It is well known that violation types tend to vary by rank. In my experience, escaping is probably the most common report overall, with timeout pretty close to disconnection in frequency, give or take a little.
Not speculative at all—it’s arithmetic. Maximum is 7.5 minutes, much better than a 20-minute main time, but I rounded off because someone can cancel in the first five moves. On the other hand, your counterargument really is speculative about the behavior of players. Many escaping reports involve long amounts of time because the tab was held open. I’ve seen them.
One argument for Fischer is it might be beneficial for Chess players coming to the game as they’re already using that time. One less excuse for them to not stay!
If people think their game is lost, sometimes they just stop playing, annoy their opponent and get reported. If main time is higher, more time is left to annoy the opponenent.
And to be precise: Stalling (playing obviously useless moves) is not escaping (stopping to play).
Conrad isn’t just concerned about games in his own rank range, so his rank doesn’t matter.
My comments have nothing to do with my game experience. As a moderator for 2.5 years, I produced 147,000 lines of report responses (so I’ve been told). As a participant in the Community Moderation experiment since last Christmas, I have seen and voted on more than 3,000 reports.
As I’ve already pointed out, what you quote has nothing to do with Fischer. It was about large main time in byo-yomi games, and has no intended relevance concerning byo-yomi versus Fischer.
Keep in mind, viewing through a moderators lens will also skew your perception. If all you see is reports it is easy to assume bad behavior is rampant, even when the average user isn’t experiencing it often.
It’s similar to emergency room doctors who see danger in all activities the rest of us consider inane.
But if the popular proposal for Automatch challenges to create a linked CG, and for compatible CGs to be picked up by Automatch is implemented, this is no longer a concern, right?
Agreed, if we’re removing options. I’m generally against removing options, especially when the proposal mentioned above would fix the issue without removing options, but if automatch users really want reduced options, Fischer is a lot more flexible
This shouldn’t be an issue if the popular proposal I referenced above was implemented, as CGs never take long to get picked up
Same; I like keeping 10s/move fischer increment, and then adjusting the main time to customize how fast or slow the game will be. Gives the maximum amount of flexibility with the minimum amount of weird time banking in the endgame where the decisions become increasingly trivial
Yeah, I find it strange why 5m+30s3x byoyomi is popular: if you spend 30 seconds per move, which is not unreasonable on those settings, you’ll be into byo-yomi after move 20. At that point, why not just have no main time at all? As it is, it’s basically just a bonus bit of flexibility if you happen to get into a fight in the first 20 moves or so
That’s what I normally play: 20m+10s/move fischer
yep, that was nerfed to 30s (uneccesarily imo), thankfully partially restored to I believe 1 minute, but I hope that it is reverted to the full 5 minutes as I argued for in the Antistalling and Antiescaping Features Thread
Not as I see it. AFAICT it would only double the load on server resources without doing anything to streamline the kinds of games users are playing. The idea is to make custom games more niche, and push users towards bigger pools of common settings, so that most people find games faster, and only the truly passionate dedicate themselves to leaving those pools in hunt of specific rules and game options.
Double what? That load seems neglectible. (Games offers, how heavy is this??
Doing anything to streamline? At reverse Idea being to help to get games in a wider choice.
Wrong choice IMHO. Destroy what is working well. Instead elaborate on integration (see my post before)
Forcing and formating instead of helping by pushing alternative ways which could be appreciated for their similarities. Not the way I hope for.
I had let this go, since it seems too trivial to fight about. But since so many people, those I respect and those I don’t, seem to think this comment is relevant, I guess I must defend myself. I don’t think there is anything in my posts in this thread that indicate an assumption of bad behavior, let alone rampant bad behavior. My posts speak strictly to the effects, regardless of whether the escape was caused by maliciousness or inadvertence (such as being unavoidably called away or losing power to the house).
If the goal is to minimize developer effort and impact to player choice, couldn’t we remove the “require” option from autofind (rules, time keeping, and handicap)? This would leave the “prefer” functionality for those who want it and which is already developed and working.
It would immediately address the concern about internally split AF player pools. While leaving custom games available for those who want specific features.
This assumes people are using AF because they aren’t as picky or don’t feel sophisticated enough to choose their own settings. And assumes CGs are working fine as the sentiment seems to imply.
It also would not drastically change the UI while leaving the decision on changing the defaults as independent.
(19x19 handicap is as different from 19x19 even) as (13x13 even from 19x19 even)
I would just stop to ever use AF if its would be impossible to turn off handicap
This is something that I suggested in the other recent automatch thread and I agree it would be a really simple way of preventing fragmentation with one easy change.
I also agree with this. I personally think handicap should just be disabled for all automatch games. The whole point of rankings + automatch algorithm is to facilitate even matchups, so having another system sitting on top of that which does the same thing seems redundant.
I think it’s fine to have handicap as one of the more niche settings that can be covered by custom games if a player really wants it.
Handicap isn’t redundant though. It’s the only real thing mapping ranks onto ratings.
If you didn’t do that you’d have to just arbitrarily decide rating bands to roughly maybe winrates and call those “ranks” but they don’t really mean much.
Alternatively you’d have to use another system entirely like on Fox where rank progression/regression is winning/losing X out Y games on a rolling basis. That kind of goes along the winrate route above in a sense.
If automatch finds a matchup with a similarly ranked player, then handicap will not apply in any case. But if it doesn’t, then handicap allows you to find an even matchup against a broader set of players.