BBO Discussion Forums: Gib is too superstitious Total Point - BBO Discussion Forums

Jump to content

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

Gib is too superstitious Total Point

#1 User is offline   lycier 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,612
  • Joined: 2009-September-28
  • Gender:Male
  • Location:China

Posted 2016-August-27, 18:18

Hi Stephen Tu

First, I am gonna give you two special question hands, and ask you to answer two questions, ok?
===============================================================
Hand-1


Hand-2


===============================================================
Question-1
For human, how to respond normally in the two hands above?

Question-2
For Gib, would you know Gib E how to respond in the two hands above?


Stephen Tu, I would better tell you my answers.
Question-1
Hand-1 : I am gonna respond forcing 1N.
Hand-2 : Of course, pass.

Question-2
I really don't know the answers, so let Gib tell its true story.

Some day I encountered a special Gib hand at BBO.


A little bizarre responding ! This Gib hand caught my attention. So I decided to make some simulating hands to show what Gibs are doing.

Hand-3

This hand shows Gib E pass 4hcp and 4TPs when with 3-card support.

Hand-4

This hand shows Gib E responding forcing 1N with 5hcp and 5TPs when with 3-card support.

Hand-5

This hand shows Gib E pass with 5hcp and 5TPs when without 3-card support.

Hand-6


This hand shows Gib E pass with 5hcp and 6TPs when without 3-card support.

Hand-7

This hand shows Gib E respond forcing 1N with 6hcp and 5TPs when without 3-card support

Hand-8

This hand shows Gib E responding 1N only with 3hcp and 5TPs when with 3-card support.

Hand-9

This hand shows Gib E responding forcing 1N only with 2hcp and 5TPs when with 3-card support.

Hand-10

This hand shows Gib E pass with 2hcp and 4TPs even when with 3-card support.

Hand-11


This hand shows Gib E responding forcing 1N only with 1hcp and 5TPs when with 3-card support.

Hand-12

This hand shows Gib E shows Gib E responding forcing 1N even with zero hcp and 5TPs only when with 3-card support.

Hand-13

This hand shows Gib E responding forcing 1N with 6hcp and 9TPs when without support.

On the Gib CC, the exact definition of forcing 1N is " Forcing one notrumph --- 3-,6+hcp,12-TPs".
These hands above completely showed its definition isn't correct.
Its true definition :
- When with 3-card support, it promises 5-12TPs with 0+hcp.
- When without 3-card support, it promises 6+hcp with 5+TPs.

Stephen Tu :
1- Its definition looks like a multipurpose definition, is it allowed in ACBL tournaments?
2- These hands above have shows TP replacing HCP partly at least. How Gib wildly superstitious Total Point !
3- Contrary to my answers in my two questions:
Question-1 : Gib will pass
Question-2 : Gib will respond forcing 1N.
This is just a rediculous basis application of Total Point. What else? Would you have pretty better excuses to tell us?


Before you type " my points are totally wrong", I, as a layman on Gib programming, am gonna remind you that these hands above should show its bidding rules, and it should fairly say there are no Debug, there are no Bug, all the truth are there, where are your beautiful reasons?Posted Image

Gamblers and adventurers including professional athletes are generally superstitious. Be Gibs would not superstitious Total Point?Posted Image
0

#2 User is offline   Stephen Tu 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 4,100
  • Joined: 2003-May-14

Posted 2016-August-27, 19:36

View Postlycier, on 2016-August-27, 18:18, said:

===============================================================
Hand-1


Hand-2


===============================================================
Question-1
For human, how to respond normally in the two hands above?

To me, hand 1 is on the edge between pass and 1nt. I personally pass, as I don't think the potential for making game/improving contract to 2H is particularly high compared to the chance of going minus if partner bids too much. Without any real fit I think it's fine to pass with 5 balanced.

