Unknown features of FIBS - The Beaver

[amarganth]

Hi all

Do you know the beaver command? Do you know, that it works at FIBS? This article explains its usage.

Note:
This feature is only available in unlimited matches.

That's what the help of FIBS tells about:

==> help beaver

NAME
  beaver - Offering an instant redouble that is a beaver
SYNOPSIS
  beaver
DESCRIPTION
  An instant redouble of an offered double is called "beaver". You can
  use the command 'redouble' instead. 'beaver' can only used for the
  first instant redouble while 'redouble' can be used for any of several
  instant redoubles in a row.
SEE ALSO
  otter, raccoon, redouble

To let the redouble work, you must first set the redoubles to a value > 0. As I know, this variable is set to 0 with every login. So you must set it before you start to play an unfinished match.

==> set redoubles 4

Both players must set the value > 0. The number of redoubles during the game is the lower value of this two persons. After that, you can start an unlimited match.  Let's assume, the cube is in the middle with a value of 1.

If your opponent now doubles, you will get the following message from FIBS:

playerX doubles. Type 'accept', 'reject' or 'redouble'.

If you now do a redouble, the cube will show 4 and you own it. That's a beaver.
May be your client does not support that 'redouble'. So you must enter it as a command.

==> redouble

Enjoy!

Note:
A good player said to me: "If a beaver occurs, one of the players is an idiot. The player, who doubled, or the player, that beaverd."

amarganth

[adamosad]

Wow. I didn't know this command!!! Thx u Amarganth!!!    

I need to try it. Do u want to play an unlimited match with me???  

[amarganth]

sure!    

[wyzzz]

I beaver ya!!!  


Now what do I do with the tail???


thanks Amarganth  


cheers  


Best of all things to you

wyzzz

[NIHILIST]

My recollection is that even playing UNLIM, the FIBS default is for match play, meaning it's not necessary to turn the cube to gammon your opponent.

Beavers are exclusive to money play, so to use it in its proper context, more tweaking is probably necessary.

Beavers generally occur when at least 1 player has read the position completely wrong; however there are instances where it is both correct to double and correct to beaver.


NIHI

[diane]

Although I recognise the line was offered as a loke - just thought I would mention that I had a stuation arise in which I was beavered - in which neither player was an idiot (I think   ) A bear off situation - I offered a redouble since I had gotten ahead - and it was beavered. I won, but my oppponent told me it was a correct beaver, since the opponent was Paul Lamford - I can see no reason to doubt that!

[wyzzz]

Yes  

Oh the beaver!!!!!!  such a combination..fun pond animal of great fur and frivolity and serious backgammon move of slapped tail ... its' tale on the board's fluid surface....a salute to a wonderous tree  chewer...tis  the real reason we don't use wooden dice.. ...the beaver's tale of tails...
I perfer to think of it as both...  

cheers ... and have a terrific time...may ya all redouble your fun..

best of all to all  

wyzzz

[NIHILIST]

I'd like to see the position that generated the DOUBLE/BEAVER. It's rare that the situaton would occur in a bearoff without one of the players being drunk or an idiot. Since the 2 players were Diane and Lamford, DRUNK is not out of the question.

Paul told you his beaver was correct, did he offer an opinion on the correctness of your double ? You said you had gotten ahead which by itself means little without also knowing the checker distribution. My guess is you had a cumbersome stack or gaps in your bearoff that made misses likely.

At any rate, I'd like to see the position if you remember it.


NIHI

[Shades]

At any rate, I'd like to see the position if you remember it.


NIHI

...beaver style..??..  

[diane]

I'd like to see the position that generated the DOUBLE/BEAVER. It's rare that the situaton would occur in a bearoff without one of the players being drunk or an idiot. Since the 2 players were Diane and Lamford, DRUNK is not out of the question.

Paul told you his beaver was correct, did he offer an opinion on the correctness of your double ? You said you had gotten ahead which by itself means little without also knowing the checker distribution. My guess is you had a cumbersome stack or gaps in your bearoff that made misses likely.

At any rate, I'd like to see the position if you remember it.


NIHI

Basically - no I cant remember the exact position - and being early in the day I wasnt drunk    The count made him require a double - and since he was back in the table a 66 really - but it was within the take margin - so he took and beavered. I do not profess to understand all of that yet - but Paul does make a point of telling me if I did something wrong - since I asked him to - and he didnt say I was wrong.

As to checker distribution - mine was good for bearoff - his was less good - which is what made me consider, and then give the redouble.

[tansley]

In a gammonless situation (ie. the bear-off) beaver-point is 40% (shows the value of owning the cube, huh?).

Diane, I think nihi's right that double/beaver situations are rare, and wonder whether you might have been doubling with less than 60 % game winning chance, in which case the beaver was correct, even though you were favourite!

Apologies if not    

[diane]

Hmm, could well be - this is definately what I mean when I say I dont profess to get all of this stuff - for me - having a good chance of winning is enough - whether it was 50% or 55% or 60% I certainly could not say - but what I do know happened, is what I felt would.  There were no jokers and the game played out with me leading and winning.  I am pretty sure I dont have to play assuming my opponent will roll the 66 required to win the match - and indeed - my double (naive as I am ) was an attempt to end the match before the 66 got thrown.  It was also a lesson in taking note of who I am playing next time I am sure plenty would have dropped it. Anyways, it made the end of the match much more interesting  

It may also be worth mentioning - I am well aware I am still learning - and need a lot of instruction in this area - as a result - the stakes were fun and not costly. (In case anyone is considering taking advantage of me   ) This does not, however make me take the situation less seriously.  The event was one designed as a learning arena - and as a result - I am still confident I would have been told if I was wrong - just as I was told the beaver was correct - however - all other opinions are welcome - I take them all on board  

So - If someone could explain what you mean in this context by the statement

shows the value of owning the cube, huh?

