It works and it's immersive. Since we can't choose the corporations we fight for it's a little different in Dust.

It's more like the New Eden Corporations are hiring out of a merc 'union hall'.

We're really starting to feel the lack of a contract system in Dust. It would answer a lot of the game's needs in a completely immersive fashion, and if built with an eye to the future could eventually be our primary content generator.

Just spent the night working on missions on many accounts.

One account has 600,000sp and was fighting against veteran players for some reason.

Apparently the small time I played on it I must have done pretty well.

Aside from that, all the other accounts seemed to be pitted against groups that were on paper equally skilled.

The high SP accounts (20mil SP +) saw most matches ending with 6-10 players on either side.

The mid and low SP accounts saw most matches ending with 12+ players on each side and not one instance of redlining.

That is not to say that some matches ended up with clone counts of 20 vs 100 ect, but that was usually due to a few guys that obviously were in the wrong weight class.

Most of my waits for battles were under 3 minutes which is fine for me, as a MAG veteran I used to wait 30-35 minutes for a domination match.

I am willing to wait up to 5-7 minutes if that allows the queue to set up full evenly balanced teams.

---

From a New Player Experience perspective this change will definitely help the new player attrition rate and grow our playerbase.

As a veteran, I can say that the longer wait times and squads backing out of almost every match is a pain in the ass.

Until word spreads that they are not going to be able to dodge other proto squads and stomp on new players all day... the long waits and low player count at the end of match won't change.

We are still doing experimentation, and are now increasing the refill ratio (which is also a factor of not being put into a lost battle), you can't have it both ways guys

Thanks for the work and update. This is in line with what we are expecting. Once the system has run for a few weeks, the heavyweights will graduate to the upper bracket, and vice versa.

In a few threads, the idea of a "leave ratio", that starts at 1, but deteriorates if you leave, goes up to 2(?) if you always complete, and is a multiplier on rewards, just for pubs could work as an incentive to keep players in battles at all times.

That begs the question, are the protosquads dropping because their KDR will suffer, or ISK. Something tells me that KDR is more important.

We are still doing experimentation, and are now increasing the refill ratio (which is also a factor of not being put into a lost battle), you can't have it both ways guys

A long time ago, eve online mission agents were divided into tiers of quality. Higher quality agents paid more isk. The requirements for access to higher quality agents was based on personal standing with said agents.

What if we could place EOM rewards on a system where rewards be reduced or increased? For players that team kill or leave matches early, they would soon find themselves with reduced rewards. Playing and completing matches would increase you're standing or security status and grant you increased rewards over time.

A limit could be allowed so players could quit a match if needed or if they're disconnected due to some other reason. Say the limit is once per every 4 hours you can quit a match early without being penalized.

Sure, we can... there has to be some sort of a cut off line though
Frankly, replacing a squad (or more) that left at the very beginning of a battle =/= putting people into a match that is already over.
Furthermore, I don't consider, say a dom, where one MCC is barely into armour and the other has most of its shields to be "almost over" or "lost", there is still time to turn it around, if the team is willing to try (assuming no catastrophic clone losses on either side). I do consider a dom where the blue MCC has 2 ticks of armour and the objective is red to be more than over. There is quite a difference between putting new players into the two, yes?

That latter scenario seems to primarily happen if queued solo... is the matchmaker simply more easygoing with putting solo players where ever they fit, than whole squads?

Also, is it possible to tweak the matchmaker so, that once a battle has started, it would be less strict? As in, if a battle loses players at the very beginning, it would primarily try to find a squad (or squads) of roughly appropriate size to replace those players, maybe on a widened "scope" so to speak? Or, at the very least, try to put more players in at least within the first few minutes of the battle, length depending on the game mode, obviously.

Thanks for the work and update. This is in line with what we are expecting. Once the system has run for a few weeks, the heavyweights will graduate to the upper bracket, and vice versa.

In a few threads, the idea of a "leave ratio", that starts at 1, but deteriorates if you leave, goes up to 2(?) if you always complete, and is a multiplier on rewards, just for pubs could work as an incentive to keep players in battles at all times.

That begs the question, are the protosquads dropping because their KDR will suffer, or ISK. Something tells me that KDR is more important.

Dear players,

I want to give a short update on what's happening with Matchmaking.

TLDR; it's working, and it's definitely pitting similar players together. We will share more data once we have collected a full 24 hour cycle from DT to DT.

