Jump to content


Photo

Why 6-3-1?


  • Please log in to reply
20 replies to this topic

#1 The Prof

The Prof

    Colonel

  • NASF Committee
  • 1,515 posts
  • Coat of arms
  • Platinum Scout

Posted 14 March 2017 - 05:30 AM

The 6-3-1 scoring system, which awards 6 points for a win, 3 for a draw, and 1 for a loss, is used in most Stratego tournaments at this site, and also in the real life WCs, but is it the fairest system?  I say no.  In ranked games, a draw is counted as half a win and half a loss for the purposes of adjusting ELO, and as far as I know, there are no complaints about this.  However, the 6-3-1 counts a draw as only 40% of a win and 60% of a loss for each player!  This is because there are 5 points up for grabs, since each player receives a minimum of one point, and in the case of a draw each player gets an additional 2 of the 5 points while the remaining point disappears.  Another way to see this is that 10 draws give the same number of points as 4 wins and 6 losses, a 40% win percentage.  Perhaps this already does not seem right to you.  Here are some additional reasons why I do not think 6-3-1 is the best system:
 
1)    A player who draws against two strong opponents has played two good games, whereas a player who wins one and loses one probably just had one good game.  Yet the two good games count for less, 6 points, than the one good game, 7 points.
 
2)    Draws in tournament games are treated differently than regular ranked games, effectively applying a penalty for players who draw in a tournament.  
 
3)    When a game is at a point where each player must take a risk to win the game, neither player should feel compelled to take a risk that they deem unwise.  If one does decide to risk, then a draw penalty creates an asymmetry that gives a benefit to the passive player.  For example, suppose I put my chance of winning at just under 50% if I risk while my opponent has no chance to win if I do not.  Normally I would ask for a draw and it would be accepted, but if a draw is only 40% of a win, then it is in my interest to risk.  But this now weirdly gives my opponent, who had no chance to win, the benefit of a higher than 50% chance of getting the victory. 
 
The main arguments in favor of 6-3-1 that I can anticipate are that 1) it is beneficial to discourages draws and 2) that giving players the equivalent of half a win could lead to collusion between players to agree to draw in advance of playing out the game.  On the second point, I would challenge anyone to find a situation, whether in Master’s Divisions, WC, or Champion’s League where a 2-1-0 system would give an incentive to collude where a 6-3-1 system would not.  If such a situation does not exist, then 6-3-1 provides no benefit against possible collusion, which I don't feel is likely anyway considering the high integrity our tournament players.
 
On the first point, it is not necessary to discourage draws by penalizing them because it is rare that both players would play for a draw, whether in ranked games or tournaments.  Sometimes the lower ranked player will be happy with a draw, but the higher ranked player won't be.  Playing for a draw often results in a loss anyway.  Draws typically happen when players who are both trying to win end up in the fairly rare situation in which neither can capture all the pieces or the flag of the other without risking too much.  More often, it is the case that a game is decided by a difference of one piece, than it is that it ends in a tie.  Stratego is not football (soccer).  Ties are not so common, as our game really balances on a knife’s edge.  Why should we discourage draws anyway?  They are often the result of a well-played game by both sides and should be scored as such. 
 
The NASF used a 1 – 0.5 – 0 system in our Regular Season event last year, and we did not see any more draws than usual (3 in 85 games).  For our next event we plan to use a similar 2-1-0 system.   If you want to still have 0 points as an option for no shows, then 5-3-1-0 could replace 6-3-1-0 as a system that would not penalize draws.  Are there any reasons I have not considered that make 6-3-1 superior?

  • Lonello, match, --Wogomite-- and 1 other like this

#2 TheOptician

TheOptician

    Marshal

  • Tournament Manager
  • 3,093 posts
  • Coat of arms
  • Platinum Captain

Posted 14 March 2017 - 11:50 AM

This debate centres on whether you think a win and a loss is better than 2 draws.