Hand 2, personally I would bid 1nt intending to usually try to play 3d or maybe spades if partner shows long spades or does FG jump shift. Since I think I can take a lot of tricks in diamonds and bidding rather than passing makes it harder for the opponents to find their heart and/or club fit, and even if I go down in 3d it's likely fine vs. the opps making something. Though passing is not crazy I'd rate it below bidding. I would also suggest that trying to form your arguments about TP based on hands containing 9+ cd suits is super weird. 9 cd suits are extremely rare animals, one doesn't program software to cater to deal with extremely uncommon hand patterns. You try to cater to the more common hand types. Now one could perhaps introduce rules to handle these super rare distributions in a reasonable manner if it one crops up in a bug report, but it would be pretty low on the totem pole to fix.

Quote

Question-2
For Gib, would you know Gib E how to respond in the two hands above?

Not without testing, but I'm sure you'll tell us.


Quote

Some day I encountered a special Gib hand at BBO.


A little bizarre responding ! This Gib hand caught my attention. So I decided to make some simulating hands to show what Gibs are doing.

Responding 1nt isn't bizarre. It's a common 2/1 treatment to use a forcing NT raise with weak spade raises in the ~4-6 pt range. The bizarre part is the jump to 4S which should show the 3 cd LR hand type not the weak raise type. Preference to 3s is normal on the second round, this is simply a bug that ought to be fixed.


Quote

On the Gib CC, the exact definition of forcing 1N is " Forcing one notrumph --- 3-,6+hcp,12-TPs".
These hands above completely showed its definition isn't correct.
Its true definition :
- When with 3-card support, it promises 6-12TPs with 0+hcp.
- When without 3-card support, it promises 6+hcp with 5+TPs.

Yeah, it's pretty standard for humans to play forcing NT as:
[1] 6-12 hcp, 2- M
OR
[2]
4-6 tp, 3M.
With some debate on exactly what range to use for [2], it basically goes up to the floor of the direct raise, and there are advocates for 8+ (constructive, not particularly popular in my estimation), 7+ (semi-constructive), 6+ (non-constructive). GIB does 7+ for the direct raise which happens to be what I prefer. But others like to raise a little weaker.

GIB seems be doing something more like 5-6 tp, 3M for the second set of hands, the weak raise type.

So it's not really a complete description for the forcingNT. Just add one to the list of many not totally accurate GIB descriptions. Note the human language description is just for the ease of use of humans, doesn't necessarily completely show what the GIB rules are doing underneath. It's nicer if they match, but it's sometimes hard to get it to print something accurate for the human description. Also the GIB bidding language doesn't really have a way to describe bids that have disjoint sets, e.g. 2nt rebids that are 12-14 OR 18-19.

Quote

1- Its definition looks like a multipurpose definition, is it allowed in tournaments?


Yes, it's allowed. Yes, human experts do this routinely.

Quote

2- How Gib wildly superstitious Total Point !

I fail to see how this qualifies as "superstition". Computers aren't superstitious. They follow rules layed out by humans. I fail to see how these examples at all illustrate anything to do with your apparent belief that TP are the root of all evil. For one thing, you don't really say anything about what you think GIB ought to have bid instead on these hands and what your criteria are for bidding or passing or bidding 1nt.

On some of the example hands where GIB tries a 1nt bid, and then ends up in 4s, a lot of them rate to make on a normal honor from CAKxx lead. Yes, they can go down on double dummy best leads. But that's bridge, sometimes you get overboard double dummy. But you'll show profit if you do well on those hands on average as opps will often not find the best lead. I suppose you wanted GIB to pass instead?

The idea behind bidding 1nt with the weak raise hands is:
- a direct raise would be too encouraging, partner will blast game or make game tries and tend to go down when you have 5 pts instead of the expected 8 or so, K stronger on average. Putting some weak raises in 1nt tightens up your direct raise range which makes bidding more accurate over that bid.
- bidding rather than passing makes it harder for the opponents to come in because you are still wide ranging, and it's also hard for them to balance after your eventual 2M because you might only be on 7 card fit.
- game is still possible if partner has a huge hand, so may want to keep the bidding alive on these 5 TP hands.
- even if you go down in your eventual contract, it's not super likely that you would have bought the contract for 1M making exactly 1, because the opponents will usually balance, and if you are down it's OK if the opps are making their contract.

Now sometimes you will go down and passing instead of 1nt would have allowed you to go plus. But that's bridge, it's a game of percentages, and people who include weak raises in 1nt estimate that they are net plus on these hands vs. passing. Doesn't mean that any particular example you might pull out might be one of the losing cases.

