developers are free to post on these forums. Very few of us actually do because it's extremely intimidating and/or time consuming. It's very easy to post the wrong thing and make a "promise" to the community that no one intended to make. Once we say we're working on something, we're not allowed to "take it back". It's set in stone. Also, because we are open with you and do not hide behind an anonymous handle (like all of you have the luxury of doing), we often times get personally attacked and threatened. Most great developers I know just love being head's down making or playing games. The "public speaking/posting" part of the job is downright scary and intimidating. It often feels like there is no winning. As a result, there are a small few of us who do most of the posting here. Two weeks ago, I was offsite all week without posting access (I cannot make forum posts from my phone for security reasons). The week after that, I came into the office 1 hour later than I normally do (I was feeling extremely fatigued and rather than waking up at 5:40am like I do most days, I woke up at 6:30am). The end result was that for 2 weeks I haven't posted at my normal rate. I apologize that it's been a quiet two weeks but that doesn't mean that we -- the OW team -- haven't been working just as hard as we always do and are not dedicated to making this game great. Just because there are "a lot of threads" or "a lot of upvotes" about a certain topic, does not mean we're not paying extremely close attention to not only the community feedback -- but more importantly -- the game itself. Of course the team talks about Mercy. Of course we talk about Ana. But we talk about all of the heroes. If we post, "we're talking about Mercy" immediately there is an expectation that she is going to be radically changed in the next patch when the truth is, we might just leave her how she is for a while. We're not allowed to post that here without our bosses (and I am talking literally here) receiving emails from some of you demanding we be fired. It's not exactly what I would call a safe environment for creative people to openly express their thoughts and feelings. Anyway, we always appreciate feedback when it is expressed in an open, direct and constructive way. Player feedback is what has made this game great. We will try to communicate as much as we can. We love being a part of the community and we don't ever want to view ourselves or be viewed as being separate or removed from that. Thanks for being patient with us. http://us.battle.net/forums/en/overwatch/topic/20759346970?page=4#post-75 - Jeff Kaplan
How I think Blizzard can fix D.va PROPERLY
by
Cidd
As a long time D.va main, I have loved her through all of her stages. Yes even when her mech killed her. Due to this I want to pitch an idea I have always had for her matrix. A solution that will benefit everyone. Obviously having D.va's matrix halved was not a good solution to her problem as a tank. My example of the kind of tank D.va should be more like, is Zarya. Zarya is rewarded with charge by protecting her team and herself with bubbles. This is what makes Zarya's gameplay that of a Tank role. So why not just make D.va's matrix 2 seconds base, but last up to 4 or so seconds provided she is deleting enough Damage, ults, etc. This idea is obviously not perfect, but my intention isn't to push for specific stats on her matrix, it is just to push towards making D.va what she is straying even farther from. A TANK. Before her matrix was too simplified and it annoyed players, it just didn't feel good to be covered by it and it took little skill to delete Ults that people worked hard to generate, sync, etc. With a change like this, it makes D.va's matrix easier to use, but actually keeps it viable. A 2 second matrix regardless of what she uses it to eat, is simply too short. One more thing. People seem to want D.va to get her armor back and be some sort of bruiser tank. In no way should Blizzard EVER give her back her old armor. This does not solve the problem of her being an Ult sponge, and is an aspect of the old D.va that was removed for a reason. I don't want an Overpowered Monster D.va. I wan't my favorite character to just be effective, and fun to play both as and against.