Consider this in the context of the battlefield.

If you win and lose a battle, you have gained land and lost land - you are net even in land and troops.

If you draw two battles you are also net even in land but your troops have been practically eliminated.
  • Master Mind likes this

#3 sevenseas

sevenseas

    Major

  • Banned
  • PipPipPipPipPipPipPipPipPip
  • 1,037 posts
  • Coat of arms
  • Platinum Marshal

Posted 14 March 2017 - 11:57 AM

I agree a win + loss should be worth more than 2 draws. We should be rewarding players for taking chances and making an effort to win the game. 


I play as Sevenseas & Don't Cry

#4 --Wogomite--

--Wogomite--

    Lieutenant

  • Members
  • PipPipPipPipPipPipPip
  • 578 posts
  • Coat of arms
  • Platinum Marshal

Posted 14 March 2017 - 01:03 PM

I agree a win + loss should be worth more than 2 draws. We should be rewarding players for taking chances and making an effort to win the game. 

 

The Prof has already stated why what you said gives the stronger player a disadvantage. 

 

3)    When a game is at a point where each player must take a risk to win the game, neither player should feel compelled to take a risk that they deem unwise.  If one does decide to risk, then a draw penalty creates an asymmetry that gives a benefit to the passive player.  For example, suppose I put my chance of winning at just under 50% if I risk while my opponent has no chance to win if I do not.  Normally I would ask for a draw and it would be accepted, but if a draw is only 40% of a win, then it is in my interest to risk.  But this now weirdly gives my opponent, who had no chance to win, the benefit of a higher than 50% chance of getting the victory. 
 
 

Two draws considers you an undefeated player yet a win and a loss gives you more points?? I agree with The Prof that two draws are two very good games played where as a win and a loss is a 50% win record and nothing to brag about.

Sevenseas, you said we should be rewarding players for taking chances which some would consider lotto?!? A player should learn how to play the game better and they won't have to take chances. Using risk/reward well is not taking chances, the only "taking chances" that will take someone from a draw situation to a win would be lotto. Why would we want to encourage that?? I am all for making an effort to win the game but let's try to earn it, not get lucky for it. 


  • match likes this

#5 sevenseas

sevenseas

    Major

  • Banned
  • PipPipPipPipPipPipPipPipPip
  • 1,037 posts
  • Coat of arms
  • Platinum Marshal

Posted 14 March 2017 - 01:11 PM

A player should learn how to play the game better and they won't have to take chances. Using risk/reward well is not taking chances, the only "taking chances" that will take someone from a draw situation to a win would be lotto. 

 

I take chances nearly every game and I'm 1000+. I think it's boring if everyone plays like Napoleon.

 

I'm not just talking about lotto, I am talking about close endgames. For example:

Player 1: Colonel, Major, Captain, Lieutenant x3, Sergeant x3, Miner x4

Player 2: Colonel, Captain, Lieutenant x3, Sergeant x3, Miner x4

 

It is even except player 1 has an extra major.

 

If two draws are worth the same as a win and a loss, player 1 has no incentive to make progress in this close endgame. Both players would agree to a draw.


I play as Sevenseas & Don't Cry

#6 TheOptician

TheOptician

    Marshal

  • Tournament Manager
  • 3,093 posts
  • Coat of arms
  • Platinum Captain

Posted 14 March 2017 - 01:14 PM

It is not accurate to say that rating a win and a loss better than 2 draws encourages lotto.

It rewards victory - how a player wishes to go about that victory is up to them!
  • sevenseas likes this

#7 --Wogomite--

--Wogomite--

    Lieutenant

  • Members
  • PipPipPipPipPipPipPip
  • 578 posts
  • Coat of arms
  • Platinum Marshal

Posted 14 March 2017 - 01:16 PM

The stronger player will usually win every time. If two players draw then most of the time they are equal skill. What The Prof is saying is that when these two equal skilled players draw each other, it counts closer to a loss then a win and in turn will only damage your standing in whatever event rather then be a neutral one. The reward  (at least not an anti-reward) should go to the players that could both fight for a draw, not hurt them. Now that would be a match worth watching. 