There is a slight increase in queue time, but that was intended as we are certain players will want to wait up to 120 seconds for a good match, instead of 45 for a bad one. Compared to competitors and ladder/ranked gameplay games like BF/LoL, waiting time can be up to 10 minutes in the higher brackets. Again something we can tune, from now on.

We have been messing around with settings all day and none of them have broken anything, the logic is very stable and issue-free.

We are tuning the minimum teamsize to start games, and also the status of a running game, and whether to refill or simply let it die since it is already lost. There is an unresolved issue of big strong squads leaving battles. If this continues, we may have to figure out a way to "stop" that behaviour.

A few reports of uneven sides, which is to be expected.

Complaints of redlining are 1) anecdotal, if we trust our metrics, and 2) will continue to happen, just not as often against new players. The logic uses a Mu_threshold to split the playerbase into two queues, above and below that Mu_threshold. Players below should NOT be fighting players above, unless in extreme cases such as absolutely no other players can be found to start a new battle, or no eligible refill battle can be found.

I look forward to sharing more data, such as average Mu variation per battle, showing that more similar Mu players are fighting each other, and hopefully, that the winning margins are overall getting lower. That is the real indicator of more balanced matches, that the winning Ambush team was on average less clones than before, and the winning Skirm/Dom team, has less MCC health remaining, indicating that the match was closer than before.

Happy hunting

If you are "truly" better and close to the threshold, you will quickly graduate to the upper bracket by winning matches, and vice versa, if you were unluckily good as a new player, you may end up dropping (safely) to the lower bracket.

How much does that one kill cost?

Thanks for the work and update. This is in line with what we are expecting. Once the system has run for a few weeks, the heavyweights will graduate to the upper bracket, and vice versa.

In a few threads, the idea of a "leave ratio", that starts at 1, but deteriorates if you leave, goes up to 2(?) if you always complete, and is a multiplier on rewards, just for pubs could work as an incentive to keep players in battles at all times.

That begs the question, are the protosquads dropping because their KDR will suffer, or ISK. Something tells me that KDR is more important.

Someone suggested that you change the button for cancelling queue from O to +ö. I think this is a good idea.
If you can also make the squad 're-queue' automatically when someone joins and/or leaves the squad, or if you make it so that anyone that joins the squad is automatically put into the queue with the rest of the squad that would be awesome.


There also _needs_ to be a lore reason for matchmaking...
(Oh, I something, something money, come on give me lots of honey.)


Someone also mentioned the idea of a reward for 'sticking it out', perhaps a high 'doesn't leave battles' rating would entitle mercs to demand that their employers reimburse them part of the value of the fittings that they lose in battle? Since it would only mitigate part of the ISK loss it wouldn't be directly exploitable. (It would have to be less than on par with simply selling the things & stuff.)


Edit: Can you make our ELO-ratings from before matchmaking being implemented decay over time, or rather as we play more matches under the new matchmaker, as it would be 'fair' if the old newbie-farming-ratings mattered less than the new up-against-peers-ratings. Oh and can you put our ELO-ratings on the EOM screen?

This was removed for various reasons.

Great points & ideas all round. Love the idea of mitigating loss for mercs who stick it out.

Perhaps when a squad ditches the match like that we can have the ISK payouts spike dramatically based on the end-of-match Mu disparity.

People aren't going to scream overmuch if their payout jumps x2-x3 in ISK/SP (insert amount based on how lopsided) when the mu ratio between teams becomes heinously lopsided.

Because what happens in the end is going to be pro squad wimps out leaving the miscellaneous masochists to pick up the slack. If there was a situation like that you'd see me gleefully playing last man standing vs. 16 bads without complaint.

Rattati, what is your take on the OB mechanics, now that one major part of redlining has been removed?

They are anti-comeback at their very core and are generally used whenever the losing team dares to re-group.
In pubs they also increase the power of squads, which are already overpowered due to teamwork.

We are still doing experimentation, and are now increasing the refill ratio (which is also a factor of not being put into a lost battle), you can't have it both ways guys

To prevent people from leaving, you need to fix the people getting dropped from squad deploy issue!

To prevent people to drop because of hard matches, we should have a coresponding stat, call it the contract violation ratio.

To further incentivize not leaving a match, this stat should influence your payout, so if you leave regularly, you will get less ISK per battle. If you constantly stay in, you will get more ISK. Together with the increase in warpoints for actually pushing the objective, this should hopefully give enough incentive to neither leave, nor AFK in the redline. If people stay for more than two thirds of the match in the redline, they should get no payout whatsoever. And yes I think sniping (in a tank or as infantry) is not helping your team enough to justify any payout!