Jeff Kaplan commented on "Report system is a failure"
by
BlizzNews
You make some really good points and we agree. The reporting and punishment system is currently one of our prime focuses. You will be seeing a developer update about this very soon (we recorded it last week -- not sure when it's going live). We have short, medium and long term plans and we haven't done a great job in communicating all of these things to you. In the short term we are re-evaluating every punishment and are in the process of converting silences over to suspensions. We're also increasing the length of suspensions. Pretty soon, we hope for silences to go away all together and only have suspensions and bans for punishments (with a few exceptions like forced BattleTag changes for naming violations). Also in the short term, we've been testing an email notification system to let you know when someone you've reported has been actioned. We cannot provide a lot of detail in these emails but they only go out if an action has taken place (meaning your report contributed to someone getting suspended). The other short term thing we just did was ban a number of accounts for boosting in Season 5. These accounts were actioned either yesterday or are in the process of being actioned. They will lose all Season 5 rewards. In the medium term, we're looking to completely ban repeated Competitive offenders meaning if you've been banned for more than X seasons, we will ban you from ever playing competitive again. We also want to increase the email notification system to other forms of reporting as well as work on an in-game notification system to let you know that your reports are doing something (this will edge more into the long term section -- the in game part). For Season 6, we're going to be way more aggressive with boosting/throwing or any sort of SR manipulation. We have some very smart people who are getting very good at detecting this behavior and we are actively building systems (and punishments) around SR abuse. Also in the medium term, we have a new series of punishments we're going to try that escalates much more quickly (so you don't have egregious cases like the one you're referencing). Basically, extreme offenders will "strike out" of the game much quicker. As part of these increasing punishments, we're looking to make it so that offenders get blocked from Competitive play much sooner -- more details on this as we get closer. In the long term, we really want to work on systems that encourage positive behavior and reward good players. It really bums us out to spend so much time punishing people for being bad sports. We like making cool, fun game systems -- that's what we do for a living. But because people seem to lack self-control or because people like to abuse anonymity and free speech we're put in a position of spending a tremendous amount of our time and resources policing the community. We will do this as it is our responsibility but we'd like to spend more time rewarding good players rather than having to focus on poor sportsmanship and unacceptable bad behavior so much. Like it or not, this is an "us, the OW community problem" and not just an "OW team problem". For better or for worse, we're in this together. We're working hard to make changes. I hope you all do too. http://us.battle.net/forums/en/overwatch/topic/20758737651#post-16 - Jeff Kaplan
Pro/Streamer/Analyst Respond to Jeff Kaplan's Game Balance Post
by
Underwatch
Jeff Kaplans post about the current metagame and radical balance changes: https://us.battle.net/forums/en/overwatch/topic/20757706588?page=2#post-24 @TrivPrivet says: "this message from jeff is exactly why i got faith in the OW team to make OW great, keep up the hard work " https://twitter.com/MisfitsTviQuE/status/883457831231516673 @Alicus says: i fully agree. i really enjoyed the level-headed take on balance. https://twitter.com/AlicusOfficial/status/883492533594095617 @Mendoksaii says: "This is why Jeff is my dad, read it. Seriously." https://twitter.com/Mendokusaii/status/883461317524205568 @CaptainPlanetOW says: "One thing he touches on that bears repeating is that the pro-level usage of heroes is essentially meaningless the further you stray from GM" https://twitter.com/CaptainPlanetOW/status/883432528056594432 @EeveeA_ says: "It's such a good read, dude. I really look forward to them testing more things on PTR, and specifically Res needs long-term changes." https://twitter.com/EeveeA_/status/883462122285518849 @ctHeurix says: "Absolutely worth the read: " https://twitter.com/ctHeurix/status/883466627483914241 @ggDOA says: "Fantastic post by Jeff Kaplan on the state of the Overwatch "meta" at the moment. " https://twitter.com/ggDoA/status/883460153663660032 What do you think? Comment below Sources from: https://www.reddit.com/r/Competitiveoverwatch/comments/6lz9m7/prostreameranalyst_responses_to_jeffs_post/
Stale Dive meta drops Overwatch PCbang player base below League.
