[GP/UI] Increase Escalating Spoils to Avoid Stalemates

[FabledIntegral]

Really? That surprises me, simply because in Sequential people will end with 5 cards in which it is significantly more beneficial to kill the person as you are nearly always guaranteed a double cash, in which you could then use to kill another 5... etc. In freestyle it's not possible to eliminate a 5.

I do read your whole posts, and please don't bring the rank thing into this conversation, it makes you look dumb not me. I never even referred to it, and my rank fluctuates all the time, I was even a Major a few weeks ago for a short duration.

I still prefer Yeti's 6/5 proposal. Even if it won't end stalemates *as quick* there is still an end in sight. Sure you may have to wait like 10+ turns to get to that end, but you can position yourself in the meantime.

[jiminski]

Really? That surprises me, simply because in Sequential people will end with 5 cards in which it is significantly more beneficial to kill the person as you are nearly always guaranteed a double cash, in which you could then use to kill another 5... etc. In freestyle it's not possible to eliminate a 5.

I do read your whole posts, and please don't bring the rank thing into this conversation, it makes you look dumb not me. I never even referred to it, and my rank fluctuates all the time, I was even a Major a few weeks ago for a short duration.

I still prefer Yeti's 6/5 proposal. Even if it won't end stalemates *as quick* there is still an end in sight. Sure you may have to wait like 10+ turns to get to that end, but you can position yourself in the meantime.

hmm I had not considered anyone had looked dumb, only that you had missed something pretty key and were pointing out that we had missed something pretty key but very basic...

You guys... stalemates don't necessarily have to go to the point where cash exceeds an opponent's armies.......

.
I apologise, i took your pointing out of this to be assumption that we were not well versed in the fundamentals of Escalator. I could only put that down to your perception that we had not learned them.. and that you had.
. Anyway my jovial style had not intended to get us into a slanging match. let's end it here, with you calling me dumb and me apologising for misinterpreting you?


And Yes, due to the Sequential structure, in the top games players sometimes overtly, sometimes without direct communication, work to halt the progress of a likely kill. As everyone knows the likelihood of an imminent cash and the value, it is easy to calculate the viability of a kill and what will stop its attempt. Now because the odds are crunched exhaustively. players rarely take shots with bad odds, so hangings and open games are far fewer!



My point is that the level of cooperation will increase and perhaps even with such a degree of foresight, as to make my increasing cash series into a stalemate.
Blocks become multi purpose; 100 troops placed on Siam, very simply, could preclude the cash of the last player with 300 from taking any player out... and then we are back to square one. This is not that the game will be lengthened it may very likely be stalemated.

In very simplistic terms (as i can't be arsed to be specific): *
100 cashes and blocks to stop the 120,
120 can't take out anyone due to previous block and nor can the 150 so they work together to block the 190,
190 works to stop 240 but also with 240 to stop 300.
Eventually you have, in effect, 800 troops working against the last cash of 300.

Now i agree stalemates may be less likely due to the troops we are talking about! But I am adding the uncertainty of the human factor: it may be that the strategy of the game changes to anticipate the troops to an even heightened degree and far from speeding up and making the game more aggressive, it becomes more defensive.


* i used my figures as they are in 10's and it's easier for me to add up in 10's.. also to show that i am not against this level of increase but that i have my doubts, i would prefer to question my own suggestion too.
Yeti's figures are very similar and the same thought process could be applied.

[FabledIntegral]

People hardly ever work together like that, and it's very hard for two people to work together on the same block, simply because rarely is there ever two territories to block like that. I mean, that could happen before stalemates happen as well, so... I don't think it's an issue. Either way, I don't want to get into an argument either, wasn't my intention, I'll simply take it as a misinterpretation on both sides...

[jiminski]

People hardly ever work together like that, and it's very hard for two people to work together on the same block, simply because rarely is there ever two territories to block like that....

Freestyle and Sequential are a world apart.

yes people do work together to postpone the end of the game and retain a chance at victory. the fact that sophisticated, multifaceted blocking is difficult is exactly why the best Sequential players are just that.

