I think this is a new suggestion. Why bootrate % as is have some snags. This is something to think about programming in an Unity mod as well. And I'm not going to put it in Uservoice, partially since I don't have Uservoice account, but also since I'm not sure myself of the best way to deal with them yet and want others' inputs. I'm offering these suggestions but this is not at all comprehensive or the best ways to deal with them. Anyhow, I'll start the list.

• Snag: You're booted in a game where you turn into AI while booted and come back. Coming back is in truth worse than not coming at all, as you can get booted again. You can get booted twice in the same game, and it counts as twice as well. That's how some experimental alts have over 100% boot. You shouldn't be punished for coming back to a game!
  
  Answer: The easy (as in easy to progamme) answer would be just to set the boot to worth 0.5 boots if you come back, but that's pretty abusable, and still leaves a lot of snags.
  
  The middle answer here'd be to multiply the boot by the % of turns that you were gone for. Or perhaps set it as % of time. So if you were gone from turns 3-5 in a game that lasted 8, instead of getting one boot on your profile, it'd be 0.375 boot.
  
  However, games hinge more than just on how long you were gone. The hard answer'd be if you factored in some army counts and income and lands in team games where you left to see how far you set back your team. And outside of team game, this would be important in FFAs since your opponent suddenly becoming stupid AI would tip the game for you if you easily conquest them. And what about custom scenarios?! I don't think I'm experienced enough to make an equasion like this, I would leave it to Beren or someone who knows their stuff on this. But that's the hard answer. And it still wouldn't be flawless, likely. But it'd be better.
  
  
• Snag: being booted in a team game or FFA is the same as getting booted in a 1ag1. It shouldn't be counted as the same, since it's one thing to forget a game with your friend where autoboot was accidentally set (since it's the default option! I have problems with it being default as well but that's another tale), a totally different one if you abandon your teammates in a ladder game. Yet they're counted as the same!
  
  Answer: The thought is clearly to make boots in team games worth more. But how much more? Getting booted in a normal 2ag2 harms your team much more than being booted in a 5ag5. I made a small equasion to deal with this, assuming going off the assumption that getting booted in a 2ag2 should be worth 2x than a boot in a normal 1ag1. I made a small equasion to determine the multiplier in a rational way.
  

  1+((1/T)*2); T = team size
  2 -> 2x
  3 -> ~1.7x
  5 -> 1.4x
  

  However, I thought about this for a bit, and wouldn't that be offset by how many players you would be disappointing? If you're booted, it might not matter as much to your team in a 5ag5 than a 2ag2, but you'd be disappointing 4 folk instead of 1. So perhaps it balances out and should be 2x for always?
  
  Then factoring in custom scenari and perhaps lucky autodistributions, which I've no thoughts on where to begin.
  
  One thing I do know, is that abandoning your teammates in ladder games should count for another 2x multiplier. That's pretty inexcusable if you do that for something so important.
  
  NOTE: all of what's written above is assuming that all teammates don't get booted at the same time, since if they do, then it's as harmless as getting booted at 1ag1. However, this might need some thinking as well, as for exemple one exasperated player could mail the other teammates expressly saying s/he won't go on with the game and that would mean that it's pretty pointless for them to go on, so they get booted as well. Obviously this prick should be punished in bootrate more than his teammates, but how can the system tell? Perhaps make "Report" button for it but instead of banning or suspending him, it just adds to his bootrate if you give proof that he is the reason for abandoning the game? But that's a hassle. However, this applies to very few games, so it doesn't matter so much.
  
  
• Snag: Being booted in "Surrenders must be accepted by all players" games. Seriously, it sucks so hard that this is the default option (but what's default and what's not is another tale). This is where most boots come from on my main. Wanting to leave a game where you've lost in a FFA or diplomacy should not mean waiting for the other lazy players to accept it. You pretty much have to pay to leave some games with an extra boot on your profile. Why!? I kind of understand the grounds for putting this in team games, but if you get booted because of this option in some 1ag1 where someone is trolling you by not eliminating you or some FFA where there's always going to be some lazybones who don't heed such things, that should be different!
  
  Answer: Preferably, remove the "Surrenders must be accepted by all players" option f̶r̶o̶m̶ ̶W̶a̶r̶l̶i̶g̶h̶t̶ from the default settings. Most times this happens, it's in games where this setting is overlooked; someone forgot to tweak it to insta-capitulate. This is unlikely to happen though :/ Fizzer defends his default settings hard.
  
  And, obviously, make it weigh less than other boots. But how much less? IMT, The multiplier should be [(1/P)*(1/T)-0.005], where P is number of players needed to accept surrender before it works, and T = number of turns spent waiting for others to accept surrender, and negative values = 0. Perhaps something should be extended to other players who get booted from these games since they know it's futile to seek capitulation, but I have no ideas how to do that while still punishing those who did get booted just since they forgot about the game.
  
  As for the players that get trolled in normal team games with "Surrender must be accepted" to basically force them to get booted, there should be some detection system of stagnation, and/or perhaps a Report button like earlier suggested. You might say that noone would care enough about report button, but some folk get really picky and angered by such things. Notably, this clan before it faded away (https://www.warlight.net/Clans/?ID=286), could boot you if you ever got trolled like that! IDK how big of a hardship this is though, troll games like that. I hope I'm not giving anyone any ideas...
  
