Board Game Arena

If you've seen my post in "Suggestions" here is some of the key parts again. I'm just going to say up front, There are hardly any games against real players in the top 30. They're all playing the robot. By some fluke I got into the top 20 but will anyone play? No. So you're forced to play players that if you lose then you' losing 35 elo. What is the point?

Here's what I'm going to start doing in three days. Go into games where a top twenty player is playing the robot and ask they why? That they should be playing real players and that gaining elo from playing the robot is ruining the game and making it no fun for others. And if I make enemies THEN THAT'S THE PRICE I'LL HAVE TO PAY!! Someone talk me out of it.

Anyway, here's some recent data:

Of the current top 12 players only 2 have not played the robot at all. Leaving 10 of the top 12 who have played the robot. Of those 10 only 2 have played any real players in their last 50 games. Those two have played the robot 46 of 50 and 31 of 50, respectively. That means that in the last 50 games for each player in the top 12 who have played the robot at all, overall they have played the robot 467 out of 500 games. It's obvious that most of the top 20 are only playing the robot to maintain their position in the top 20.

Continuing, I propose that the current rankings are contaminated with robot elo gain so all previous robot elo should be eliminated. My proof for this contamination:

(these are not in order) In the top 15 in ranking, these players have played the robot:
Played robot 160 out of 180 total games
Played robot 160 out of 161 total games
Played robot 318 out of 545 total games
Played robot 225 out of 256 total games
Played robot 212 out of 311 total games
Played robot 486 out of 596 total games
Played robot 117 out of 130 total games
Played robot 177 out of 228 total games
Played robot 150 out of 169 total games
Played robot 1884 out of 2743 total games

That's 10 of the top 15 in ranking. Two others have played to robot exclusively if their last 50 games.

Conclusion: Players are using the robot to artificially boost their ranking and gaining top spots and staying there.

Someone please explain how this is fair.

Funny that there really are people who spend their time with this apparently silly robot just to get into the top ranking. I never look at this ranking as in luck based games like Can't Stop this probably is never of great value. Normally it should be quite random who is on place 1, isn't it?

Or: Even if there is no robot I would say that the top rankings in luck based games are in some kind or the other manipulated. Restricting the tables to "better" players, switching off ELO once when on the peak etc.

Why is this robot as bad as it is? If it would play better then this discussion would not occur.

Normally it should be quite random who is on place 1, isn't it?

It's true that a game so involved in dice has a much larger element of luck than many other games on BGA. However, the fact that the robot plays so 'poorly' shows that a person can exercise some control over the luck aspect. If it was purely luck then over time players would fall to a mean of 50 percent. However, this isn't what actually happens. Some players are able to maintain a winning percentage between 55 and 60. This means that some players have discovered strategies that help them win slightly more than the average. This is the fun of the game.

Even if there is no robot I would say that the top rankings in luck based games are in some kind or the other manipulated.

Well, some will say that Las Vegas games are manipulated and the odds favor the house (which they do, by the way). But people play anyway, why? Because people want to try to overcome those odds. They know the house odds are small and that they could go on a winning streak. And they can try out their strategies and see how they work. But if you heard that a casino down the street cheats people and it's hard to win there, people would stop going there. If it doesn't feel fair people will stop playing. This is what is occurring with Can't stop. It's not fair so people will slowly stop playing.

Solution 1: increase robot IA (use rules of 28).
Solution 2: remove BGA robot.
solution 3: playing with robot automatically turn ELO off.
my 2p

DrKarotte wrote: Normally it should be quite random who is on place 1, isn't it?.

No it's not "random." Random means there are no choices to be made, and you do have choices. One of those choices is deciding when to stop and that, in theory, should be based on probabilities - people play the odds. I don't win more cause the statistics seemed to be against me and I turned into a chicken on that game. I later (on another site) tried to be a hair less chicken and I actually started doing better, by doing nothing but "rolling just one more time" than I had before. I'm still not great at it cause I'm still a bit on the too-chicken side. But I can certainly tell the difference, which would not be the case if it were completely random. Not to mention that there are also choices about which numbers you take.

There's always ways to work around luck, at least to some extent, making it so it's not completely random. Anyone who thinks the winner is totally random just hasn't found any of these.

But, to the point about the silly robot and ELO, just automatically having it turn off when playing the robot is such an incredibly easy solution to this problem. Or having it just not reward any ELO even if it claims to be on.

Liallan wrote:

DrKarotte wrote: Normally it should be quite random who is on place 1, isn't it?.

No it's not "random." Random means there are no choices to be made, and you do have choices. One of those choices is deciding when to stop and that, in theory, should be based on probabilities - people play the odds. I don't win more cause the statistics seemed to be against me and I turned into a chicken on that game. I later (on another site) tried to be a hair less chicken and I actually started doing better, by doing nothing but "rolling just one more time" than I had before. I'm still not great at it cause I'm still a bit on the too-chicken side. But I can certainly tell the difference, which would not be the case if it were completely random. Not to mention that there are also choices about which numbers you take.

There's always ways to work around luck, at least to some extent, making it so it's not completely random. Anyone who thinks the winner is totally random just hasn't found any of these.

