[cherdano]

In my view, there is some very basic human psychology that makes it difficult to evaluate the effects of opening or overcalling light without biases.

Say you reach 5 vulnerable 23 hcp 3NT games. They all should go down. Two of them make on misdefense.

I think human instinct is to say "Damn, I got lucky" on the two of them, and "Damn, I overbid again" on the other three. On these three you know you were in full control, and you yourself caused your side to lose IMPs. But it's hard to give yourself the same credit for the IMPs you won on the two boards where you gained two IMPs.
This is perhaps even more true when you know you aren't playing the strongest opposition the world has seen. But when you gain IMPs against opponents in the US team trial semifinals, it becomes a little bit easier to credit yourself for the IMPs (partly as you are more likely to understand why your opponent went wrong - you can construct the hand you could have held where the opponent's defense would have been necessary).

A good comparison are football coaches decisions. 4th down with 1.5 yards to go for the endzone, early in the game. Basically any mathematical model says it is clear to try for a touchdown. NFL coaches go for the field goal way too often. But if they go for the touchdown and fail, they know they cost their team 3 almost certain points. If they go for the field goal, nobody can claim they cost their team four points for certain.

Humans tend to overvalue certain gains or losses compared with uncertain gains or losses, even if the expected gain or loss in the latter case is the same.

[Joe_Old]

Sorry, Kenberg, wasn't meaning to focus on you unduly.

Actually, one major reason that experts are bidding differently now than 60 years ago is that tournament hands are different than they were 60 years ago. Those of us who were playing when computer dealt hands first started appearing were shocked by the "wild" distributions. There were numerous complaints. The reason was that humans don't (usually) shuffle effectively, and human dealt hands tend to be flatter.

Don't think that people didn't notice. Early in the 1970's, many better players were already devising ways to evaluate and bid distributional hands. Work's HCP counting was discarded first. Losing trick count and other systems were devised. Basically, the experts learned that bidding systems then in vogue worked much better in club games (human dealt) than tournaments, and moved toward big club systems that left room for weakish one level bids. It's no coincidence that Precision was such a success.

Therefore, one major reason for the proliferation of weak calls by experts is that they've found that they work, and one major reason that they work is that hands actually played in tournaments are more distributional. Now that clubs are using computer dealt hands (not to mention online clubs) players who want to compete most effectively have to move to methods that best address the hands.

[fourdad]

They do it because they are better than the rest of us...and they KNOW it!!

[benlessard]

What would be more pertinent is just check cases where one table open and the other didnt and there was imps win or lost.

I believe opening 1C light is losing imps, while 1M light is a winning style.

1D depend on the lenght promise

[kenberg]

Joe_Old, on 2017-September-06, 10:47, said:

Sorry, Kenberg, wasn't meaning to focus on you unduly.

I was in no way offended.

But since there were several comments, I expanded a bit. Adam asked why a non-expert does not emulate an expert. Afaik, I am the only respondent so far to say something along the lines of "Hey, trust me, I am a non-expert. Here are some of my reasons for the choices that I make". No doubt an expert can make better bidding choices than I do. But if the question is why I make the choices that I do, I could well be in a better position than he is to give my reasons.

No doubt others have other reasons.

I'll add another point. I think much bridge instruction gets the cart (the bidding) before the horse (the play). If a result is bad, or for that matter if a result is good, it is useful for a player to review the hand to see why. Unless he has a reasonable ability at playing the hand, he is apt to come to the wrong conclusion. An extreme example: In a recent online game I put a partner in 6C. The bidding was not sophisticated but there we were. To make, the clubs had to be no worse than 4-1, they were 3-2, and two of the side suits had to split no worse than 7-1 and 6-1 respectively. They split fairly evenly. Down 1. When this happens, the need is not for more sophisticated bidding techniques.

[dokoko]

kenberg, on 2017-September-06, 13:36, said:

I was in no way offended.

But since there were several comments, I expanded a bit. Adam asked why a non-expert does not emulate an expert. Afaik, I am the only respondent so far to say something along the lines of "Hey, trust me, I am a non-expert. Here are some of my reasons for the choices that I make". No doubt an expert can make better bidding choices than I do. But if the question is why I make the choices that I do, I could well be in a better position than he is to give my reasons.

No doubt others have other reasons.

I'll add another point. I think much bridge instruction gets the cart (the bidding) before the horse (the play). If a result is bad, or for that matter if a result is good, it is useful for a player to review the hand to see why. Unless he has a reasonable ability at playing the hand, he is apt to come to the wrong conclusion. An extreme example: In a recent online game I put a partner in 6C. The bidding was not sophisticated but there we were. To make, the clubs had to be no worse than 4-1, they were 3-2, and two of the side suits had to split no worse than 7-1 and 6-1 respectively. They split fairly evenly. Down 1. When this happens, the need is not for more sophisticated bidding techniques.

Right you are!

[dokoko]

fourdad, on 2017-September-06, 11:48, said:

They do it because they are better than the rest of us...and they KNOW it!!

Not quite!

They can do it because they are better (i.e. they are more apt to handle the situations that arise).

They do it because it is theoretically sound to do so (when you can handle it).

[dokoko]

cherdano, on 2017-September-06, 07:49, said:

In my view, there is some very basic human psychology that makes it difficult to evaluate the effects of opening or overcalling light without biases.

Say you reach 5 vulnerable 23 hcp 3NT games. They all should go down. Two of them make on misdefense.