#8 sevenseas

sevenseas

    Major

  • Banned
  • PipPipPipPipPipPipPipPipPip
  • 1,037 posts
  • Coat of arms
  • Platinum Marshal

Posted 14 March 2017 - 01:21 PM

The stronger player will usually win every time. If two players draw then most of the time they are equal skill. What The Prof is saying is that when these two equal skilled players draw each other, it counts closer to a loss then a win and in turn will only damage your standing in whatever event rather then be a neutral one. The reward should go to the players that could both fight for a draw, not hurt them. Now that would be a match worth watching. 

 

But what about the example I set? It could conceivably be a win for Player 1 but some assumptions will have to be made. It could potentially result in a win for Player 2 depending on how tricky one's bomb arrangement is, and how many smaller pieces Player 1 wastes attempting to figure out his opponent's setup. 

 

Under an equal draw rule, it rewards players for being passive. Player 1 in a equal draw rule will likely not attack as there is no reward in doing so. 


I play as Sevenseas & Don't Cry

#9 TheOptician

TheOptician

    Marshal

  • Tournament Manager
  • 3,093 posts
  • Coat of arms
  • Platinum Captain

Posted 14 March 2017 - 01:39 PM

I disagree that either system encourages or discourages being passive/aggressive.

This is because being passive or aggressive in itself doesn't make you more or less likely to win the game.

What makes you more likely to win the game is how effective your strategy is - and this could be passive or aggressive.

It isn't about rewarding or penalising a TYPE of play, it is simply about measuring a win and a loss against two draws.

One thing is for certain - it is not currently 'unfair' - as it is the same for all players.

Now if I was playing in a tournament that was 2-1-0 as opposed to 6-3-1 and I was faced with a 50-50 shot at winning - as I neared the end of the game this would inform whether I took a chance or not (by weighing my odds of success against the points I could win/lose)

So in that sense 'risk' is marginally rewarded. (But again it's not unfair - as the players are using the same rule book!)

#10 sevenseas

sevenseas

    Major

  • Banned
  • PipPipPipPipPipPipPipPipPip
  • 1,037 posts
  • Coat of arms
  • Platinum Marshal

Posted 14 March 2017 - 01:48 PM

I disagree that either system encourages or discourages being passive/aggressive.
 

 

The point payoff  may influence one's judgement in the endgame, but not anywhere else. 6-3-1 encourages one to take risks in close endgame, where it may otherwise be decided as a draw. 


I play as Sevenseas & Don't Cry

#11 The Prof

The Prof

    Colonel

  • NASF Committee
  • 1,515 posts
  • Coat of arms
  • Platinum Scout

Posted 14 March 2017 - 05:06 PM

If you win and lose a battle, you have gained land and lost land - you are net even in land and troops.

If you draw two battles you are also net even in land but your troops have been practically eliminated.

 

By this reasoning, a player should receive more points for capturing more pieces (troops) or taking the flag (more land), and the player who surrenders with most of his army still intact should get more points than one whose troops die in battle.

 

This debate centres on whether you think a win and a loss is better than 2 draws.
 

 

Is a win and a loss really better than two draws?  Suppose a player in Division 1 who was not expected to do well plays Hielco and Nortrom and wins one and loses one.  You might be tempted to think that he got a lucky win.  But if that same player were instead to draw both of these games wouldn't you think "Wow, that guy can play!"?


  • --Wogomite-- likes this

#12 GaryLShelton

GaryLShelton

    Flagbearer

  • Moderators
  • 5,478 posts
  • Coat of arms
  • Gold Spy

Posted 14 March 2017 - 05:25 PM

The NASF used a 1 – 0.5 – 0 system in our Regular Season event last year, and we did not see any more draws than usual (3 in 85 games). For our next event we plan to use a similar 2-1-0 system.