by
Tok
Prominent streamers taimou, rainbow, and Yongbongtang are voicing concerns over the dive meta and its effect on the players base of overwatch. A common theme in the competitive world is the need for balance and variety. At the current moment, it seems that about 10 of the 24 heroes are played consistently on the ladder in the dive meta which hasn't changed in since the end of season 2. Sponsorship for apex is down and koreans are analyzing the issues. --------------------------- Still have the same requests as before. Increase hero output from blizzard to balance the hero distribution. 2 main tanks, 2 main healers, 2 off healers, stale 222 dive comp. -Blocking koreans going to bangs to mess with NA players made them have to play meta. -Blizzard should be releasing many more heroes at once and testing them more actively for feedback on the ptr. Not as focused on animations and cosmetics early on can implement them more slowly over time. - Current money model does not fix the main issues with the game. Skins do not bring content or balance to the game. Mission packs around the new heroes is a preferred model. - Overwatch is now below League of legends in pcbang player base. - Tournament viewership numbers not reaching expections and many complaints about the replay system. -Current blizzard focus on the highlight system, potg's, and lootbox sales are not matching the consumer desires in the korean scene. Complaints about the game are making sponsors pull out. https://www.reddit.com/r/Competitiveoverwatch/comments/6l74f3/yongbongtang_overwatch_usage_is_showing_signs_of/ https://www.reddit.com/r/Competitiveoverwatch/comments/6htbsw/taimou_i_havent_truly_enjoyed_playing_overwatch/ https://slingshotesports.com/2017/03/28/league-of-legends-remains-no-1-in-pc-bang-percentages-month-after-new-overwatch-season-starts/
A body of thoughts on SR derivation, info towards calcating it and nuances missing for accuracy
by
Tok
There are sort of some ideas about SR that I think are close to right but not exactly clear enough that you could derive and test models. (I don't really plan on doing this again for a Blizzard game.) I guess there's not really a good place for this topic and it's pretty popular so I thought I'd put some things here. Its hard to have a technical discussion of SR with people that don't understand math. I'm going to introduce the math enough to have the points of the discussion. The theory here is not with the actual mmr, but more on reasoning with optimal SR and the choices that Blizzard is making. ------------------------------------------- Blizzard's history of skill calculation- ------------------------------------------- During the brood war era, you could match with any player and play any player. It was kind of up to the players to find out how to balance the games on their own by trial and error. Custom servers were being made for brood war with some look at your history (general wins and losses). The best players were often trying to play the same players over and over who were better and their win-loss record didn't give an indication of skill. The guess on how to solve it came from giving you an approximate skill rating and using that as a basis for guessing how to weight how many points you get from a win or loss (c-=1 a+=9). Thus people were estimated simply just as weights(a term meaning you scaled the points won or lost based on the difference). After a while the majority of players were in the middle and people started to realize that the distributions were must like population iq with a mean, median,etc. (https://en.wikipedia.org/wiki/Sum_of_normally_distributed_random_variables) They looked at books from engineers and found a simple system which estimated a person mmr using code from Elo. Thus, a players mmr estimates were born with elo. Blizzard adopted the community work on this and used it in warcraft 3. (https://en.wikipedia.org/wiki/Elo_rating_system) There are several issues to using an elo system, such as it does not work for non-normal distrubted data, finding an accurate game weighting factor (k value), selective pairing from players being available on different regions or different times of the day, and the tendancy for top players to deflate to maintain the distribution. World of Warcraft- Burning Crusade MMR.(2007-2011) During the release of arena with the burning crusade a similar system was used, and gradually over time Blizzard realized there were flaws in the mmr calculation. They started to look at ways to balance the system and brought on Ghostcrawler. He took time to look at the PVP matches in wow and then started to define other criteria to understand how to rate players such as gear rating they started to work on including gearing factors into the mmr. (https://worldofwarcraft.com/en-us/news/7702178) Starcraft 2- During the release of starcraft a new problem appeared in terms of estimating sr that didn't exist as much in wow. The players have the same relative skill but one race or another is much more favored onto a map. Map win rates were also included into the weights and after a while they learned on how to improve this weighted rating. IF THERE IS ONE LINK YOU CLICK ON TO UDERSTAND THIS HISTORY THIS IS YOUR LINK.