I think human instinct is to say "Damn, I got lucky" on the two of them, and "Damn, I overbid again" on the other three. On these three you know you were in full control, and you yourself caused your side to lose IMPs. But it's hard to give yourself the same credit for the IMPs you won on the two boards where you gained two IMPs.
This is perhaps even more true when you know you aren't playing the strongest opposition the world has seen. But when you gain IMPs against opponents in the US team trial semifinals, it becomes a little bit easier to credit yourself for the IMPs (partly as you are more likely to understand why your opponent went wrong - you can construct the hand you could have held where the opponent's defense would have been necessary).

A good comparison are football coaches decisions. 4th down with 1.5 yards to go for the endzone, early in the game. Basically any mathematical model says it is clear to try for a touchdown. NFL coaches go for the field goal way too often. But if they go for the touchdown and fail, they know they cost their team 3 almost certain points. If they go for the field goal, nobody can claim they cost their team four points for certain.

Humans tend to overvalue certain gains or losses compared with uncertain gains or losses, even if the expected gain or loss in the latter case is the same.

You are surely right!

I think most experts will "go for the touchdown" - i.e. bidding thin games consistently - because they know it's right (and need no outcome to prove it).

That said I think opening light has other goals. Imho the aim is to be in the bidding more often, forcing opps to overcall rather than open and enabling partner to preempt by raising our light opener which he couldn't had we passed.

Light openings only lead to light contracts if used occasionally. Opening light by design otoh need not lead to thin games as partner expects it and can adjust the standards for game bidding (I guess most experts do).

Below expert level, however, the gain from consistently opening light will not outweigh the loss from not adjusting subsequent bidding correctly. And opening light occasionally is losing strategy at any standard.

[sfi]

One of the people who plays for Australia every so often tells the story of when his crowd of juniors started playing Regres back in the 80's. They had really good results with the forcing pass system, and one of the top players commented on how dramatically it improved their performance. The comment was "we can handle the forcing pass fine, but you're now having to open all those 8-12 counts and we no longer have freedom to win all the part scores!"

Even at IMPs, part-score battles make up a large component of the game. If you open light, can judge well in the auction, can read your opponents effectively and can play your cards well, you have a good chance of picking up lots of 6 IMP swings. This one of the biggest areas of difference between experts and the second tier in the field, IMO.

Yes, there is a cost in definition for game and slam auctions, but those are much less likely to be contested. You therefore generally have time to sort out the ranges and make up for the initial wide range.

[sfi]

dokoko, on 2017-September-10, 03:12, said:

Light openings only lead to light contracts if used occasionally. Opening light by design otoh need not lead to thin games as partner expects it and can adjust the standards for game bidding (I guess most experts do).

Good point - it needs to be partnership expectation rather than merely individual flights of fancy.

[PhantomSac]

Think about how much time you have spent going over your constructive bidding auctions (where the opps pass, or you open and the opps bid), compared to how much time and definition you have spent on defensive auctions. Almost everyone has spent way more time on the former. Even assuming equal amounts of time spent on both, there is much more definition in the constructive auctions since there are far more possible defensive auctions to go over. Even an auction like 1D 1S p 2H is so much wider range (for both the overcaller and the 2H bidder) than if you had not opened. Wider ranged auctions make it harder for them to get to the right spot. And it's not like this is a flaw in the 1S or 2H range, if the opponents just passed when you opened that would make opening light even better! This is just the nature of defensive auctions being less good than unopposed auctions.

This makes it very valuable to open first. Every good opp if given a free run is going to get to the right spot. Additionally it's much harder/riskier to get in later. You see time and time again people having a 9 or 10 card fit and it goes say, p 1S p, whereas if they had opened it might have gone 1H 1S 4H. Any time you have a big fit and pass throughout is a missed opportunity to pressure the opponents, find a save, or push them up etc. Basically opening is the time when you are at the least risk to bid, you don't have to balance and risk going for a number etc.

The vulnerability matters a lot, generally white/red you will have far more saves, or the ability to jack it up higher, whereas red/white you won't. I think a style where you open sound vul and aggro NV makes a lot of sense.

You do sacrifice some constructive bidding accuracy by opening light (even if your partnership is aware of it, again remember that the opponents WILL bid, it's not really a problem if they don't bid, but sometimes after opening 1H and the opps bid 3S you have to do something and you're too high, this happens even with sounder openings!). The biggest downside to opening light is really when the opps reach a normal contract and have far more info in the play than they would have if you just passed. The better the opps, the more of a problem this is. But it is definitely worth it IMO to try to make it harder for them to reach the best contract, and to win the partscore battles/make sure you get to your making fits and partials.

[Jinksy]

In response to the original question, 'why don't club players do it?' I think there are a couple of perfectly sensible possibilities I haven't seen anyone give:

Quote

Think about how much time you have spent going over your constructive bidding auctions (where the opps pass, or you open and the opps bid), compared to how much time and definition you have spent on defensive auctions.

For a lot of club pairs, this ratio is much less skewed towards constructive auctions, simply because they haven't spent that much time going over *any* auctions. And to the extent that they have, they'll often have made a pragmatic decision to reach for common off-the-shelf agreements (like 1m 1x / 1N is 12-14HCP) because it requires less time to agree, and less effort to remember deviations from what they're used to. So Justin's reasoning, while it might still apply, would apply a lot less.

Quote

The usual explanation I see "they are much better declarers" doesn't seem right -- they also face much better defense and I suspect the gap in defense is actually more than in declarer play.

Assuming that you're right that the defence gap is bigger, declarer averages substantially more relevant decisions to make over the course of a hand, since they have more control over the play, and (relatedly) have a higher proportion of the cards that will determine the outcome of the tricks. So club declarers could play comparatively better than club defenders and still do worse on average than expert declarers facing expert defence on the same hand.