The Prof, one thing I would ask about is the "0" for participation in the NASF scoring. You didn't really speak about this much, but it makes sense to me. Winner take all as it were. But if you do that, what would be the difference between a 2-1-0 system and a 6-3-0 system? If you eliminate the participation point, you now make your 40% situation a 50% one. 10 draws equals 5 wins, not 4. So should you just be campaigning for the doing away with the participation point?

i77rs4m.jpg

The complete GS&F Rules can be found here: http://forum.strateg...rum-rules-2016/

Draw Refusal Rules, specifically, can be read here: http://forum.strateg...604#entry339604


#13 Lonello

Lonello

    General

  • Honorary members
  • PipPipPipPipPipPipPipPipPipPipPip
  • 2,226 posts
  • Coat of arms
  • Gold Marshal

Posted 14 March 2017 - 05:57 PM

It is quite funny because I was actually present at the (live) meeting in the '90's this system was established. It was actually me personally to object in that meeting. We called it an ALV, which is a general meeting of the membership / the Strategocommunity.

 

I used some of Prof's argumentation too. But we lost the fight (I was spokesperson of our club). Erik van den Berg originated all of this, but I am not sure if he is a player here. Many Rotterdam players are secretive players, so I don't know. And since it was adopted wildely afterwards, I think you can consult with a lot more people nowadays who endorse it. Not me. I'm still with The Prof :D  :lol:  :rolleyes:


  • The Prof likes this
Lo

#14 The Prof

The Prof

    Colonel

  • NASF Committee
  • 1,515 posts
  • Coat of arms
  • Platinum Scout

Posted 14 March 2017 - 05:58 PM

The Prof, one thing I would ask about is the "0" for participation in the NASF scoring. You didn't really speak about this much, but it makes sense to me. Winner take all as it were. But if you do that, what would be the difference between a 2-1-0 system and a 6-3-0 system? If you eliminate the participation point, you now make your 40% situation a 50% one. 10 draws equals 5 wins, not 4. So should you just be campaigning for the doing away with the participation point?

 

It's true that 6-3-0 is functionally equivalent to 2-1-0, so yes, doing away with the participation point would have the desired effect, but then I'd say that 2-1-0 is preferable on the basis of simplicity. 



#15 TheOptician

TheOptician

    Marshal

  • Tournament Manager
  • 3,093 posts
  • Coat of arms
  • Platinum Captain

Posted 14 March 2017 - 06:43 PM

I don't actually think that a win and a lose is better than two draws - or vice versa.

Most football leagues play 3-1-0, where a win and a loss is rated considerably better than 2 draws - so in comparison with that it is a compromise.

The advantage of the 6-3-1 system is that it (mildly) encourages a result, which is likely to mean (marginally) less ties.

Maybe TC would consider using both systems employed in different tournaments.

I just object to it being described as unfair or favouring a certain playstyle.

#16 GaryLShelton

GaryLShelton

    Flagbearer

  • Moderators
  • 5,478 posts
  • Coat of arms
  • Gold Spy

Posted 14 March 2017 - 07:37 PM

Well, I'm certainly no high ranked or experienced tournament player, but I offer this idea from the peanut gallery:

American football has settled upon 7 for a touchdown and 3 for a field goal. So how about 7-3-1 for Stratego here?

It seems 7 points for a win rewards a victory handsomely, and although I spoke in favor of the zero participation point above, actually awarding zero points for participation might have the undesired effect of encouraging draws because it increases the risk of attempting victory.

