|
|
Post by dinnerplain on Jan 16, 2024 9:56:56 GMT -5
In the final sub-event of the current maxipad set, the first task is win 5 games.
How hard is that?
I have a baby account and an account I play actively.
I've been playing both simultaneously. Kind of.
The baby account I just click "To Battle" and let it play without doing anything - tanking if you like. The senior account I attempt to win.
The results?
After an equal amount of time, both have the same win rate.
How is that possible?
Because the match making will sometimes skew the balance towards red, this happens with enough imbalance that a player missing doesn't impact the outcome of the game. Not only that, because imbalance is programmed in, it would appear that there are games where it is effectively programmed you will lose.
I will add that this is happening in Expert/Diamond, not Champion.
|
|
|
|
Post by killmenow on Jan 16, 2024 10:23:20 GMT -5
In the final sub-event of the current maxipad set, the first task is win 5 games.
How hard is that?
I have a baby account and an account I play actively.
I've been playing both simultaneously. Kind of.
The baby account I just click "To Battle" and let it play without doing anything - tanking if you like. The senior account I attempt to win.
The results?
After an equal amount of time, both have the same win rate.
How is that possible?
Because the match making will sometimes skew the balance towards red, this happens with enough imbalance that a player missing doesn't impact the outcome of the game. Not only that, because imbalance is programmed in, it would appear that there are games where it is effectively programmed you will lose.
I will add that this is happening in Expert/Diamond, not Champion.
in master league you can play ...if you want too lol majority of games are stacked against us anyways (multiple ocho's flattening everyone) 8 out of 10 games are won by the top 2! players (usually heavy pay to winners) and even if you are playing you'll be lucky to be counted in most of the time anyways EG : top 2 doing millions and the rest doing maybe 600k at a push this is because they are maxed out and anyone they cross goes down almost instantly (it's not a fight) this is one of the reasons i see tons and tons of ranged setups because it really isnt worth it so sit idle and take a few shots if someone comes into view....been in many many games where the top 1/2 are doing 8+ million damage and the rest scraping by on 5-600k ...it makes the games pointless... everyone hides and waits for bot cooldowns and don't move unless they are up , half the players sit idle and don't move from the spawn ....seen lokies and skyross's run the first beacon and sit there on that beacon idle for the rest of the game.... not to mention the magical super low level bots that on occasion do many millions of damage with weapons you wouldn't even take a second glance at lol , lvl 6 and below bots in masters with lvl 6 or below weapons with hangers sort of similar to when i started the game lol |
|
|
|
Post by titando on Jan 16, 2024 13:29:31 GMT -5
Thanks for testing. I've often wondered about it. This probably happens in any league. I've never understood why they need to manipulate the win rate. Just create a somewhat fair match and let the cards fall where they may. I don't see how it would impact their revenue. If people have money, and suck or don't suck, they are going to spend it.
|
|
|
|
Post by Thunderkiss on Jan 16, 2024 21:35:49 GMT -5
How exactly would match making "skew the balance towards red"?
|
|
|
|
Post by waestern on Jan 16, 2024 22:03:13 GMT -5
it should also be noted that during programmed losses, your bot speed is slowed, and win streaks it is increased. Take note with your baby accounts after you've upgraded a few levels and it becomes very apparent. Pixo usual maliciousness and own cheating forced onto the players is absolutely bonkers. It's simply not possible gor new players to get anywhere in the game without paying which of course is why diamond and below is saturated with exploiting and tanking.
|
|
|
|
Post by dinnerplain on Jan 17, 2024 4:11:40 GMT -5
How exactly would match making "skew the balance towards red"?
As has been observed before, matches fall into 3 groups: 1) red has stronger bots than blue (skew to red)
2) red and blue have roughly equal bots 3) red has weaker bots than blue (skew to blue)
In scenario (3), the match making skew can be strong enough that 5 blue bots can beat 6 red bots, sometimes 4 can beat 6 but not so often.
Then there's the "tanker influence". I don't know if that is part of the calculations for (1) & (3).
I suspect that part of the calculation for the above is how likely is it that all 6 bots will show up in the game.