In reality, high level, Sequential escalator games consist of all players working together to stop each other winning, whilst independently trying to kill them all.

go and have a look through some of those games at the top of the board; you will see that the behaviour which exists now need only be refined.

look at these for example: (remember that much of the cooperation is unspoken too)
Game 1187144 short but still with cooperation
Game 2041005 only 14 rounds but an attempt to lengthen the game was made through cooperation
Game 1815231 look at the debate between key and sub over killing 2 men
Game 1286325 ended in round 15 but after constant communication and debate, including request for block.
Game 1279083 a long game with lots of debate, communication and cooperation
Game 949774 and this one because i was in it and it could have lasted for ever.
Game 752196 OPEN GAME with some of the best players at the time explaining their moves.

[hatterson]

The simple fact is that you have basically 3 options.

1.) Make the cards escalate at such a level that it is *impossible* to get stalemates. This would require something like a turn in being worth double or triple the previous turn in.

2.) Keep it as it is, have stalemates.

3.) Implement something like Yeti's proposal. There would still be the possibility of stalemates but only if the players actually try to make it a stalemate. This would require several (at least 3) players acting in unison to 'block' each other.


Imo option 3 is the best. Option 1 is too gimicky. I wouldn't want to be in a game that's lasted 300 rounds and then have it ended because cards got out of control and I happened to be in the wrong spot to turn in. Option 2 allows for stalemates too easily imo. If you get a tiny bit of cooperation early in the game and you get to the turn ins being worth 50 or so you can usually push to to stalemate indefinitely.

Option 2 means that you're only in stalemates if you want to be and even then with a little clever maneuvering you can probably get out of it.

[jiminski]

The simple fact is that you have basically 3 options.

1.) Make the cards escalate at such a level that it is *impossible* to get stalemates. This would require something like a turn in being worth double or triple the previous turn in.

2.) Keep it as it is, have stalemates.

3.) Implement something like Yeti's proposal. There would still be the possibility of stalemates but only if the players actually try to make it a stalemate. This would require several (at least 3) players acting in unison to 'block' each other.


Imo option 3 is the best. Option 1 is too gimicky. I wouldn't want to be in a game that's lasted 300 rounds and then have it ended because cards got out of control and I happened to be in the wrong spot to turn in. Option 2 allows for stalemates too easily imo. If you get a tiny bit of cooperation early in the game and you get to the turn ins being worth 50 or so you can usually push to to stalemate indefinitely.

Option 2 means that you're only in stalemates if you want to be and even then with a little clever maneuvering you can probably get out of it.

I agree with the essence of the post, but i think that we are not just set into 3 options.
this is why i am still questioning this. If a change is to be made let us beat it out and really consider it! Otherwise it is like buying and moving to a one bedroom house when we might want to have 15 kids in the near future.

My point is that the human factor adds a huge level of uncertainty. It is not a case of playing for a draw, it's that at the top level it is so tight that they play not to lose, with the hope of finding a tiny chink in the defences to win!

So perhaps there is a middle way which anticipates human adaptation without ruining the nature of the game!

Perhaps the answer is in using Yeti's Series at 100 but then making a further doubling of the increase at 2 or 300 then another at 600 and so on. In this way the game has a chance to progress naturally but has more chance at finality; even with the most cunning and flawless play from all in any game.

I just think that we need a change which will last as a solution and if we are lazy about it we will not have one.

[hatterson]

The simple fact is that you have basically 3 options.

1.) Make the cards escalate at such a level that it is *impossible* to get stalemates. This would require something like a turn in being worth double or triple the previous turn in.

2.) Keep it as it is, have stalemates.

3.) Implement something like Yeti's proposal. There would still be the possibility of stalemates but only if the players actually try to make it a stalemate. This would require several (at least 3) players acting in unison to 'block' each other.


Imo option 3 is the best. Option 1 is too gimicky. I wouldn't want to be in a game that's lasted 300 rounds and then have it ended because cards got out of control and I happened to be in the wrong spot to turn in. Option 2 allows for stalemates too easily imo. If you get a tiny bit of cooperation early in the game and you get to the turn ins being worth 50 or so you can usually push to to stalemate indefinitely.