None of this really has anything to do with use of TP or not, in my view. It's a matter of how you want to utilize the F1nt. TP is just for hand evaluation purposes. I for one definitely support use of TP, or at least some other method of taking distribution into account, as I DO think 1nt is reasonable on xxx x xxxxxxxxx -, but definitely want to pass and not bid 1nt with xxx xxx xxxx xxx.
0

#3 User is offline   lycier 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,612
  • Joined: 2009-September-28
  • Gender:Male
  • Location:China

Posted 2016-August-27, 22:50

View PostStephen Tu, on 2016-August-27, 19:36, said:


The idea behind bidding 1nt with the weak raise hands is:
- a direct raise would be too encouraging, partner will blast game or make game tries and tend to go down when you have 5 pts instead of the expected 8 or so, K stronger on average. Putting some weak raises in 1nt tightens up your direct raise range which makes bidding more accurate over that bid.
- bidding rather than passing makes it harder for the opponents to come in because you are still wide ranging, and it's also hard for them to balance after your eventual 2M because you might only be on 7 card fit.
- game is still possible if partner has a huge hand, so may want to keep the bidding alive on these 5 TP hands.
- even if you go down in your eventual contract, it's not super likely that you would have bought the contract for 1M making exactly 1, because the opponents will usually balance, and if you are down it's OK if the opps are making their contract.

Now sometimes you will go down and passing instead of 1nt would have allowed you to go plus. But that's bridge, it's a game of percentages, and people who include weak raises in 1nt estimate that they are net plus on these hands vs. passing. Doesn't mean that any particular example you might pull out might be one of the losing cases.

None of this really has anything to do with use of TP or not, in my view. It's a matter of how you want to utilize the F1nt. TP is just for hand evaluation purposes. I for one definitely support use of TP, or at least some other method of taking distribution into account, as I DO think 1nt is reasonable on xxx x xxxxxxxxx -, but definitely want to pass and not bid 1nt with xxx xxx xxxx xxx.


Helgemo had ever said that guessing is a worst part of this game.
What you said is just a kind of your guessing.
For example :
My question Hand-2

If you really respond forcing 1N, maybe you are looking for your trouble.
If you play against me, when I gain enough scores, I am gonna keep silent, but when I gain little scores, of course, I will call director to look for your trouble since you can't follow your CC " 1N=3-,6+hcp,12-TPs", there is over 6hcp differences on your CC , maybe with secret conventions here.

You are not God, that's to say that too many luck may not be with you in many situations. For example, in a bad day, some worse matter may occur.


Result : 4Wx-3 800
Note it is me to make this penalty double at first, then leave the table and let Gib play.

Now you see 800 scores enough to le me keep silent.
If don't respond forcing 1N, I have good reason to believe 1 will be a final contract, its result is 1W= 80.



0

#4 User is offline   Stephen Tu 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 4,100
  • Joined: 2003-May-14

Posted 2016-August-27, 23:37

View Postlycier, on 2016-August-27, 22:50, said:

Helgemo had ever said that guessing is a worst part of this game.
What you said is just a kind of your guessing.
For example :
My question Hand-2

If you really respond forcing 1N, maybe you are looking for your trouble.


Why don't you poll this hand in some other non-GIB forum, see whether other people think you should bid or not?

Quote

If you play against me, when I gain enough scores, I am gonna keep silent, but when I gain little scores, of course, I will call director to look for your trouble since you can't follow your CC " 1N=3-,6+hcp,12-TPs", there is over 6hcp differences on your CC , maybe with secret conventions here.

If you are playing against me in a human tournament, then in ACBL we simply announce forcing NT. If you ask what it is, I will say "normally 6-12 hcp, 2- spades, or 4-6, 3 spades" which is perfectly accurate. OK, so I pick up this 1 in 3000 hand or whatever it is where I happen to choose to bid on less HCP than partner and opps normally expect because of the extreme shape. So what? If you call the director, he's just going to shrug. There's not going to be any adjustment. If you consider it a psych, so what, psychs are legal.

