Gib, you got me again

[thorvald]

[hv=pc=n&s=sa53h54da93ckqjt3&d=w&v=0&b=8&a=p1hp2c3d3s4d]133|200[/hv]

What should i bid now?

 

This is a challenge, and you will have the best hand in regards to HCP, but you still need to find a bid.

 

We know partner is short (single or void in ♦)

 

Question is if partner could be 5-6 in the majors, what about 4-6 would GIB then bid 3♥?

 

When you have decided you can take a look at the hand

 

 

 

[hv=pc=n&n=sk872ha9762dca752]133|100[/hv]

Why suppress a 4-card support?

 

And ♥ 5-1 if you bid 4♥

 

 

[sfi]

Unlike many of your other hands, the robot's actions here look reasonable. You can still have four spades, so it takes an easy opportunity to show them. It can always show clubs later.

 

I can't see any reason to do anything but pass over 4D. You're in a game-forcing auction with no clear action. I agree you might have a problem if RHO had passed, but here you would have just bid 3NT.

[thorvald]

Unlike many of your other hands, the robot's actions here look reasonable. You can still have four spades, so it takes an easy opportunity to show them. It can always show clubs later.

 

I can't see any reason to do anything but pass over 4D. You're in a game-forcing auction with no clear action. I agree you might have a problem if RHO had passed, but here you would have just bid 3NT.

 

Yes, you have a point, but thinking about it, Double might be the right bid - I will check the continuation and return.

 

I am not a 2/1 player normally, so I also make mistakes. I am so used to a system, where showing fit is the highest priority

[thorvald]

Observation no 1: If undisterbed bidding, the BOT will just bid 3♣ and not show the ♠-suit

 

Observation no 2: Over pass, the bot will bid 5♣, and describe the hand as 4=5=0=4.

 

Observation no 3: Over double the bot will not show the ♣-support.

 

Observation no 4: If the opponents just bid 2♦ the bot will not bid 2♠, but 3♣

[pilowsky]

To make it a fair test, it may be an idea to play challenges against robot where it isn't best hand.

After all, the object of the exercise isn't whether you think there's a better way of doing something, it's about who gets the best result over a reasonable number of hands.

Using "best hand" deals is a bad idea since one player knows something the others don't.

 

We can all find examples of bidding and play that leads to bad results - and get some enjoyment by posting them, but this doesn't indicate a structural inadequacy with GIB.

What I mean by that is that the 'purpose' of GIB is not to play 'perfectly' - by definition impossible.

 

Others here complain that robot Bridge is a pointless exercise.

I disagree. Practising a simple 2/1 GIB system - especially using not best hand deals - has helped me go from miserable to almost useless - a considerable improvement.

[sfi]

Observation no 1: If undisterbed bidding, the BOT will just bid 3♣ and not show the ♠-suit

 

Observation no 2: Over pass, the bot will bid 5♣, and describe the hand as 4=5=0=4.

 

Observation no 3: Over double the bot will not show the ♣-support.

 

Observation no 4: If the opponents just bid 2♦ the bot will not bid 2♠, but 3♣

That feels uncommonly well thought out. It shows the clubs if you still have room to show spades, but shows support if it has to. Over double I'm guessing it assumes you are penalising and it passes?

 

The problem with it bidding 3C in an auction like 1H - 2C; 3C is that you might bid 3S to probe for 3NT. But in my experience that's too advanced for the bots.