- (http://www.teamliquid.net/forum/starcraft-2/332391-how-to-calculate-your-mmr-from-results-of-1-match) Players realized the the weighting function therefore had to have map weights and complained. In starcraft 2, the idea then came if players want the best practice they can get they can play all the maps and have their sr suffer for losing on maps which they are not favored. The opposite scenario happens where a player looking to show they are the best player could stand out as the top player by the ladder by vetoing certain maps and having their sr climb. Blizzard decided to introduce vetoes to this end. The top players on the ladder looking to get scouted tended to have vetoes, where the pros tend to have lower sr but ignore vetoes. Thus, Blizzard recognized the importance of a veto system. Overwatch- Enter Overwatch, Blizzard had a rough idea of the need to include hero choice, relative hero rating, and map ratings in terms of weights to influence sr. The 5estion of how to include this weights and select players for the games were being tested during the beta of the game. One of the first issues that blizzard ran into was that the player base was not fitting into a normal distribution. the distrubution had a set of palyers grinding very hard, and a set which grind at a smaller rate. This meant that the normal distrubtion was biomodal. The players grinding very hard were basically only queueing with other players grinding very hard and those who weren't were separated from them. This meant that queue times were very long in overwatch. (fenn3r for example waited over 45 minutes for a game on his stream). Blizzard recognized this to be an issue. And had to figure out what to do. To some degree this decision led to Overwatch being the way it is to this day. The solution to some degree was basically to make it so that a second factor which influenced your ability to queue with higher level players would be that players that played often would increase their chance of playing with players at either end of the spectrum. This helped with times of the day where the player base is lower and leads to an optimal ladder period. This level balancing has been proven multiple ways and often pros are paired with very high level single trick underperforming players playing during certain times of the day. SR for players that play often is inflated to match the 2 distributions to remove the bimodal behaviour and force it to a normal distrubution. Additionally, they chose not to have map vetoes, this means that players which tend to train for a specific thing weill not have the optimal sr, and that the most abusive and versatile players on the ladder will have the highest ratings. They gave no real reasoning behind having ladder be the way that it is choosing that they want to have representation on ladder maps which players tend to have closer matches rather than representation on maps which players tend to win. This lead to the current cherry picking system of the maps and the players cherry pciking heroes which gives anomalous data not fit for the SR system. With this history in mind, I'm going to cover some of the mathematics and theory to the point where hopefully a person who reads a Blizzard post could in theory predict the factors which lead to a rating and create models to test their theories. ------------------------------------------- Math Primer, estimation theory, and random walk- ------------------------------------------- I guess its worth defining some important concepts to someone with calculus as their background. In general you can write math as a group of numbers collected at time intervals (discrete) or you can view it as never having a hole in the information and predicted. (continous) Both are important, in estimation (using data to test a hypothesis) since a continuous value is used over a period of games to generate the data. Lets assume a random walk model leading two different final distribution functions, a normally disributed function like a gaussian function, and a confined equpartition model with time as the variable rather the the displacement as the variable. Random walk leads to the logistic curve as the expected score of player is is the inverse of increase of a raised exponenent player ratting difference, normalized by the total distribution size E_a=1(1+10^(R_b-R_a)/n) This leads to a bayesian probability where E_a=Q_a/(Q_a +Q_b). q_a=10^R_a/n. (https://en.wikipedia.org/wiki/Bayesian_inference) Bayes theorem implies that when the probability of an event is low, the likelihood of a false positive is increased. An example in medical terms is that when you have a medical test, when the prevalence of the condition is low, the likelihood for having a false positive significantly increases which means that low prevalence conditions require more testing to ascertain accurateness and a tendency for inaccuracy occurs. The implementation of the ELO system required a certain aspect which was that equal representation was apparent and that factors. The ELO system was then heavily modified to add weights to generallize the normal distribution by summing normally distributed random varibles. (https://en.wikipedia.org/wiki/Sum_of_normally_distributed_random_variables) By using a characteristic function the independant varibles are summed in the exponents to generate the f-function different between the skill rating of 2 players. https://en.wikipedia.org/wiki/Characteristic_function_(probability_theory) This acted like a convolution between multimodal kernals which is the method of how weights algorithmically over discrete data result in a normal