But, to the point about the silly robot and ELO, just automatically having it turn off when playing the robot is such an incredibly easy solution to this problem. Or having it just not reward any ELO even if it claims to be on.

But no amount of strategy or "working around luck" will help you if luck isn't on your side. That's why I don't pay attention to ELO ratings in this game. I've easily beaten experts and have also been beaten by beginners.

I've seen games where players were having a hard time progressing on numbers that occur more frequently (6, 7, 8) but others were somehow having more luck with the outer numbers (2, 3, 11).

Can't Stop is a game that's 90% luck and 10% strategy. That's why if someone wants to become a Master, they have to cheat and play the Robot. The odds of becoming one while playing real people are very, very slim.

And OP, I totally agree about the BGA Robot (this is probably my fourth or fifth complaint about it on this forum). All the mods have to do is simply turn the ELO off, yet they keep ignoring our concerns, for some reason...

Google "rules of 28 can stop" and start counting. It helps !

But no amount of strategy or "working around luck" will help you if luck isn't on your side. That's why I don't pay attention to ELO ratings in this game. I've easily beaten experts and have also been beaten by beginners.

I've seen games where players were having a hard time progressing on numbers that occur more frequently (6, 7, 8) but others were somehow having more luck with the outer numbers (2, 3, 11).

Can't Stop is a game that's 90% luck and 10% strategy. That's why if someone wants to become a Master, they have to cheat and play the Robot. The odds of becoming one while playing real people are very, very slim.

Thanks for your support of the 'something needs to be done' movement, by the way. I would amend you comments slightly, though, that while the game is certainly luck based there is an element of strategy that a player can pursue. For instance I try to look at the percentage of wins in my last 100 games. That lets the odds shake out a bit and reveal the results of my strategy. I'm doing well if I maintain 58 to 60 percent. Even when I go on an inexplicable 5 of 25 streak I can still get back to 55 percent.

Lots of players take low odds chances over and over because it's glorious the one time it works but over time they just average 40 percent wins. I see this game like investing in stocks; those who do well are the ones who build a strategy based on what's likely to happen and then discipline themselves to follow it and take chances only when it means something like gaining some big advantage or at the end when you're going to lose next turn, etc.

And OP, I totally agree about the BGA Robot (this is probably my fourth or fifth complaint about it on this forum). All the mods have to do is simply turn the ELO off, yet they keep ignoring our concerns, for some reason...

One other small point is that along with turning off elo, they have to make it so that playing against the robot doesn't count as a game played also. Otherwise, all the players in the top 20 will simply play the robot once a month and they'll never play a real player and they'll hog all the top spots.

Why is that important? Because BGA promotes the idea of Prestige. A top 20 spot gets you 1500 prestige points. For this reason alone, BGA should fix this problem. Essentially, we have players gaining prestige who don't deserve it and players who do deserve it are locked out.

I think turning ELO off for bot battles is a perfect and simple solution.

I also will protest the 'this is a pure luck game' sentiment. I've heard plenty of players say the same. 90/10 is a gross misrepresentation.

There's tons of strategy and it centers around managing probability. I'd put it more 60/40 luck/strategy, perhaps 70/30. There are certainly times when the dice seem to be against you but if you have a strategy and follow it consistently you will more often than not come out on top.

gdBlaze wrote:

Thanks for your support of the 'something needs to be done' movement, by the way. I would amend you comments slightly, though, that while the game is certainly luck based there is an element of strategy that a player can pursue. For instance I try to look at the percentage of wins in my last 100 games. That lets the odds shake out a bit and reveal the results of my strategy. I'm doing well if I maintain 58 to 60 percent. Even when I go on an inexplicable 5 of 25 streak I can still get back to 55 percent.

Lots of players take low odds chances over and over because it's glorious the one time it works but over time they just average 40 percent wins. I see this game like investing in stocks; those who do well are the ones who build a strategy based on what's likely to happen and then discipline themselves to follow it and take chances only when it means something like gaining some big advantage or at the end when you're going to lose next turn, etc.

Totally agree here. Any game with luck can be unlucky for the best of players, and lucky for the worst of players, in any given game. Really, any given turn. There's another high luck game on Yucata where certain strategies can work sometimes, but they aren't strategies the top players use - I'm in the top 10 and I don't play that way, but I can get beaten by those who do, but not the majority of the time. (In fact, one particular "strategy" turns it into pure luck, which makes the game pointless.)

But some people aren't into that, having luck affect a given game so much, even if in the overall scheme of things there's some strategy. Each to his own. But the first comment was that it's just "random," which is what I took issue with. But I also wonder when someone over-exaggerates luck.

Anyway, I also agree those games shouldn't count at all. I don't know whether non-ELO games count as anything, other than showing up in a list that you played them. If that's true, then no-ELO should work on its own.

It's weird how the site is somewhat contradictory, implying one place that it's all just for fun ("can't you have fun without statistics?"), but in other respects makes ELO important, also stating it's more fun to play someone near your level and then that whole prestige issue, etc. It's all in good fun, but it's also a shame anyone has to make it so important as to essentially cheat with the bots. I wonder if they're proud of themselves.