Option 2 means that you're only in stalemates if you want to be and even then with a little clever maneuvering you can probably get out of it.

I agree with the essence of the post, but i think that we are not just set into 3 options.
this is why i am still questioning this. If a change is to be made let us beat it out and really consider it! Otherwise it is like buying and moving to a one bedroom house when we might want to have kids in the near future.

My point is that the human factor adds a huge level of uncertainty. It is not a case of playing for a draw, it's that at the top level it is so tight that they play not to lose with the hope of finding a tiny chink in the defences to win!

So perhaps there is a middle way which anticipates human adaptation without ruining the nature of the game!

Perhaps it is using Yeti's Series at 100 and then making a further doubling of the increase at 2 or 300 then another at 600 and so on. In this way the game has a chance to progress naturally but has more chance at finality even with the most cunning and flawless play from all in any game.

I just think that we need a change which will last as a solution and if we are lazy about it we will not have one.

The thing is Yeti's proposal has these increases. With his 6/5 proposal is that it's a simple geometric sum. If you take the limit as # of turn ins goes to infinity

<bla bla bla math that no one except me cares about>

If there are 5 or more people still in the game (including you) then we'll assume armies are roughly evenly distributed. That would mean each person would have roughly 1/5 of the armies presently on the board. Now assuming no significant battles have occurred that would mean that each player (including yourself) would have roughly 1/5 of the armies on the board. Now if you knew all that math above every time you turn in you will be deploying roughly 1/5 of the armies that have ever been deployed. So previous armies was X and you had 1/5 X. Now you're deploying another 1/5 X. New total armies is X'. X' ~= 6/5 X. You have 2/5X armies or ~1/3X'. All other players have 1/5X or ~1/6X'. In other words you have double the armies of the next person. This means that as long as you're connected to their territories you can eliminate them. Then you can turn in again, eliminate another person and so on.

Now lets assume we're in a stable game with 3 people. Same scenario, roughly = armies for each person, or 1/3X. You turn in and deploy 1/5X. You have 1/5+1/3 X= 3/15+5/15x = 8/15X ~= 4/9x' ~= 44.4% of the armies on the board. Given that the attacker has the advantage over the long run in large battles, 44.4% of the armies is enough to eliminate at least 1 opponent (turn in again) and then get the last guy.

Assume there's 8 people in the game. Each person has 1/8X you still deploy 1/5X. You now have 1/5+1/8X = 13/40X ~= 13/48X' You opponents have ~5/48X' You have nearly triple their armies so you can eliminate 1 or 2, turn in and bla bla bla you get the point.


So like I said:
You can leave it as is have frequent stalemates.
You can use DiMs numbers and just have the game end and the winner be based on whoever gets the lucky turn in slot.
You can use something like Yeti's numbers and only have stalemates in the extremely rare cases where everyone works in perfect unison to block someone. This is extremely hard in the long run and even a tiny mistake by one player would end the game.

I like option 3 the best because it gives you a chance to end the game via strategy and not turn position.

[jiminski]

you don't have to limit yourself to 3 options.

[hatterson]

you don't have to limit yourself to 3 options.

I don't see what a 4th option can be.

Option 1 make stalemates fairly easy.
Option 2 make stalemates hard but still possible.
Option 3 make stalemates impossible.

Option 1 is now.
DiMs numbers are an example of Option 3.
Yeti's numbers are an example of Option 2.

Sure you can create other solutions that fall under option 3 and others that fall under option 2 but the crux of it is picking which option to go with. Then it's number details not a directional debate.

[jiminski]

you don't have to limit yourself to 3 options.

I don't see what a 4th option can be.

Option 1 make stalemates fairly easy.
Option 2 make stalemates hard but still possible.
Option 3 make stalemates impossible.

Option 1 is now.
DiMs numbers are an example of Option 3.
Yeti's numbers are an example of Option 2.