distribution(F_a). F_a(Q_a-Q_b)=E[e^it(Q_a-Q_b)] A priori estimation. Giving a player an inital placement can be used towards the process of a priori estimation. You use bayesian estimation and divide a smaller estimate over a known distribution. While match differences occur, the initial placement it appears gives an estimated player weight across all roles. It uses a previous modifier g_a. F_a=conv(F_a,G_a)/Int(F_p, G_p) where p is the population estimate. https://en.wikipedia.org/wiki/Maximum_a_posteriori_estimation The descriptors for using this have been enhanced significantly in public forums using machine learning algorithms which selectively isolated the key data points. Blizzard's algorithm does not use this approach but sticks to the bayesian framework of weights using sums of normally distributed random variables. http://www.cs.ru.nl/bachelorscripties/2016/Wietse_Kuipers___4317904___Improving_matchmaking_with_game_data.pdf Performance of Machine Learning Algorithms in Predicting Game Outcome from Drafts in Dota 2 http://ieeexplore.ieee.org/abstract/document/7861165/ ------------------------------------- Its likely that blizzard has in fact looked at these other methods and is beginning to test them to some degree but historically and based off empiracal evidence given, its likely that they are not used currently. Post hoc thoughts about progression since the beta until now. The addition of using a map a-priori estimate and defining the weights have been added to overwatch since release. The population data is what is missing to calculate the weights for a map estimate in the bayesian framework. The community attempts to understand population data stem from multiple facets of information given by blizzard. https://us.battle.net/forums/en/overwatch/topic/20745504371#post-3 A post by blizzard says that factors include. Individual performance on heroes, opponent sr, map win rates, game penalties, prior estimates from quick play or other services. At some point it seems that staying on a single hero and forcing other players to switch is the best case to maintain sr even more than winning. Losing and having a worse player rating means that you drop much faster in the bayesian approach used. The bayesian approach forces as much as it can for win rates to be 50% and the level balancing which occurs all appear to give idea as to the penalties that are experienced from external factors. The tendancy for cherry picking in both players and the maps means that the most skilled players will not be the highest on the ladder ever. This has always been the case for blizzard games though and that sr is not designed around having the best players on the top of the ladder but rather giving certain styles and edge in getting the highest level games.
Overwatch Campaign Game
by
SirNightmare
On the Blizzard careers page, a job has opened up for a Lead Software Engineer, Engine, saying: Blizzard Entertainment is seeking an experienced engine engineer to work on a robust first-person engine for an unannounced project. The ideal applicant has demonstrated ability to lead a team of engineers to success, has a vision for how a scalable, performant and technically excellent engine should be built, and has a passion for adopting state of the art rendering technology to implement visually stunning games. Another job opened up for a Lead Software Engineer, Tools, saying: Blizzard Entertainment is seeking an experienced tools engineer to work on an unannounced project that utilizes a robust first-person engine. The ideal applicant has demonstrated ability to lead a team of engineers to success, has a vision for how tools should be built, and has an unwavering passion for developing content creation tools. A third opened u for a Senior Technical Artist, Animation, saying: lizzard Entertainment is seeking an exceptionally skilled technical animator to join us in developing a character and animation pipeline. Your charge will be to create rigs and tools for animating a wide range of characters and creatures. You will help guide modelers and animators in the creation of this ensemble, supporting the crisp gameplay and stunning visuals that are hallmarks of Blizzard’s work. Our ideal candidate has extensive experience rigging on one or more successful AAA console or PC games. We are looking for someone with extensive experience with 3D animation tools, great collaboration skills, and a passion for Blizzard games. You’ll be part of a team just as excited about making great games as you are, and together we’ll work to create some of the most epic entertainment experiences ever. These are all jobs related to an unnanounced, robust FPS, with many characters and creatures. This could definitely point toward a second Overwatch, featuring a full single player campaign. It could be a project for Overwatch, but it says it is for an unannounced project, not Overwatch; and it has been claimed that all DLC's will come with Overwatch. Therefore, it is most likely going to be a second Overwatch, based on a campaign. It has also been previously stated that it is hard to make PvE in Overwatch's engine, which would explain why this project needs Software Engineers. This makes a lot of sense. Sadly, however, this will probably take years to create, if these jobs are jest opening up. https://careers.blizzard.com/en-us/openings
I think Blizzard should change the definition of SR to be OVERALL skill at Overwatch.
