Join Bridge Winners
All comments by Max Schireson
You are ignoring the author of this comment. Click to temporarily show the comment.
That’s the worst.

In theory they are entitled to this information and often don’t get it. Then they get it and it was superfluous on the actual hand, and could have steered them wrong.
Sept. 27, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Theoretically I could make a case that I have fulfilled my obligations by saying it’s a new agreement and she has a history of forgetting but I feel like it’s hard for them to know how seriously to take that possibility… so I should just tell them that based on my knowledge of partner and what we’ve all seen it really seems like she forgot.

That feels like the only practical way to put them on equal footing to me in terms of knowing my partner, which I think is their right (even if it is infrequently actually given).

Technically they aren’t entitled to the knowledge of the hitch or the inference from it but they almost certainly saw the hitch and the jference felt so clear given what I knew of partner, so I think practically it’s just partner info I am giving them when I synthesize. But it’s messy/murky.
Sept. 26, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
To be clear I don’t think the standard of what most club opponents would do is useful, and I didn’t intend to ask that question.

What I thought made it blurry is that that inference was available to me based on a combination of:
1. My knowledge of my partner, both that a forget was possible and an “impossible” 3N was possible; in my view it is unambiguous that opponents are entitled to this info.
2. The bridge knowledge that she should have never have reason to think over 4S; my opponents are not entitled to this inference. Imo this raises the likelihood of a forget - maybe 20% a priori - to something like 80-90% (because there is something like 2-5% that she would just hesitate for no reason, so most likely the hesitation is the misbid case).

The question is whether I can just leave it at “possible she forgot” which feels like a generic throwaway when he is persistently asking about other possible agreements, or do I need to go further (as I did at the table) which gives him (a simple) bridge inference to which he is not technically entitled, but may have made (perhaps subconsciously) anyway.

If I were playing against an opponent that I knew would easily make that inference then I think I must explain it as I did, because he is entitled to my knowledge about partner and already has the other half of the puzzle. I shouldn’t force him to say “what might account for the think” that I know he is trying to understand. But if the opponent doesn’t catch the impossibility of the think, do I then need to go the extra mile in emphasizing the likelihood of a forget? That is murky for me.
Sept. 26, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
A sum of independent events approaches a normal distribution over the long term… I don’t think the individual swings are very closely approximated by a normal distribution because of the idiosyncrasies of game bonuses etc which make certain swings more common than others, but the sum of many of them will be pretty close and my instinct is that 200 boards is comfortably enough for this to be a good approximation.

I don’t think this methodology is perfect but I think it gets us enough in the ballpark to understand that differences of eg 0.1 or 0.2 are probably in the noise and eg 1.0 or 1.5 are probably real.
Sept. 26, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Danny,

Appx 5.5-6 IMP typical single SD in even matches (I have seen this in various places, not sure the original source)

For 200 boards total SD sqrt(200) * single board SD gives about 80 imps, which is about 0.4 per board

This would also say that a theoretically even match would be +- about 40 imps over 60 boards, which feels ballpark right to my experience.
Sept. 26, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve,

Yes, this partner is a bit forget-prone (similar forgets in the past).

Also this partner is “creative” enough to possibly bid 3N natural here (which should be impossible).

With other partners I might know it wasn’t a forget because they would never bid an “impossible” natural 3N here.
Sept. 26, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Marty,

If a top pair has an expectation of +1 per board, sometimes they will land at +1.4 and sometimes they will land at +0.6. They might then be at the top or near the top. The Butlers will usually (but not always) confirm what we expected.

However when an average pair has a great event and winds up at +0.8 right up in with the top pairs, it can easily be a fluke. Since there are quite a few average pairs, we should expect that many of the unfamiliar names near the top (or familiar names that are not expected to be near the top!) are there through luck.

If you have a reasonably accurate prior view of where pairs belong, you will find that view is often confirmed… but that doesn’t mean that surprising results mean you should change your views; it is exactly the surprising results that are most likely (but not certain) to be flukes.