Sure you can create other solutions that fall under option 3 and others that fall under option 2 but the crux of it is picking which option to go with. Then it's number details not a directional debate.

no you are being simplistic; just as an example, a fourth option could have a staggered increase after a number of cashes.



you don't have to limit yourself to 3 options.... you can have another step as i say! but i am most strongly the advocate of not trying to be right but to get it right, as i can not honestly say who is right. Therefore i am the advocate of sucking and seeing.

So would it be possible (if we got that far, we do not even know if Lack is interested yet) of trying various series to see what does solve the problem.

1. Yeti
PreviousCash + (PreviousCash / 5)

100 - 120 - 144 - 172 - 206 - 258 - 310

2. DiM

100 - 120 - 160 - 240 - 400 - 720 - 1360


3. Yeti's for 7 cashes and then a doubling of the increase at various stages after.

100 - 120 - 144 - 172 - 206 - 258 - 310 : next step : (2/5 increase)365 - 447 - 547 - 666 etc


I Agree that the option with least impact to game dynamics is the correct one. But there is no point having a halfway-house. I think that option 1 could prove to solve most stalemates but there could be a significant amount still at the top of the board. I just don't think we are quite there yet... not far away though perhaps.

I am honestly not smart enough to know what the long-term result would be of the mid-range series of increases... and i truly do not believe many are! Maybe you are Hatterson but i would still prefer to find out empirically, rather than on debatable evidence.

[hatterson]

no you are being simplistic; just as an example, a fourth option could have a staggered increase after a number of cashes.

That would fall under either Option 2 or Option 3 depending on how it was worked out.

If it was worked out so that it doubles after 100, quadruples after 200, octuples after 300, etc. then it would be option 3.
If it was worked out so that it increases a set amount after 100, 200, 300 etc. then it would be option 2.

Actually now that I think of it, option 1 and option 2 are really suboptions.

The core debate is a yes or no question. Do you want to have the possibility of stalemates? Yes or No.

If No then implement an increase that will absolutely make it impossible to stop a player from eliminating someone and turning in an extra set thus gaining enough armies to eliminate the rest. DiMs is an example. Perhaps too extreme but that's for another debate. Another example would be something like Yeti's with a higher fraction. Say 4/3 or 3/2 or even 8/5. Yet another example would be Yeti's where the fraction increases either gradually or jumps at set intervals.

If Yes then either leave it the same, implement a step increase plan that makes it slightly harder, or implement something like Yeti's which makes very hard.

[hatterson]

I am honestly not smart enough to know what the long-term result would be of the mid-range series of increases... and i truly do not believe many are! Maybe you are Hatterson but i would still prefer to find out empirically rather than on debatable evidence.

If you stay at the same level then as time goes to infinity you are deploying (increase-1) of the armies deployed so far.

ie, at 2 you are deploying roughly the same number of armies as has been deployed to the point so far
at 3/2 you are deploying roughly 1/2 the number of armies as has been deployed to the point so far
at 3 you are deploying roughly 2 times the number of armies that have been deployed so far.

[jiminski]

I am honestly not smart enough to know what the long-term result would be of the mid-range series of increases... and i truly do not believe many are! Maybe you are Hatterson but i would still prefer to find out empirically rather than on debatable evidence.

If you stay at the same level then as time goes to infinity you are deploying (increase-1) of the armies deployed so far.

ie, at 2 you are deploying roughly the same number of armies as has been deployed to the point so far
at 3/2 you are deploying roughly 1/2 the number of armies as has been deployed to the point so far
at 3 you are deploying roughly 2 times the number of armies that have been deployed so far.

no, Maths is fine Hatt, but what we are attempting to do is predict the behaviour and adapted joint strategies of the best Sequential players here. have a look at the game numbers i listed before, the last excercise Game in particular.
We can not predict the validity of these series.

[hatterson]

I am honestly not smart enough to know what the long-term result would be of the mid-range series of increases... and i truly do not believe many are! Maybe you are Hatterson but i would still prefer to find out empirically rather than on debatable evidence.

