Just received what I'm hoping is only the first response:
Thanks for the email! Really appreciate the questions!
Let me take a stab and answering some of your questions cursorily. And if you need you can follow up with, Eric and Jerry, the true masters, with more questions in more depth.
1. From what we researched, it's a pretty good basis for a match system that a large number of successful multiplayer game employ. However, just research on what others did was not sufficient. We actually applied glicko2 ratings over all of our historical matches, and found that there was a high and statistically significant correlation between the ratings and the match results. Therefore, we decided that glicko2 was worthy of adoption.
2. We look at ship and team strengths in addition to individual strength, to make sure that we mirror the team strengths of each side as much as possible.
3. For rank do you mean levels? If so, the levels are not directly correlated with ratings. I mean, there is obviously a difference between a lv 45 player vs a lv1, but leveling does not account for actual performance, but only for how long someone has played the game generally, so it's not always the best reference for balance.
4. Yes, but I really won't disclose the details. The reason why we don't, and any game does not, disclose details is in part because then we introduce sample and response bias, and open the door for people to game the system.
5. By ranking system do you mean leveling? Whether progression is necessary is a subjective question and answer. Ultimately, if we think it is fun to do and that the players will enjoy it, then we'll do it. With any progression system, it is to an extent cosmetic.
6. Generally we look at ship and team strengths and balances as well as individual strengths.
7. There are many. Speed, scalability, compatibility with console and PvE, balance, sociability to name a few.
8. It does. Right now AI is a bit of a handicap as you know.
9. Quite a few, but I actually think the hundreds of feedback from players through 6 months testing far outweigh that.
10. Well, if you're thinking levels or say achievements, it is a much better reflection of actual player performance. Levels are not really a good reflection of performance. Level, especially now, has a heavy time component. Achievements is just really farming. People game it (me included).
11. This is a sensitive question, because we are here to serve, and not to judge individuals. But from what we've seen, objectively speaking, community regulating itself has been less than satisfactory. Of course there are exemplary cases and our community is in my opinion the best in all games. I really don't want to get into an anecdotal debate about some good cases of great matches vs some examples of stacking etc though. I don't want to call anyone out, and I don't want to judge individual players etc. If we're all honest with ourselves, I think the number of issues with the old system should be self apparent. And I think if we are all honest with ourselves, we can also be objective with the issues that the new system fixed while acknowledging that it is not perfect.
12. Great question, partly it's performance. This is a bit implied perhaps, as teams and individuals who do well to affect the outcome of the match may be presumed to have better teamwork. I guess what I'm saying is, this is a highly teamwork oriented game, if a team does well to win for example, then we'll reward and give more credit to that. Another part is progression. But here we're thinking of doing more as well. Thinking about an idea where we reward people for staying in matches for example, and not bailing out.
Let me just also say that I really appreciate you writing us. And I absolutely didn't take it as rude. If my answers are too brief in parts or need more clarity, please let us know. Having this type of inquiry and reasonable discussion is great, as opposed to the forum, where people with different voices and valuable things to contribute are drawn out by a few voices who want to shout the loudest. Amusingly, these few voices shout us down with just about everything from balance to game play to now the match system.
Tangentially, I do want to publish more system related stats in a write up, but we want to collect data over at least a week or two. So I'll try to write something up next week. But so far based on stats, we know where some of the areas in the match making system where issues arise. We also know where some of the areas that match making is doing well in. We know how much more efficient it is over the old system, and we know how well we are doing balance wise overall compared to the old system. In the forums sometimes I don't even want to bother with stats or making an argument. I give a stat and got people telling me I just made it up lol. I'll wade back in in a bit to see if people can at least have reasonable arguments.
Thanks! Howard