by
Tok
The idea of having this is to put sr as a ranking of overall skill at Overwatch. One could say that a mercy player is a top mercy player. This is true, but they may or may not actually be good at Overwatch OVERALL compared to another mercy player who plays her at a similar level, but who can also play many other heroes at a higher level. I propose that Blizzard changes the definition of SR to be OVERALL skill rating. So here's an idea how implement this. What I'm proposing is a 10% role based rule. This means that your competitive ladder games has to have 10% on each role. The roles would be the same is in the current competitive scene. This again requires that the heroes be labeled as they are in the tournament scene and not in Blizzards current method. Each game you play where you do not have played 10% of all roles, you are penalized with a loss even if its a win. This does allow that you spend up to 50% of your time focusing on your role of specialization while still having sr represent OVERALL skill. -------------------------------------- 6 roles- Main tank-Rein, Winston Off tank- Orisa, Zarya, Roadhog, Dva, Mei Main DPS- Soldier, Pharah, Tracer, Widow, Mccree Main healer- Mercy, Ana Off healer- Lucio, zenyatta Flex- Genji, Hanzo, Junkrat, Bastion, Reaper, Torbjorn, Symmetra, Sombra. -------------------------------------------- The time spent in a 6 stack does not affect your total time in the ladder. Thus if you played only one hero in a 6 stack you'd never have a penalty. Thus, teams looking to practice specific heroes, would have more incentive to ladder as a team and have a separate team sr and solo sr.
An unoffcial/secret leak or just a mere coincidence?
by
iNormal
I, myself spent more than 1000 hours on Team Fortress 2 before I switched to Overwatch. Now I usually play Overwatch more than TF2 but sometimes I go back and visit TF2 for a few hours. Yesterday, I was just wandering around Youtube to watch some Overwatch animations made by some TF2 Youtubers and I subcribed for a channel called Fanic. A Youtuber who make SFM ( Source Filmaker) videos about TF2 for the Saxxy Award, an Oscar Award alike but created by Valve that range varies from TF2 to Half-Life 2, Left 4 Dead series, Portal 2 and DOTA 2, etc... but for Valve's games only and the majority of the SFM videos are for Team Fortress 2 entry. The aim of the Saxxy Award is attract each Valve's game communities to comepete to each other with animations create by each individuals; bringing entertainment to the each own community perspectively and any animation won will receive a Saxxy Statue. In some cases, the finalist winner will have a high chance to become Valve's Employee and working with their favorite game but that require the plot and quality of the animaton must be outstanding. He (Fanic) recently made some SFM videos about Overwatch. And there is a video for entertainment purpose called "Tracer's Highlight Rare Easter Egg [Overwatch SFM Animation]" , which feature a few memes like Shooting star music, iDubbz TV's "gay" reference and Tracer dancing in her own butt. But thing that worth mentioning here is Tracer's dance in the video is oddly similar to the dance emote of Tracer in the recently Overwatch Anniversary Update. Furthermore, the update just live nearly 1 week but the video was made 4 months ago and it got more than 11000 views. So there could be a few possiblities (my own conspiracy): 1. This is just a mere coincidence, because both were based on the Charleston dance of the 20s. 2. Blizzard is secretly recruiting or collaborating with some or maybe many animators on Youtube to spread or leak their infos about the upcoming updates like they did on Reddit under the name of any Redditers. I put the source to the video down below, you guys can check it. And you guys can also check out his channel too, he is just a small Youtuber, support him if you can. The 1st link is the link of the video itself, the second one is Fanic's channel. The decisions belong to you guys. Have a nice day. https://www.youtube.com/watch?v=BQ7ek2hFSAg https://www.youtube.com/channel/UCp2p0bwcBlAQVSmoGqWX5cA
Blizzard app idea
by
Shaggy