If you stay at the same level then as time goes to infinity you are deploying (increase-1) of the armies deployed so far.

ie, at 2 you are deploying roughly the same number of armies as has been deployed to the point so far
at 3/2 you are deploying roughly 1/2 the number of armies as has been deployed to the point so far
at 3 you are deploying roughly 2 times the number of armies that have been deployed so far.

no, Maths is fine Hatt, but what we are attempting to do is predict the behaviour and adapted joint strategies of the best Sequential players here. have a look at the game numbers i listed before, the last excercise Game in particular.
We can not predict the validity of these series.

What I was trying to show is that if you set the fraction in Yeti's to 2 then you *cannot* stop the person as they are getting more armies on the turnin then all other players have combined. That would fully eliminate stalemates.
If the fraction is at something very low, like 101/100 then your turnin has functionally the same problem that exists now, it's a drop in the bucket in terms of overall army size and blocking is easy.

That's why the decision needs to be made (either on the community level or the lack level) if stalemates should be *impossible* or just really hard. Once that is done then you can choose an algorithm and test it.

[jiminski]

I am honestly not smart enough to know what the long-term result would be of the mid-range series of increases... and i truly do not believe many are! Maybe you are Hatterson but i would still prefer to find out empirically rather than on debatable evidence.

If you stay at the same level then as time goes to infinity you are deploying (increase-1) of the armies deployed so far.

ie, at 2 you are deploying roughly the same number of armies as has been deployed to the point so far
at 3/2 you are deploying roughly 1/2 the number of armies as has been deployed to the point so far
at 3 you are deploying roughly 2 times the number of armies that have been deployed so far.

no, Maths is fine Hatt, but what we are attempting to do is predict the behaviour and adapted joint strategies of the best Sequential players here. have a look at the game numbers i listed before, the last excercise Game in particular.
We can not predict the validity of these series.

What I was trying to show is that if you set the fraction in Yeti's to 2 then you *cannot* stop the person as they are getting more armies on the turnin then all other players have combined. That would fully eliminate stalemates.
If the fraction is at something very low, like 101/100 then your turnin has functionally the same problem that exists now, it's a drop in the bucket in terms of overall army size and blocking is easy.

That's why the decision needs to be made (either on the community level or the lack level) if stalemates should be *impossible* or just really hard. Once that is done then you can choose an algorithm and test it.

sorry mate, i have not studied this in detail but are you basing the increase on 2 fifths increase to the last increase? (not 2/5 of the total cash increase) anyway that is only an example, as it is likely not enough.

[hatterson]

sorry mate, i have not studied this in detail but are you basing the increase on 2 fifths increase to the last increase? (not 2/5 of the total cash increase) anyway that is only an example, as it is likely not enough.

The numbers are actually the same. A cash being worth 7/5 of the previous cash (a 2/5 increase) is the same as increasing the increase by 2/5. And 2/5 would be plenty increase in the long run to make it impossible to stalemate.

It would start at 100, then go to 140, then to 196, then 274, then 384, then 538. If you notice the pattern the turn in that is 2 behind of yours is nearly double what yours was (1.96x). The turn in 3 behind yours is worth between 2.5x and 3x yours (2.74x). The turn in 4 behind yours is worth nearly quadruple. You will get a scenario where Person A turns in first and is the weakest person, then someone who shares a border with him will at some point turn in 2 or 3 after him and be able to easily wipe him out.

[FabledIntegral]

So... I was just thinking, what would be really cool is if you could, at say cash 150, increase all bonuses by 10 times... for example in classic, Europe/NA would give 50 armies, SA/Oceania 20, Africa 30, Asia 70. Think about it, it would cause a lot more attacking, autoing last second, accidental hitting each other (in freestyle), especially if the cashes were still at 150 (which means the army count is usually around 700 at this point). Because if you hold Europe for 5 turns... 250 armies... you're friggin' getting more than a cash.

Well at least I think it would be cool.