Interpretation of noisy data is hard.
Sept. 26, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
A few thoughts on Butlers:

1. Over 200 boards, I would expect the standard deviation to be around 0.4 imps per board. Butlers of eg +0.7 probably mean better than average and -0.7 probably mean worse than average but even that isn’t certain. The difference between eg +0.8 and +0.5 is mostly meaningless, but the difference between eg +1.1 and -0.8 is very very likely to be real (but still could be a pair of flukes where one average-ish pair had a *great* tournament and another average-ish pair had a poor tournament.
2. Opposition matters a lot.
- in a Swiss it’s *much easier to get a good Butler on a poor team because your opponents are less strong
- in a RR format like the BB, strong teams may play their weaker pairs against weak opponents and make sure their strongest pairs are playing in the matches against other teams in contention - not only are imps in close matches worth more VPs, but swings against other teams in contention effectively count double.

So while I don’t think they are totally meaningless, they definitely need to be viewed with some significant caveats.
Sept. 26, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
I would also like to add that restricting the strongest players may also drive away other players who want to compete against the best.

Our youth NABCs are great learning events, but not serious events that the strongest eligible players are excited to win. Each year that I see strong players not enter it makes me sad. There the issue is both masterpoint limits and event structure so it’s not a perfect analogy, but I do worry that limiting eligibility based on skill/experience drives away more than just the ineligible players.
Sept. 21, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
I have mixed feelings, but (like my feelings about any event I will never be eligible for) not strong ones.

I think junior bridge is great. The participants seem to have a blast and many of them grow into being top level competitors; some are while they are already eligible for juniors. Players who have won or are played in the Bermuda Bowl play U26, and last year we had a Bermuda Bowl veteran play U21 (!).

I think this is fine. I just wonder, given that U26 already has world class competitors, whether it makes sense to have a U31. Funding and mentoring for junior bridge is not unlimited, and I worry that this would come at the expense of other categories.

Sure, it would be great to see some of our young pros in the U31, but we might also see them in the Bermuda Bowl. We might also see fewer U26 or U21 teams sponsored by their NBOs to make room for the U31 teams, and I am worried that that tradeoff won’t be a net plus.

As for Peg’s idea of limiting it to newer players, I do think there is something to be said for serious events for newer players. This is especially true for new players who are not eligible for any age or gender restricted events. That said I think it’s hard to implement that limitation well today, and I wonder if a new-ish player event might be more useful than Peg’s contemplated new-ish U31 event. (Disclaimer: I might be eligible for the first but wouldn’t be for the second, so I may be biased).

I feel like the grinch for not jumping for joy about more junior events, but I do have concerns.

I am curious how three groups feel about this:
1. Players currently planning to compete in U26 who expect they would instead play U31 despite eligibility for U26, because they want to compete at the highest/most prestigious level of junior competition (this assumes U31 takes off and becomes the strongest and most prestigious junior event).
2. Players currently planning to compete in U26 who expect they would continue to do so (unless perhaps to their surprise they made a U31 team)
3. Players who are no longer able to compete in U26 who would be eligible to compete in U31

This doesn’t capture the resource issue I worry about, but it does get to potential dilution of a long established U26 event, which I also worried about.
Sept. 21, 2019
Max Schireson edited this comment Sept. 22, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
John,

Do you have specific continuations agreed?

Why does partner do with a good spade raise?
Sept. 9, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
I debated whether *all* applicable hands would open 1N, but after some thought left it at “most”.

The hands that I imagined might at least consider bid this way with would be good 17s with a 5 card spade suit that I was planning to rebid 2N, but when partner makes a 2/1 I don’t want to have to bid to 4N to show my values.

Besides the intellectual inconsistency of first deciding the hand was worth 18 then backing down I dislike this auction - it is very awkward if partner has a good spade raise - so on further thought I am pretty confident I would never actually bid this way (even apart from you stating that you would not know what it meant)… but if a partner who had read those notes did, I would think the agreement would still apply, and that they would have good reason for me to understand what they are showing.

All that said, I am not sure I would like the agreement if it was ever going to come up, so I now regret sharing what I did.
Sept. 8, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
For what it’s worth, I think “outrageously out of line” is closer to a correct description of Michael’s *comment* than “perhaps just a little inappropriate.”

Knowing Michael a bit, I am confident in his positive intentions. I suspect that Kevin likely was also confident that Michael genuinely expected the comment to be helpful and welcome, and this colored his comments.

Michael was well meaning, and I think his behavior - by which I mean both the comment and the motivation behind it - was not at all outrageous. But Michael is someone who wants to and does learn, and I think it is good for him to see very clearly that
- not everyone values learning at bridge in the way he does
- the norms for interactions at the bridge table definitely do not include making comments such as the one he made
Sept. 7, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Michael R: I like your rule. It is more accurate, but also harder to apply.