The only reason that you can drop leagues is because when you win while tanking, the wins generate a lot fewer league points than the loses where you finish 5 or 6.
|
|
|
|
Post by killmenow on Jan 17, 2024 5:21:00 GMT -5
How exactly would match making "skew the balance towards red"? many games i've played 1-2 games per 10 are skewed to us you'll often find (when checking the results/player profiles) that the majority of MY team are super low level players using basic bots some as low as level 5 (incl weapons) nats/fens/leo's you know the more basic bots and these are normal players who if you observe they go for beacons they push forward they fight so not idlers and tankers , they are basically slaughtered time and again by the opposing team using max mk3's (incl weapons) best bought bots they could have and full teams of them v's full teams of none payers most of them barely scratching anything mk2 at all. now when it comes to those 2 games....WE have multiple mk3 players and a few lowbies ...those mk3 players clean the entire map leaving very little if anything for anyone else to do. i've also observed many times over ....some of my team are diamonds- lower , i'm master league 3/2 so basically they are little more than fodder/target practice and barely worth the effort for those pay to winners but that is the point right there.... people want to win and the more they win the more special they feel tempting them to pay more and more , if you put people on par with each other many of them will quit because they pay to win not pay to be beaten to a pulp by a more skilled player. i also will note that i have 10.5k games played over a long period so i know how to play pretty well and can and have beaten many pay to winners when i do this though i become targeted by all of them lol (not a brag at all because it's annoying) they will literally ignore everyone else and all of them will target me , they will time and again ignore all beacons ignore the rest of my team and chase me around the entire map and i suspect these are paid pixonic stooges , they need fodder ....they need people to lose and get aggravated into spending , they need people to see how great their new stuff is at max level so they don't want people beating them. why do i suspect this? - a few times when this happens i'll just opt out as it gets tedious and pointless and join another game ...low and behold these same people again on a new map- new game rinse repeat , now i left these people in the other game and immediately joined another only to find them in that same game doing the same thing - they target YOU! |
|
|
|
Post by killmenow on Jan 18, 2024 6:53:20 GMT -5
it should also be noted that during programmed losses, your bot speed is slowed, and win streaks it is increased. Take note with your baby accounts after you've upgraded a few levels and it becomes very apparent. Pixo usual maliciousness and own cheating forced onto the players is absolutely bonkers. It's simply not possible gor new players to get anywhere in the game without paying which of course is why diamond and below is saturated with exploiting and tanking. i noticed this before AND again today (bot slowed and damage decreased for no reason other than game rigging) i don't remember offhand the map name , large grassy map where everyone always rushes the center beacon. on this map you have beacons directly left/right depending on your spawn side and in my imugi -2x immune amps- speed , i can always launch from where i am and get to those beacons in one flight without walking and always fast except when game rigging takes place in which case my bot is slowed on the ground and in flight and i fall well short of what is normal , on top of this when i fire at a bot my weapon damage is far far lower than it should be ....normally i can dot up a mk2 bot to it max hp so dead or not if it hides it still dies , on these rigged matches the dot effect seems to only effect 1/10 of that bots hp even if the bot is far lower than mk2 i don't pay so there is no buffs taking place and also i'm not effected by any kind of slows since the game just started and almost everyone is out of range anyways , this takes place every few games and is more noticeable on this map stated because from launch to landing gets me exactly inside the line of the beacon except when rigging takes place |
|
|
|
Post by T34 on Jan 18, 2024 18:01:11 GMT -5
Been saying it for a long time. The Match Maker determines your destiny…. mostly. Provided data to support this years ago and tested it extensively. Years ago the event chains included tasks like 60 or so wins. Good luck playing that many games. Short cut at that time was to start the game, bail in 30 secs (LPQ wasn’t triggered by such behaviour back then). That netted about 5 wins every ten mins while watching TV. The more astute players scripted all this…..
The way I think The MM works is as follows. Note, this is summarized at a high level (but still long):
1. Under normal circumstances, the MM allocates reds and blues according to league cup points. Both sides have similar combined league cup points. This includes all leagues, excludes squad games. High Champions League cup players are disadvantaged by this hence the monthly reset.
2. The MM-er will also balance the league level of players on each side. Something like two high cup players (eg high Champions League), to medium cup players (M1 and Champions League border) and two low cup players (eg, masters).
3. You will occupy each of the positions as per the above. For example, once you will be the strong player in your team, than a medium level player and then the least capable player. This will ensure that you get a fair share of all positions on the table on the winning / losing team. Statistical anomalies and interference by squads and OP tankers/whales disturb this.
4. The match maker will search more than the standard deviation in terms of league spread when there are insufficient players in the queue.
This would work most of the time but normal statistical anomalies, squads, tankers and paying customers outside Champions League can cause a player to lose too many times. Many gaming companies (most likely Pix included) do not want players to get fed up from losing too often so they have provisions to ensure the MM-er provides assistance. Google up the EA matchmaking and see how sophisticated and complex MM-ing is.