Whereas on the other side, raising the points awarded for a win one point to 7 would serve to disincline players to draw more than currently because the reward for attempting to win would be that much greater.

i77rs4m.jpg

The complete GS&F Rules can be found here: http://forum.strateg...rum-rules-2016/

Draw Refusal Rules, specifically, can be read here: http://forum.strateg...604#entry339604


#17 Napoleon 1er

Napoleon 1er

    General

  • Honorary members
  • PipPipPipPipPipPipPipPipPipPipPip
  • 2,607 posts
  • Coat of arms
  • Platinum Marshal

Posted 14 March 2017 - 11:23 PM

I take chances nearly every game and I'm 1000+. I think it's boring if everyone plays like Napoleon.

 

 

...hey hey ... what does it mean playing like Napoleon :) ? ...I'm also around 1000 and in our last games I have a win ratio against you of about 2/3 of the games, why? because I learned that against reckless players like you the tip is to let these players feel they are leading the game but surprise them with a bomb in a unusual position where they would most likely lotto with a high piece. After that step my material advantage usually helps me to organize my pieces for board domination ... and then usually you surrender!!! ... so playing like Napoleon is a compliment, do I understand it right? :D


  • --Wogomite--, sevenseas and Fairway like this
If you don't know where you go ... you have a lot of chance to arrive elsewhere ...

#18 --Wogomite--

--Wogomite--

    Lieutenant

  • Members
  • PipPipPipPipPipPipPip
  • 578 posts
  • Coat of arms
  • Platinum Marshal

Posted 14 March 2017 - 11:40 PM

The advantage of the 6-3-1 system is that it (mildly) encourages a result, which is likely to mean (marginally) less ties.

 

The NASF used a 1 – 0.5 – 0 system in our Regular Season event last year, and we did not see any more draws than usual (3 in 85 games).   If you want to still have 0 points as an option for no shows, then 5-3-1-0 could replace 6-3-1-0 as a system that would not penalize draws.

 

 

TheO, being TC would you have a rough estimate as to how many draws you see with the 6-3-1 system after 85 games?

 

 

 

I just object to it being described as unfair or favouring a certain playstyle.

That's just it, embracing what The Prof has brought to attention brings the most organic form of the game by allowing players to not feel a dire need to prevent a natural draw. Many players are willing to risk the game knowing that losing is only two points less where lottoing could gain three for the win.  To many the risk is worth the potential outcome but with a 5-3-1 idea, a true draw will be safe rather then some end of a game fluke that determines a random winner. We are not talking about a substantial amount of draws, as The Prof has mentioned they are few and far between. What he is trying to implement is just a way to protect the draws that do come about occasionally. A lot of us have had draws in a 6-3-1 tournament and to be honest, it feels as though it is a glorified loss. A draw is a well fought game and I don't think it should be treated as a "just better than" lost match. 



#19 sevenseas

sevenseas

    Major

  • Banned
  • PipPipPipPipPipPipPipPipPip
  • 1,037 posts
  • Coat of arms
  • Platinum Marshal

Posted 15 March 2017 - 12:11 AM

...hey hey ... what does it mean playing like Napoleon :) ? ...I'm also around 1000 and in our last games I have a win ratio against you of about 2/3 of the games, why? because I learned that against reckless players like you the tip is to let these players feel they are leading the game but surprise them with a bomb in a unusual position where they would most likely lotto with a high piece. After that step my material advantage usually helps me to organize my pieces for board domination ... and then usually you surrender!!! ... so playing like Napoleon is a compliment, do I understand it right? :D

 

Correct.

 

Sadly, lack of patience is a weakness of mine and often I get bored and rush you. However, this is rarely an issue for me in other games. In fact I fare well against the other shufflers. So indeed I would agree you are  playing a lot better than you used to. :)


I play as Sevenseas & Don't Cry

#20 scottrussia

scottrussia

    Captain

  • Members
  • PipPipPipPipPipPipPipPip
  • 800 posts
  • Coat of arms
  • Bronze General

Posted 16 March 2017 - 04:24 PM

A tie is like kissing your sister.  Anything else need be said?


​Spartan Warriors

KING of the Battlefield!!!!!!





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users