Unlike the golden rule it is likely to eventually (perhaps now) help Michael with this situation.
Sept. 6, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
I think that the golden rule won’t help Michael, because he would likely be happy to hear unsolicited advice that is correct/useful.

Michael, I agree with the general sentiment that unsolicited advice is usually a bad idea. If you give it the burden is on you to know that it is an opponent and situation where it will be appreciated.

It is very hard to be confident that it will be appreciated; for example here are some reasons someone might not want this type of feedback (not all of which necessarily apply to this hand)
- they may want to move on to the next hand
- it may be embarrassing in front of their partner (imagine their partner is paying them for a lesson- even more embarrassing then!)
- it may be an affront to their ego that someone who wasn’t even born when they started playing bridge knows more than them
- they may already know their error and be upset about it
- they may believe their play was correct and not want to discuss it - perhaps even to avoid embarrassing you

It is great that you are so motivated to learn. Not everyone shares that motivation. If you stop to think about it, it might not be so surprising that someone who has 20x your bridge experience but less skill doesn’t share your passion for learning and improvement.

In summary: don’t give unsolicited advice unless you are sure it will be appreciated!!!
Sept. 6, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
My agreement is about a K extra. Yeah most of those hands have some other in this auction, but definitely not a balanced minimum.
Sept. 4, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Bill,

I think you are correctly describing the mechanics of standard 2/1, but not the value and the purpose.

The space through the 3 level is primarily used tounderstand how the shapes of the hand mesh, which is a form of exploring for slam. Players who are playing 2/1 explore for slam on every 2/1 auction, but they start doing that by seeing if the shapes make them feel there is the potential for 12 tricks, and only when that is the case do they check for controls unless they have significant extras.

This data is really important for evaluating slam.

Slam is not just about controls. You need 12 tricks. Think about 5=3=3=2 opposite 3=3=3=4 with eg 27 HCP consisting of AKQ of S, AK of H AK of D A of C. All the trump controls, first round control of every shot, second round control of almost all suits, and a good game (though still with some risk) that has no play at the 5 level.

I think there is merit to learning about trump for earlier, which your system does. I think it is more important to explore shape than controls next.
Aug. 31, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Yes!!!!!

I find that learning to see and think through whole hands is
1. Essential to playing well
2. Incredibly hard to learn as an adult

I don't really know how to learn it as an adult - still work in process for me - or how to teach it.

For what its worth, thinking hard about hands helped a lot. I found that eventually I could hold whole hands in my head and accurately play through variations. At first it took literally hours of thinking about a hand. Now it comes in 15 or 20 minutes.

Unfortunately for me it still doesn't come quickly enough to see things clearly at the table… and thinking I do and not quite having it, or stretching to see it and missing something else leads to some epic fails.

I may write more about this. I think it comes naturally to young players and only with very great difficulty to adults. I am excited that you are trying to teach it and look forward to learning more about what works/doesn't.

My only worry is that few students will be willing to put in the effort, which for me was monumental and I am still not all the way there. But if even a few do, you will be doing a very very good thing.

For what its worth in the beginning 5 is a lot. I would be happy if a beginner - or even a decent intermediate - could really grasp even one hand with some study.
Aug. 30, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Fwiw the guideline many follow is that with a good trump fit and shortness slam can be possible with two well fitting minimums. This can routinely be be 24 HCP or fewer. Tools like splinter bids, J2NT, and others are designed to help find these slams. I have certainly looked for slam when it is clear that we have less than half the HCP (but those cases are rare).
Aug. 28, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Bill,

The way I think about slams is that you need two things:

1. the ability to take 12 tricks
2. the ability to stop them from taking two tricks first

I agree that your system does a good job of identifying one risk to item 2 at a low level.

However, multiple posters have suggested that the key to finding low HCP slams is the shape that enables 1. Your system seems to take away from this.

I suspect that if you focus on 1, and find that the hands fit well and have the potential to take a lot of tricks, it will be relatively rare that you are in danger at the 5 level.

It’s great that your system has helped you find a lot of low HCP slams. Is it possible that with continued focus on that area of bidding, but prioritizing the fit and trick taking potential of the hands, you could do even better?
Aug. 27, 2019
.

Bottom Home Top