How does matchmaking work?

Is it based on just infamy, or does it have something to do with your win streaks? I don't know, so if anybody can help, it would be appreciated.

 

Only devs know for sure but it is mainly infamy based (on ranked matches), it also tries to put the same mks on both sides (because we forummers asked so much, now must of us realize it was a mistake).

 

Match making works by matching 5 friends and 5 unfriends. when they are matched the battle starts and we fight on our boat to win the game. If we lose we start over again. Hope that helps.. keep sailing.

 

Thanks!

 

Infamy.


EDIT: what's with the likes?

 

It's obvious that BB matchmaking (MM) takes ship levels and an infamy range into account, but I've heard it's also based on some variation of the Glicko player rating system.

If so, the basic idea is the more you win, the higher your rating goes (and vice-versa). As your rating increases, the more you're expected to contribute to your team. In other words, your competitors get tougher and/or your team gets weaker.

A rating system would explain why the vast majority of players have the same win percentage - within roughly 1%. Otherwise, without such a rating system, many top players would likely have an 80% or higher win rate, get bored, and move on.

A rating system is intended to keep games more evenly matched and more interesting .... but, it can also be frustrating if you're used to winning every match.

So, when playing these types of games, it's best to have the right mindset. If your team is terrible or your opponents overpowering, it's a sign you're rated highly and doing well. Eventually, with some time and skill, you'll progress far enough to be able to stand toe-to-toe with anyone, and have a shot at winning those end-match 1v2s and 1v3s.

/sermon ended. cookies and donation basket in the back.

 

Ok I'm too tired to actually google what does the Glicko rating system mean but let me explain how the BB matchmaking works and why it makes perfect sense that most players have ~50 % win ratio.

Matchmaking in Battle Bay is based mostly on infamy and just a little bit of ship mk level. Win or losing streaks don't affect the matchmaking, they are only included when shuffling the teams from already picked 10 players.

Infamy based matchmaking actually drives everyone near 50 % win ratio because if you win more than lose you will gain infamy and always face tougher opponents until you can't win anymore. Then you lose infamy until you get back to comfortable range again.

Then what about the best players? They are also near 50 % win ratio because they play against other "best" players. If someone at 6k infamy had 80 % win ratio that would mean that some other 6k player should have only 20 % win ratio.. At least theoretically, right?

Now some of those 6k players do have win ratio slightly higher than 50 % because they are forced to drop to 4k every two weeks. So for some time they get to battle with players who are below their level. After some days they usually have climbed high enough to be matched with each other and win ratio goes down near the 50 % again.

 

It doesn't work.

 

Relax gringo, it was sarcasm.
On a serious note, the mmatchmaking was fine during the first tdm event, before exploiters started playing with gear score. So there's still hope.

 

Ok. I'm ready. Oh, just FYI, player rating algorithms like Glicko-2, ELO, and TrueSkill are used in a variety of games from League of Legends to most Xbox live games (like FIFA).

Based the confident tone of your post, I must assume that you have personal first-hand knowledge of exactly how BB MM works - knowledge that is indisputably, irrefutable fact beyond question.

So, I apologize for my educated guess that some other rating system might be at work.

Ok, I'm following you .. for most players.

Theoretically? Umm, I'll give it a go - Assuming the 80% win rate player only matched against the 20% player, won 4/5 games (or 80 out of 100), and the player with a 20% win rate didn't lose infamy after losses or was still able to match the 80% player no matter how big of an infamy difference .. then theoretically, yes.

But, if a shark eats a smaller shark, does that mean there are no other smaller sharks to eat? Why would the success of one person be carried solely by one person? You lost me there.

Yeah, 'beginning of season only' players would have an inflated rate. Ok, I'm with you.

However, the assumption that all of the best players have equal skill, or at least an equal chance of losing, in a random matchup (based on nothing more than two infamy team totals) is beyond the limits of my imagination.

I routinely match against the top players on the leader board with well over a thousand infamy difference, so I'm not sure they only play the other best players. It seems more like a 1 or 2k pool of players or so.

Perhaps there's always another top player matched against them on the other side, and 8 other players there as cannon fodder. So, maybe.

Although, to use a sports reference, are Durant and James going to win at a 50% rate playing each other for 100 games? Doubtful, but I suppose it could happen. What if they played others too. Harden? Curry? The rest of the top 50? It seems unlikely that every single one would have something close to a 50% win rate. What about #1 vs #30? Still 50/50 with 8 random players teamed off?

What about fleeting? What would happen if you put the two best on the same team versus the 3rd and 4th best (with 6 randomly assigned players)? Would that be a 50/50 split? Probably not, but maybe .. what if they got matched against the 15th and 16th. Still 50%? What about a random team of 5 not used to playing together?

Since not everyone follows sports, what about sharks?

There are going to be a few, maybe even 6 or 7, that are the top of the food chain. And, they might fight each other occasionally, and they might lose occasionally, but combined with all their other fights against smaller sharks, it's not going to be anywhere near 50% - with random players selected then teamed off.


Anywho, as I said, you obviously have intimate knowledge of the internal workings and coding of BB MM, so I'm not questioning your word as gospel.

But ... it still doesn't make perfect sense .. or any sense .. especially when you look at matches and try to guess how they matched teams under the theory that it's based "mostly on infamy and just a little bit of ship mk level".

The ship levels usually match, but teams would be very different if you selected them based on infamy totals.

 

So, infamy is simply a funneling tool? Then it's based on some sort of balancing .... or player rating system .. interesting.

I'm curious how you know the order of the selection process, but I don't see any reason why it couldn't work like that.

 

Yeah, as a lawyer, I have to watch writing hearsay and opinions as fact without having an authoritative source to back it up .. but such is forum life.

 

Unfortunately I don't have the time or motivation to answer everything you pointed out but let me tell how @YerJokinArnYer or I know some of the details behind matchmaking system etc.

It's all told by the devs on this forum (and the old forum, which is no longer accessible). There are so many posts here so it's really hard to find individual posts for source anymore but somewhere in the thousands of messages the devs have provided here you can find that information. Don't underestimate how much you can learn from this forum.

And so far we haven't got any reason to believe devs aren't telling us the truth so we can assume everything they say is in fact true.

 

As I said ... I'm sure your words describing what you remember the developers saying on the inaccessible old forum however long ago are still gospel today.

As a former accountant, I enjoy numbers and figuring out why they work the way they do.

So, you can imagine my excitement when I saw the same exact patterns in BB that exist in almost every other player ranking and matching game that's come out semi-recently.

But, alas, I understand my mistake now .. It was simply one of those astonishing coincidences (that probably happen every day ... somewhere) that the win rates are identical to those games and that team matches had so little to do with infamy totals of the respective teams. I was excited to think that the MM was based on something more intuitive like a player rating system.

Anywho, I suggest you guys mention it at the next team meeting or shoot them a text telling them to check it out .. According to several research papers (that you can find with a 2 second Google search), these systems are the gold standard for matchmaking games now.

 

Let me finish this with this picture
I came back to play this game for just checking and found out the matchmaking is still trash. It's even worse now I'm able to play with suckers with lower than 500 infamy. Fantastic! I watched he died himself so bravely. Devs.. I know you are monitoring articles. Thanks for amazing job and please let me know when you guys will stop servicing this stupid game so I can request refund

 

Hey man, there's nothing wrong on that picture just a tanker placed in the place he/ she really belongs