Warning long post >.< Not sure if anyone has thought of this already or even if this has been discussed. If not then hopefully this idea could become a reality. So I was thinking about how it would be a pretty cool idea if Blizzard made a mobile app that connected to your Battle.net account kind of like how Under watch requires you to sign into your blizzard account so it can access certain info. But like what if on this app you could access some content that you would normally be able to access in game such as Whisper messages and friend requests and even view highlights that you saved. In the official overwatch forums and In a recent lifestream during the overwatch anniversary Big daddy Jeff talked about a feature that might be implemented into the game to save and view your highlights. So i thought that would be cool if you could aces stuff like that on the go say if you wanted to show your friend your sick highlight that should've gotten POTG but didn't. You would easily be able to access your Highlights through the app instead of having to record the Highlight from your computer with a capture card or recording software or recording the screen with your phone. Another thing that could be used in the app would be like viewing your stats like hours and stats on a hero or checking your SR and see if decay is bringing it down. This app doesn't just have to be for overwatch either. I don't play too many other blizzard games besides overwatch so I'm not sure what content you would be able to access for WOW or HOTS. But it would be cool if you could view if your friends are online. Like the Blizzard game launcher. Maybe even message them that you'll be online soon via whisper. I just thought of these few things for a blizzard mobile app. Hopefully some reads this and comes up with some more features that this app could potentially have. Hopefully we can make this a thing if it proves to be a worth while thing that could actually help players in the community and deepen the experience for overwatch and blizzard games in general. We all know blizzard is capable of making apps (hearthstone and the Blizzard Authenticator app) so why not make one where you access content from the game itself? http://www.rebootreload.com/overwatch-jeff-kaplan-forum-questions/
Overwatch Skin Gaining Chance
by
BitzOutBoy
Okay, I wanted to talk about a few things here. First of all, I think that fans should be able to send skin/character ideas to Blizzard just for the creativity (I don't think I'm alone on either of the things I'm going to say). For example, https://www.google.com/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=0ahUKEwiwm47f96LTAhUJ32MKHaHVAIMQjRwIBw&url=http%3A%2F%2Fkewminus.deviantart.com%2Fart%2FOverwatch-Junker-Zarya-skin-concept-662485876&bvm=bv.152180690,d.cGc&psig=AFQjCNEWSzY77_4aztKG1eIn0qmCKC4JcA&ust=1492224481322585 (sorry for the length). Secondly, I don't know if it's been mentioned on here, but I think there should be a week in the year, once a year, to get every past skin possible; it would be how you normally do it, loot boxes. The reason for this is that new players can't even have a chance to get skins that they wanted to get before buying the game, but they can't because that event is over and it's unlikely that the same event would be repeated with the same skins being the only reward. This is because the Overwatch team has to also keep older players interested. So I have a couple of options/ideas to think of if they are possible: - Have a week where players have all event skins able for purchase and unlocking by lootbox. - Every time there is an event for a holiday/misc events (summer games), add skins, but also make previous skins able for purchase and unlocking. - Have every skin, spray, voiceline, etc able for purchase and unlock, but make them significantly harder to get than legendaries (with chance) and for purchase the price would be tripled (common: 75c, rare: 225c, epic: 750c, legendary: 3000c) like they normally are.
Heroes of the Storm is Borrowing Some of Overwatch's Style
by
Schleepwaker
While Blizzard, creator of both Overwatch and Heroes of the Storm, does a good job of giving each of their games a unique feel and style, they do a good job of including tiny references in their property as a friendly nod of acknoledgement for the existence of the other games. A perfect example of this would be like how in Overwatch, you can unlock player icon's that are the symbol for the three classes in Starcraft (another Blizzard game), Zerg, Protoss, and Terran. Similarly, Heroes of the Storm itself made the bold move to include playable heroes that are alost direct ports of the characters themselves from Diablo, World of Warcraft, and Overwatch, so you can play as Overwatch's Lucio inside of Heroes of the Storms' MOBA style. Recently, they decided to make the move to accompany Overwatch in the Loot Chest craze. Now, everytime you level up in HOTS, you will earn a loot chest, just like you would in OW. The contents of the chest will contain 1 item of at least Common, Rare and Epic, matching the quality of the chest. Accompanying this change, they are planning to add a lot more cosmetic items to collect, and as well, they are planning to adjust the leveling system. "Before, it was too easy to gain early levels, and later levels felt like a slog. The patch changes all of that by slightly increasing the experience needed during the first few levels, then reducing the amount needed to progress later by approximately 75 to 80 percent. Blizzard is also removing the Player and Hero Level caps. Your Player Level will now be the sum of all Hero levels, so whenever you gain a new Hero level, your overall Player Level will increase too." https://www.engadget.com/2017/03/29/heroes-of-the-storm-loot-chests/