It's not any secret convention. It's a judgment call to bid with no HCP because holding extreme shape. That's perfectly allowable via bridge laws. You are entitled to know a partnership's *agreements*, not the contents of a hand. And it's perfectly legal to deviate from partnership agreements, when partner is going to be in the dark about it. I will normally bid 2d or 3d on most sequences, partner will expect 6-7 diamonds usually and about 4-7/8 pts or so, because such hands are way way more common. Instead of this I turn up with 9 diamonds and 0 hcp once every 10 years. So what? Perfectly legal.

Quote

You are not God, that's to say that too many luck may not be with you in many situations. For example, in a bad day, some worse matter may occur.

Of course I don't expect bidding to always turn out better than passing. Posting a single bad result proves nothing. I am expecting to be ahead net by bidding instead of passing. You may disagree, but I think you are wrong. You can post a poll in another part of this forum if you want to solicit other opinions.

Quote


Result : 4Wx-3 800
Note it is me to make this penalty double at first, then leave the table and let Gib play.

Now you see 800 scores enough to le me keep silent.
If don't respond forcing 1N, I have good reason to believe 1 will be a final contract, its result is 1W= 80.


Uh ... with 9 freaking diamonds, why the hell am I not trying 5d making instead of 4S if partner bids 3nt showing balancedish hand? Bidding 4s is pretty horrible. Leaving in 4s-x is insane. If partner shows balanced I am definitely playing in diamonds. Only if partner rebids large number of spades and might have 7+ of them with void diamond am I going to consider staying in spades since I'm off so many honors.
Note opps are cold for 4H and 5C. Lots of good human North may bid an off-shape 1nt overcall to not miss this game, sadly GIB will not really consider it.

Have you not heard of the don't put down 8 cd suits as dummy rule?

Now if GIB is choosing these things, then it's just bug in GIB on the followup auction. Doesn't mean the 1nt is a clear mistake. It's also arguable that 19 balanced as west is supposed to bid 2nt over 1nt not 3nt, that's the normal bid in 2/1.
0

#5 User is offline   lycier 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,612
  • Joined: 2009-September-28
  • Gender:Male
  • Location:China

Posted 2016-August-28, 07:36

Posted Image
0

#6 User is offline   Stephen Tu 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 4,100
  • Joined: 2003-May-14

Posted 2016-August-28, 08:15

Wow, super mature lycier.

See this is why debating with you is practically pointless. You try to come up with an example of why bidding 1nt is bad. I make a completely valid point that with a 9 cd suit that East ought to bid it and make a game instead of going down 800. Instead of admitting that you are wrong, or trying to refute my point, you call it crying. This is not constructive debate whatsoever. You can't convince people of anything when you refuse to specifically address an opponent's individual arguments and instead just dismiss it all generally as crying. Convincing arguments need specific breakdowns, refutation of opponent's individual statements. "Your statement is wrong because of ____, ____, and ____". Just repeating "you are wrong" without any explanation over and over is just childish. In a debate class you'd get flunked.

Maybe I shouldn't bother and just put you on ignore.
0

#7 User is offline   iandayre 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,114
  • Joined: 2013-December-23
  • Gender:Male

Posted 2016-August-28, 12:54

View PostStephen Tu, on 2016-August-27, 23:37, said:

Why don't you poll this hand in some other non-GIB forum, see whether other people think you should bid or not?

[/size]
If you are playing against me in a human tournament, then in ACBL we simply announce forcing NT. If you ask what it is, I will say "normally 6-12 hcp, 2- spades, or 4-6, 3 spades" which is perfectly accurate. OK, so I pick up this 1 in 3000 hand or whatever it is where I happen to choose to bid on less HCP than partner and opps normally expect because of the extreme shape. So what? If you call the director, he's just going to shrug. There's not going to be any adjustment. If you consider it a psych, so what, psychs are legal.

It's not any secret convention. It's a judgment call to bid with no HCP because holding extreme shape. That's perfectly allowable via bridge laws. You are entitled to know a partnership's *agreements*, not the contents of a hand. And it's perfectly legal to deviate from partnership agreements, when partner is going to be in the dark about it. I will normally bid 2d or 3d on most sequences, partner will expect 6-7 diamonds usually and about 4-7/8 pts or so, because such hands are way way more common. Instead of this I turn up with 9 diamonds and 0 hcp once every 10 years. So what? Perfectly legal.