I would appreciate it!!

[NIHILIST]

Doubling in a bearoff with less than 60% game winning chances does not necessarily mean a beaver is correct.

I have 1 checker on my 5 point and 1 on my 2 point, on roll, owning a 2 cube. You have 1 checker on your 1 point. My GWC is about 53%. The proper action is DOUBLE/TAKE. Beavering is dead wrong.

This is because the game ends with my roll, there is no redouble so no gain by beavering. I still win at a rate of 53% regardless of the value of the cube.

Cube ownership is something whose value can't be overstated. The main value is that you are in the game until the end, you can't be doubled out.

One of the earliest expensive lessons I learned about the value of the cube involved a time-honored proposition. I start games with 1-1, my opponent owns the cube. Play that for a while and the value of cube ownership becomes obvious.


NIHI

[tansley]

Quite right nihi. I was trying not to get too complex.

The beaver point is 40% for a fully live cube (the potential for an efficient redouble later), 50% for a dead cube (as in nihi's example), and anywhere between depending on the potential efficiency of the cube.

As for Diane's question on cube ownership, if you assume a fully live cube, then taking ownership of the cube when your game winning chances are 40% immediately increases your game winning chance to 50% (hence the correct beaver).

The reason is that although your cubeless gwc is 40%, the chance of you reaching, say, 80% gwc is 40/0.8 = 50% (ie. winning 80% of this gives the original 40% wins).
But you can cash a win at 80% gwc (hence the need for an efficient redouble).
Thus if you own the cube, you'll win 50% of games.

[dorbel]

In a chouette many years ago, UK expert Rick Janowski spotted a Kauder paradox, i.e. a position that is a correct double and also a correct beaver. These occur in positions where the player on roll is an underdog, but has a strong gammon threat. The point of his double that it activates his gammons under the Jacoby rule. Rick doubled, got beavered and duly lost 4 points, while the wimps in the team who wouldn't go along with him only lost one, which is fairly typical! This was the only unconstructed true Kauder paradox position that I have ever heard of.

[diane]

The beaver point is 40% for a fully live cube (the potential for an efficient redouble later), 50% for a dead cube (as in nihi's example), and anywhere between depending on the potential efficiency of the cube.

As for Diane's question on cube ownership, if you assume a fully live cube, then taking ownership of the cube when your game winning chances are 40% immediately increases your game winning chance to 50% (hence the correct beaver).

The reason is that although your cubeless gwc is 40%, the chance of you reaching, say, 80% gwc is 40/0.8 = 50% (ie. winning 80% of this gives the original 40% wins).
But you can cash a win at 80% gwc (hence the need for an efficient redouble).
Thus if you own the cube, you'll win 50% of games.

OK - where to start with this - the third paragraph I am happy to not understand for some time to come.
The second paragraph - how does the gwc automatically increase by taking the cube? - erm surely it takes some good rolls for that?? Sorry - have to ask some dumb questions - cos I just dont see this!!
The first paragraph - erm - what is a fully live cube (trying to resist the temptation to holler "IS THAT CUBE ALIVE?" at my next tournament   ) ?

Can you offer any explanations that involve match progression - rather than equations??

On a positive note - I did follow Dorbels post ( I think)    

[tansley]

It's actually a lot simpler than it looks. Taking ownership of the cube increases your game winning chance.

Suppose during the game, you reach a (cubeless) game winning chance of 80%, you actually win all games  played from that position (as opposed to 80%), by doubling your opponent out.

Conversely, your opponent cannot double you out when her (cubeless) gwc reaches 80%, and you win 20% of all games played fom that position. (In fact slightly more than that because you own the cube!)

BTW NIHI, your proposition is interesting, but you should try rolling it out before you take the side of the board with the cube again.

tansley

[dorbel]

Diane, the point is this. When you give a poor double, your equity, that is your average expectation from the game goes down. However, not every poor double merits a beaver. A beaver is only correct when the double is so poor that your equity goes from positive with the cube in the middle to negative with your opponent holding the cube. This can certainly happen, but not as tansley suggests when the doubler is 60% to win the game in a race. He wrongly assumes that you can use the live cube assumption for this, but you can't. You can only do this if every recube is going to be maximally efficient, which isn't true. Sometimes you will have a cash, sometimes you will have a very minimal redouble, the variations are endless. A more realistic estimate might be to assume a point somewhere about halfway between a live cube and a dead one. I am prepared to believe that the taker might have a correct beaver in a situation where he has about 45% cubeless winning chances, but not 40%.
For those troubled by talk of live cubes and dead cubes; a dead cube often occurs in match play where the taker will never have a correct redouble, a live cube is one where the taker may have a number of chances to turn the game around and recube. A fully live cube is a rarity. the best known example is one checker on the 6pt versus one checker on the 6pt. The side on roll will win the game 81.25% of the time cubeless, but the other side can take, because all the games where he can redouble will occur in the optimal position of being on roll with a 75% chance of winning. See? if not, talk to me after class!

[dorbel]

By the way tansley, the whole point of the 1-1 versus cube ownership prop, is that the hustler starts what is essentially a level game with a pigeon. between equal players, it is probably close to a wash.

[Zorba]

Actually, if the cube is fully live you cash at 80% and opponent's takepoint is 20%.

The 18.25% TP from the bearoff example comes about because the cube is NOT fully live there. Opponent can redouble once but then the cube is dead since the game is over.