The assistance mechanism in War Robots (my suspicion based on observations over the years):
If you win too many times or lose too many times the match maker seems to introduce additional criteria while preserving the above mechanism. It seems to track strength of players allowable within the above criteria. I suspect it is based on damage dealt (not beacons captured for sure). If it wants to help you or beat you down, it will introduce low damage players or high damage players on the other team respectively. You can observe the following outcomes:
1. If you play with a weaker hanger/or don’t deal much damage, your win rate will be the same after the readjustment. League may be lower but the win rate stays. Same goes for boosters, so consider spending on boosters in order to win.
2. High cup Champions League players were (and still may be) somehow exempt form the MM-er assistance by such a design. Due to their high cups, the other team may be loaded with stronger players on average and hence the high cup Champions League player may keep on losing but topping the table on the losing team.
3. If you keep on tanking, after about 4 or 5 games, the MM-er will start allocating easier reds and your average win rate goes back into the 40-ies.
4. I’ve tested this by tanking in a duo squad. Guess what, even with 2 tankers the win rate was the same after the initial readjustment by the MM-er.
5. Also notice the tankers who are tanking on your team. The number of tankers on both teams are usually very similar. This neglects their effects. I am not saying the system identifies them and splits them but that they are identified as low scorers and balanced across reds and blues.
6. If you keep winning and then change to a weak hanger, the reds will swamp you and grab all beacons and the game is over in 2 mins. This happens for about 4 games. Same thing happens if you change from a damage hanger to a purely beacon capping hanger.
Overall, your win rate is determined by the matchmaker and not your performance nor your performance boosters, nor your hanger nor tankers or squads or whatever.
Also, the current points system pushing people up would skew the fairness of the allocations because everyone is pushed up. Low level hangers can be high in the leagues if they play frequently.
A recent anomaly which I haven’t figured out yet but it seems like resource boosters give you preferential matchmaking. 20 boosters netted me a top allocation 17 times!
|
|
|
|
Post by waestern on Jan 18, 2024 21:36:19 GMT -5
all true sir, no doubt its always been skewed, however as the actual consistant playerbase has leveled and there isn't a true wealth of incoming players pixo has interfered more and more with their own bias programing. what REALLY 「fluffernutter」s with me is that they've progressed the ability of ai accounts like you would see in the tutorial to be able to "convincingly" play as teammates in matches. I feel its not so much an issue at the higher league since I think this game is pretty much carried by the large group of commited term players in masters and experts but god damn its frustrating that they drop entire non player teams in lower leagues vs the odd one or two pretenders that cover for dropouts and tankers in top leagues. Unfortunately most 「dookie」ty developers have done this with their clown 「bum-bum」 mobile games forever so pixo doesn't even have to invest in a think tank to come up with all these shady tricks.
Sigh... sorry guys XD pls excuse my frustration, im currently leveling up another alt for the new year so I have to slog my way through every last thing pixo is pulling. I swear they'll never get a penny from me though!!
|
|
|
|
Post by gus169 on Jan 18, 2024 22:13:30 GMT -5
I've played for a long time now and my ratio has always sat just under 50%. Flip of a coin, minus my many stupid choices. I don't think MM is rigged, just incompetent. If you over-analyse incompetence, you can usually overlay a design and assign an intent. I think you guys are giving them too much credit.
|
|
|
|
Post by waestern on Jan 18, 2024 23:33:19 GMT -5
I've played for a long time now and my ratio has always sat just under 50%. Flip of a coin, minus my many stupid choices. I don't think MM is rigged, just incompetent. If you over-analyse incompetence, you can usually overlay a design and assign an intent. I think you guys are giving them too much credit. no i swear it must all be trueee, pixo killed jfk and tupac i've found their deep secret plans in the secret easter eggs on the maps |
|
|
|
Post by dinnerplain on Jan 19, 2024 7:02:29 GMT -5
... 4. I’ve tested this by tanking in a duo squad. Guess what, even with 2 tankers the win rate was the same after the initial readjustment by the MM-er. It is great to know that others have replicated the results of my experiments.
Yes, that is the theme of this thread.
If boosters make a difference them I'm screwed because I've got over 1000 of each of them.
|
|
|
|
Post by dinnerplain on Jan 19, 2024 7:04:22 GMT -5
I've played for a long time now and my ratio has always sat just under 50%. Flip of a coin, minus my many stupid choices. I don't think MM is rigged, just incompetent. If you over-analyse incompetence, you can usually overlay a design and assign an intent. I think you guys are giving them too much credit.
MM is rigged in order to try and manage the frustration/happiness of players. Sometimes it gets it wrong because of tankers. |
|