[hatterson]

So... I was just thinking, what would be really cool is if you could, at say cash 150, increase all bonuses by 10 times... for example in classic, Europe/NA would give 50 armies, SA/Oceania 20, Africa 30, Asia 70. Think about it, it would cause a lot more attacking, autoing last second, accidental hitting each other (in freestyle), especially if the cashes were still at 150 (which means the army count is usually around 700 at this point). Because if you hold Europe for 5 turns... 250 armies... you're friggin' getting more than a cash.

Well at least I think it would be cool.

Good players would know that the increase is coming and defend against it. Then you'd just have large stacks that balance each other and be at the same spot.

[FabledIntegral]

So... I was just thinking, what would be really cool is if you could, at say cash 150, increase all bonuses by 10 times... for example in classic, Europe/NA would give 50 armies, SA/Oceania 20, Africa 30, Asia 70. Think about it, it would cause a lot more attacking, autoing last second, accidental hitting each other (in freestyle), especially if the cashes were still at 150 (which means the army count is usually around 700 at this point). Because if you hold Europe for 5 turns... 250 armies... you're friggin' getting more than a cash.

Well at least I think it would be cool.

Good players would know that the increase is coming and defend against it. Then you'd just have large stacks that balance each other and be at the same spot.

"Good players." People would ditch spots for their own bonuses for sure. It's not hard to take Oceania/SA. And whilst that's happening you can be people would be threatening to auto other armies if they didn't leave. People take Europe etc already in stalemates... it would definitely help a freestyle situation that's for sure.

[yeti_c]

Just to resurrect this - as stated earlier - my series will always eliminate stalemates just as always as DiM's...

As Infinity dictates it must.

However - the series I posted is just a bit slower than DiM's explosive break - which really could ruin your day if you happen to get 2 pair at the wrong time.

C.

[jiminski]

Just to resurrect this - as stated earlier - my series will always eliminate stalemates just as always as DiM's...

As Infinity dictates it must.

hehe yes infinity dictates that at some point everything will happen! however i am sure you must have considered this Yeti; you can not know to what level the human factor and Cultural strategy change will impact.

[yeti_c]

Just to resurrect this - as stated earlier - my series will always eliminate stalemates just as always as DiM's...

As Infinity dictates it must.

hehe yes infinity dictates that at some point everything will happen! however i am sure you must have considered this Yeti; you can not know to what level the human factor and Cultural strategy change will impact.

That doesn't matter though... Inifinity is the overriding force here... the fact is - the logarithmic scale that I have proposed tends to infinity - and thus - the game although may stale out for a few more rounds - will end.

C.

[jiminski]

Just to resurrect this - as stated earlier - my series will always eliminate stalemates just as always as DiM's...

As Infinity dictates it must.

hehe yes infinity dictates that at some point everything will happen! however i am sure you must have considered this Yeti; you can not know to what level the human factor and Cultural strategy change will impact.

That doesn't matter though... Inifinity is the overriding force here... the fact is - the logarithmic scale that I have proposed tends to infinity - and thus - the game although may stale out for a few more rounds - will end.

C.

ok mate! it certainly sounds convincing and we have been beating this out in your absence too!
So you are certain that your number series, without amendment, can supecede the combined forces of all players; working in multilateral co-operation, at the same time as against all. And that it is highly improbable that they can create multi-layered blocks which lead to gridlock?

[yeti_c]

Yes.

Mathmetically speaking the amount of armies received on cash will sometime exceed the total armies on the board... thus ensuring an end...

Although there is a cuteness to this point -> As you can only advance 999 armies - said person may not actually win -> if they don't claim the kills correctly!!

C.

PS - yeah I did read through the thread... in the end though - infinity & logs are all that matter... I can & do see your point... and it is valid - for the short term...

[jiminski]

Yes.

Mathmetically speaking the amount of armies received on cash will sometime exceed the total armies on the board... thus ensuring an end...

Although there is a cuteness to this point -> As you can only advance 999 armies - said person may not actually win -> if they don't claim the kills correctly!!

C.

PS - yeah I did read through the thread... in the end though - infinity & logs are all that matter... I can & do see your point... and it is valid - for the short term...

good enough for me mate! I have been playing devils advocate a little, as i think the idea deserves proper execution!