  
• Snag: I used to be a pretty lazy player who used to play a lot of games but get booted in a lot of games as well, since I lost motivation to finish losing games with my teammates or my friends accidentally turned autoboot in a game or I couldn't surrender in them because of infernal demonic "Surrenders must be accepted by all players" option. Then I retired, and came back a year later. Now (̶a̶f̶t̶e̶r̶ ̶c̶o̶n̶v̶e̶r̶s̶i̶o̶n̶ ̶t̶o̶ ̶I̶S̶L̶A̶A̶M̶)̶ I'm a tranquil guy who drinks coffee sometimes and plays much less games (and thus handle and focus on them much better) and out of last 2 monthes, I got booted from 2% of my games. But my profile still says I got a 25% bootrate.
  
  Answer: Make an alt and make it your new main. But that's a big hassle and you can't even make any good games with alt until you've earned as many points as there are atoms in your body, and there's absolutely no grounds why to put it on player, when it can easily be solved by the system, by making it seen in the profile perhaps of how often you got booted in the last 30; 50; 100; 250; 500; perhaps any number you want to see, in addition to time spans: how often you got booted in the last month, two monthes, so on. It's easy!
  

I'll likely think up of more and add to this list, along with community suggestions as well to reasons why bootrate isn't a flawless way of sizing up how likely someone will get booted from a game.

Edited 3/18/2017 23:51:56

Yeah or you just not get booted and join the UNBOOTABLES clan?

This thread was too ribbit, I could not read it.

Maybe just don't get booted from games where turns take 5 minutes tops? What are you doing, jacking off at the same time?

You shouldn't be punished for coming back to a game!

You aren't punished for coming back, you're punished for not taking your turn again. If anything, getting a second chance and failing again should punish you more, don't you think? As a society, we're harder on criminals who re-offend after being released from prison than we are at first-time offenders.

I used to be a pretty lazy player who used to play a lot of games but get booted in a lot of games as well, since I lost motivation to finish losing games with my teammates

This sounds like the root of your problem.

I got booted from 2% of my games. But my profile still says I got a 25% bootrate.

I think it would be reasonable to only track the boot rate from the last X months, rather than all-time. It would allow people to change over time. I would agree with that change.

it sucks so hard that this is the default option

This will probably be changing to not be the default anymore in the new Unity client.

being booted in a team game or FFA is the same as getting booted in a 1ag1.

I can understand the logic of this. I'm not sure what's the right thing to do though. Part of me feels that we should just take the more severe penalty you're talking about and apply it to all games equally. Why should people get a free pass (or lessened punishment) if it's a 1v1? We should be trying to eliminate all reasons people get booted anywhere, not just focusing on one game type. Perhaps that's too idealistic though.

One thing Mercer really wants is to force people who get booted a lot to go through a training exercise where the game shows them how to surrender and walks them through surrendering in an example game (kind of like how the tutorial walks you through playing your first turn.) You'd have to complete the training before you could join new games, as a sort of punishment for having a high boot rate and also to reinforce people's brains into clicking surrender.

Edited 3/25/2017 21:46:10

If anything, getting a second chance and failing again should punish you more, don't you think? As a society, we're harder on criminals who re-offend after being released from prison than we are at first-time offenders.

The way I see it, most human players above L10 are undeniably better than the normal WL AI, and if they come back and play better than the AI would, it's helping your team, but are still punished for twice. I understand your POV though, and maybe it's better.

Why should people get a free pass (or lessened punishment) if it's a 1v1?

A lessened punishment since often the only guy you're really hurting in a 1v1 is yourself - getting lower rank in a ladder, a lower 1v1 rate, losing a tournament, so on.

One thing Mercer really wants is to force people who get booted a lot to go through a training exercise where the game shows them how to surrender and walks them through surrendering in an example game (kind of like how the tutorial walks you through playing your first turn.)

That sounds terrific!

Fizzer: "We should be trying to eliminate all reasons people get booted anywhere, not just focusing on one game type."

Mercer idea of forced surrender/not getting booted tutorial seems neat. If, then forced time-investment to go through a part of tutorial before playing any further may be the turning point. Other than that you could add penalties of making new games for certain time period, joining new games past X number of games with a notice - that if a player has too many games and has no time to commit, gets booted, than one should not create nor have that much games open at the same time.

The latter may not force people to play, commit or surrender like Mercers idea, but should teach the idea of one joining games he knows he can commit at time. And force player to finish some of his games, before making new ones.

I don´t think there is high correlation between level and boot-rate, but there may be a maximum games allowed related to level (excluding ladders, Tournaments and Coin-games), maximum Tournaments participation's allowed. This can avoid some of boots.

Fizzer: "I think it would be reasonable to only track the boot rate from the last X months, rather than all-time. It would allow people to change over time. I would agree with that change.

I would suggest 3-6 months. But not shorter than 3. If its too long, it would be easier (like now) to make a new-account, if its too short, boot-rate looses its meaning.

Although I would favour deflating boot-rate by monthly-based (12.5% up to 13%), having 1 or 2 year lasting duration and loosing its significant value 50% at least withing 4-6 months. For example 18 month duration and 13% deflation would leave boot-rate value after 5th-6th month less than 50% and after 18th month just below 10%.

I would suggest 3-6 months. But not shorter than 3. If its too long, it would be easier (like now) to make a new-account, if its too short, boot-rate looses its meaning.

As I understood, he meant X as in you can input whatever you want.