DoomFist Concept
by
Evertin
DoomFist is interesting because we know little but a lot about him. We have seen the Savior, Scourge, and the Successor. But we don't know which one Blizzard will put in the game. This is just my idea of Doomfists stats and moves. This is not official but this would be really cool if they add these in the game and make it more different than what we've seen. people are debating on whether the Savior will be in the game or the Scourge or the Successor, but i have an idea to add all three. Let me explain.(P.S. they will all be melee only so keep that in mind.) First we'll start with the Savior. -I'd say he'd stand at about 6 foot tall(size will increase throughout the Fists) -His HP can be 500(this increases throughout the Fists -He can be voiced by Terry Crews(Love him!!!!) -His melee speed can be about 1 seconds a swing. -His attack power will be 70 HP per swing(this ,like the size, will also increase) -His Shift ability can be Ground Pound, basically a pocket sized EarthShatter, Cool down can be 10 seconds and is has an area 5 meters around DoomFist. Also, the enemys are stunned for 2 seconds. -His E ability can be Metal Crunch, allows all metal based enemy's to take 100 damage.(Bastion, Zenyatta, Genji, Reinhadt, Orisa, and D.va will be effected by this) Anyone else will lose 50 of health. Cool down can be (due to over powerness) 23 seconds. (note none of these ability's will not continue to the other Fists.) -His Right Click can be a Fist launch sort of thing, Reinhardts charge but no pin only damage and a shorter amount of time based on how long you hold down the button. This also has the knockback. When held down you are immobile and cannot be stopped by taking damage. -His Ultimate's (yes 2 Ultimate's) are where he gains another fist and his damage in doubled and so is his attack speed(He cannot use any of hes other ability's while this is active) and his other ultimate is when he becomes *drumroll* the Scourge. when you become the Scourge anyone (yes i mean anyone, including teammates) within 5 meters will take 25 damage. Now with out further a due lets get on with the Scourge. The Scourge can stand at about 6ft 5 inch's tall. -His HP will increase to 600(will also increase) -He can be voiced by The Rock (Idk why i just think it would be awesome) -His swing can stay at 1 second a swing -His melee can be about 80HP a swing -His Shift ability can be Pile Driver, his fist will spin for 3 seconds knocking back enemys and does 40 DPS within a 5 meter radius. His speed increases and cannot perform any other abilitys. Cooldown can be 12 seconds. -His E ability(personally one of my favorites for some reason) is SmokeScreen, causes a blinding smoke out 5 meters from where deployed but follow Doomfist for 2 seconds.(basically you activate the ability and it stays active as you walk around for 2 seconds then stops) Speed is decreased and negates all wall hack abilitys.(hard counter Hanzo's Sonic Arrow and WidowMakers Ultimate) Cooldow is 15 Hits(every time you hit an enemy it charges up). -His Right click can Talon's Luck. Anyone who is associated with Talon (Reaper, Widowmaker, and Sombra or any future talon Heros) will gain a damage increase anywhere on the map. The Cooldown can be 15 seconds. -His Ultimates can be the same as the Saviors but you level up to the Successor. But the damage when "Leveling" up can be increased to 50. Now for the Successor. The mysterious Fist. Lets get started. The Successor can stand at around 7ft 3 in tall.(Don't ask about the weird 3) -His HP (due to his power) can be 750. -He can be voiced by (insert person here) -His melee speed can stay at 1 second if not .7 seconds per swing. -His melee power can be increased to 100 HP per swing. -His Shift Ability can be the surprise of Ground Pound from the Savior but the radius is increased to 10 meters and are stunned for 4 seconds. Cooldown can be 25 seconds or higher. -His E Ability can be Reconstruct which is where he can give 75% health to allys or himself if his or allys health is critical. -His Right Click can be I can't think of anything at this time please leave suggestions. -His Ultimate can be the Double Fist idea from the other Fists. If there is anything I have missed please let me know and if you have any other ideas to add on. Thank you for reading
New