Of course I don't expect bidding to always turn out better than passing. Posting a single bad result proves nothing. I am expecting to be ahead net by bidding instead of passing. You may disagree, but I think you are wrong. You can post a poll in another part of this forum if you want to solicit other opinions.



Uh ... with 9 freaking diamonds, why the hell am I not trying 5d making instead of 4S if partner bids 3nt showing balancedish hand? Bidding 4s is pretty horrible. Leaving in 4s-x is insane. If partner shows balanced I am definitely playing in diamonds. Only if partner rebids large number of spades and might have 7+ of them with void diamond am I going to consider staying in spades since I'm off so many honors.
Note opps are cold for 4H and 5C. Lots of good human North may bid an off-shape 1nt overcall to not miss this game, sadly GIB will not really consider it.

Have you not heard of the don't put down 8 cd suits as dummy rule?

Now if GIB is choosing these things, then it's just bug in GIB on the followup auction. Doesn't mean the 1nt is a clear mistake. It's also arguable that 19 balanced as west is supposed to bid 2nt over 1nt not 3nt, that's the normal bid in 2/1.


Of course by passing you lose the opportunity to reach the cold 5D. To me, the problem is the 3NT rebid. Most 2/1 players use that bid to show AKQxxx of S with cards on the side. 2NT should show a good 17 to 19, with 19+ hands upgraded to 2NT openers.
0

#8 User is offline   lycier 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,612
  • Joined: 2009-September-28
  • Gender:Male
  • Location:China

Posted 2016-August-28, 16:55

Responding forcing 1N merely based on TP - Gib is wildly superstitious Total Point.Posted Image

Facts are facts, sophistry is meaningless,no amount of sophistry can cover up the truth.Posted Image
0

#9 User is offline   Stephen Tu 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 4,100
  • Joined: 2003-May-14

Posted 2016-August-28, 17:36

It is a fact that 5d makes. Do you dispute that?

Sophistry means a false argument. You have not demonstrated that 5d making is false.

Since English is not your native language, I'd advise you to stick to simpler words where you truly understand the meaning. And learn how to make an actual argument rather than basically ignoring opponent's factual statements and basically claiming "lies, all you say is lies". If you cannot present irrefutable facts of your own to support your argument, cannot dispute your opponent's facts, you have lost the argument. You have devolved your argument into "I don't care what you say, I'm right and you are wrong, and I refuse to explain why because I can't, it is just my feeling". Good luck convincing bridge players that way, who rely on facts and logical thinking.
0

#10 User is offline   hrothgar 

  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 15,495
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Natick, MA
  • Interests:Travel
    Cooking
    Brewing
    Hiking

Posted 2016-August-28, 17:40

View PostStephen Tu, on 2016-August-28, 17:36, said:

Since English is not your native language, I'd advise you to stick to simpler words where you truly understand the meaning.


Silly me, thinking "Sophistry" was from the Greek...
Alderaan delenda est
0

#11 User is offline   Stefan_O 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 469
  • Joined: 2016-April-01

Posted 2016-August-29, 14:51

Total Points -- not a problem, per se, of course.
Everyone -from beginners to world-champions- uses them in different ways: hcp+distributional points+adjustments

Only...... the "walrus" WAY that Gib counts them, of course, is ridiculous.
- Same extra points for shortness, regardless of whether you found a trump suit or not.
- Same extra points for shortness in pd's suit -- always -- even when you play in NT... scary.
0

#12 User is offline   lycier 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,612
  • Joined: 2009-September-28
  • Gender:Male
  • Location:China

Posted 2016-August-29, 16:43

In the dynamic bidding sequences, Gibs do not know when to add value, when to devaluate.
On the Gib TP evaluation world, there is only addition, no substraction.
So Gibs often only make many one-sided description, this is a kind of misrepresenting description, too many rediculous story will occur.

I think this is a biggest problem.
0

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

3 User(s) are reading this topic
0 members, 3 guests, 0 anonymous users