The point is the streaks are to obvious, that there is tampering on some level. I have had several college level courses in statistics, I understand clusters, etc. This is not that in the least.
About three years ago (different account) I tracked 500 matches; its about 49.9 and 50.1 - so like 50/50. What stood out was that loss streaks usually outnumbered win streaks and were almost always longer, this from someone who doesn’t tilt easily either. I have no doubt were I too repeat this today (I plan too in OW 2) it will be the same nonsense.
I feel like you didn’t actually read my post or you wouldn’t have replied with this. There are external factors besides tampering that can produce such results.
I’m failing to see how this would in any way prove that there is tampering. Just because you dont “tilt easily” doesn’t mean there arent other external factors involved. Humans are not machines that operate with consistency all day every day.
I’m no stats expert like you but my merely mortal mind does think the streaks are too odd to be completely random. Agreeing with what DeadlyPants stated, the games feel so drastically different from each other. Rarely do I have win streaks that I feel I really worked for and rarely do I have loss streaks where I feel I even had a chance.
Another thing I’ve noticed is my loss streaks always happen after I hit my seasonal peak. I’ve gone on huge win streaks to T500 and then immediately dropped down to diamond within a week. Then I’ll go W/L/W/L/W/L/W/L for weeks and then another huge win streak will happen outta nowhere. Rinse and repeat the whole process every single season.
Well all of that’s true. There has to be some oscillation, right? Your MMR and or Sr are never static, assuming matches are typical.
My understanding was that, Overwatch would award you more MMR if you won then when you lost.
If you win against someone with a higher rank you are awarded more points, and if you lose against someone at a higher rank you lose fewer points.
If someone has more “skill” (however that is determined) than those they play against, they get awarded more mmr and or SR, so they face players with higher scores too, next time.
Ultimately intending the system such that someone should be able to equalize around the same skill as others.
I also believe that the process of getting someone to the right place requires, what amounts to trial and error. By awarding players more points until they reach a point of facing too much resistance ,and start to lose more, so they fall back to where things are more comfortable.(and subsequently overshooting in the opposite direction*) Ultimately resulting in their rank oscillating up and down through the rough ballpark the algorithm thinks they belong in. From the players perspective this would appear to be streaks of winning and losing, alternating.
Again. That is as I understand things. Please do correct me if I have made an incorrect assumption, or have something confused.
True to a point, but again, its more than that. We are not dumb; we understand basic statistics, clustering, etc. and the fact that its a process of testing. The problem is the system seems to resort to unnatural degrees of it. Sure some is player availability, mental state, fatigue, etc. but too many note the same behaviors to be just that alone.
Certainly. There will always be fluctuation in the matches. Some people might be having a good or bad day. Be tired or more alert than usual. Maybe someone decided they really wanted to try a new hero that day. There’s tons of variables that represent external factors that will influence the outcome of the match beyond what is knowable by the matchmaker. You will absolutely have some level of oscillation as a result.
I’m not entirely certain about that first one, but yes the amount of MMR/SR you gain or lose is going to vary based on the match you’ve been put into. But those are established patterns in ELO systems. A grandmaster chess player isnt going to gain much, if anything, by playing against novice players at a very low ELO. Likewise, if they are beaten by a novice, they are going to loss a lot.
Realistically there should be no real need to do so aside from the initial placement of the player. MMR is supposed to be based on a standard deviation from average, so the criteria for matching you doesn’t really need to apply arbitrary trial and error offsets to those numbers to figure it out. You should stabilize somewhere over time.
Now, that’s not to say that this isnt happening. The game could certainly recognize streaks of wins or losses and make temporary adjustment to try and account for some of the aforementioned variables. For example, if you’re tired and just not playing at 100%, it’s feasible that they might want to try and account for that in the matchmaker so you dont just end up with a full day of losses under your belt. I can imagine the same could be true for wins. However, I’m unaware of any data that indicates that this is true, and the devs have certainly never confirmed nor denied it.
However, bear in mind that even assuming this is true, this is quite a bit different than “rigging” the matches to force an outcome, as some people might claim. I wholeheartedly do not believe even for a moment that the matchmaker is rigged to force a 50% winrate. It’s both insanely stupid and really has no benefit to anyone.
Source? Because I am positive you are making that up, lol.
They have said that they will match you against stronger and stronger opponents though, as you win more and more. It’s called a ranking system, and pretty standard in all games.
Yes, ranked games tend to keep players win rates around 50/50.
This happens to me all the time. But, it’s usually more like I’ll win 2 or even 3, or lose 2 or three. But that’s just the nature of 50/50. If I flip a coin, it’s RARELY going to be heads tails heads tails heads tails. And I’m not sure why you would prefer win lose win lose win lose anyways?
I think these peoples opinions also come from a lack of understanding why they are losing, as well. I often see things like, “I played exactly the same and lost 5 in a row.” And there can even be things as simple as like, well, were you playing monkey into a reaper every game? There are so many variables in this game, I think it’s easier for people to just “blame the matchmaker,” rather than try to analyze what they did wrong in those games.
This is just something our brains like to do. The extremes are the ones we REALLY remember. When you have three easy games in a row, all of a sudden, oh it’s the matchmaker giving me really easy wins. You tend to remember those more than the 3 really difficult close matches that you won in a row (even though I’m sure that has happened many times to you as well).
Although also, with the nature of this game, it does seem like close games happen less frequently than blowouts do. Which is another simple reason that it may feel like the matchmaker is just “forcing wins/losses.” (even though blowouts happen all the time, even in OWL, which suggest that blowouts are just part of the game).
I actually was stuck in low gold for hundred of games. I bought a new computer, and was high plat in a week. Mine was more like win win win loss win win win loss win win win win loss win win