bluecalm

Thank you for pointing out error in my analysis. I run a double dummy sim starting from trick 3 (after Qs) and here are the results: Cards, winning, best: Kh 7 7 7h 480 870 6h 480 870 2h 480 870 Qd 350 642 5d 383 605 Ac 514 804 Kc 514 804 Jc 415 715 9c 415 715 7c 415 715 It agrees that Ac at trick 3 is better :)

I am tempted to try another spade hoping they duck and then play club to the T. If they take it I have 4clubs + 2 spades and only need 3 tricks in red suits which I will get as I have two stoppers in hearts. If they take a spade and switch to heart then I take in hand and play T of clubs. I have 3spades, 2hearts so they can't really afford to take it and if they duck I am trying D to the Q. It seems to me that this line works very often. I run a double dummy simulation with assumptions that W has 12+hcp, 4spades, 3-4 hearts and As. Here are the results at trick two: Cards, winning, best: Qs 603 997 Kh 7 3 7h 408 617 6h 408 617 2h 407 614 Qd 342 619 5d 333 446 Ac 531 803 Kc 528 802 Jc 16 404 9c 16 404 7c 19 407 Even double dummy playing one top club isn't the best and I think my line is the reason.

Why would he ? It's useful for him to wait and see if we bid 5♣ first. Anyway, he is not cuebidding Kh/x twice so void is the only possibility left now.

As partner bid void in hearts and lack of Ac they probably want to lead a club and get a diamond ruff. If partner has something along the lines of: AJxxxx - Qxxx ??? Does it make a difference to his bidding if he has KQx or xxx of clubs ? I don't think so so I can't really run to 6N. I would like to say "Partner, run if you have Kc but I can't imagine redouble conveys that message :-) I've never played top level bridge so I am used to assuming there is significant chance that they just did something silly. If I am 100% sure they know what they are doing it might be a worthy gamble to run. Sometimes I am going to be down 6 but sometimes I swinging ~30imps by turning -1 into =. As I don't believe location of Kc makes any difference to the bidding in this hand I think there is about 1/4 to 1/3 chance partner has it (as he should have 3-4 clubs on average). So yeah, if I am certain they are serious I should run based on that assumptions. On the other hand if they are so good and never brain fart why would they double anyway ? Just lead the lowest club and partner works it out. I can't imagine ever making decision to run in real play. He cuebid twice = 1st round control and we hold the ace. EDIT: One more thought I just had: With 3505 LHO's lowest club is his natural lead if they play 3/5. That might be a reason to double. If they played say 2/4 then the lowest one is often going to be readable as something strange to partner.

Well, I don't agree with "way more" and even more. The raiser has Kxx, the overcaller AJxxx... it's imo a pure flipwho has Ac here based on points. I don't see 2H on K + Q as less aggressive bid than 1H on AJxxx + Q... I might well be wrong about this assumption and then yeah, playing the Q would be correct. Right. I got carried away a bit. I didn't play much bridge in my life comapring to you guys.. but I find it very suspicious that raise on xxx + A is somewhat standard and overcall on AJxxx Q isn't. I wouldn't play the Q. I would stick to AJxxx being more likely and just play low Yes, this is my main point. I feel people are more likely to overcall with junk than to raise with junk. It might be very region specific and it might be very wrong even for my region. I feel somewhat strongly about it but on the other hand I didn't play much bridge in my life and almost none last two years so that might be totally off.

So do I get it right that you play a Q because overcaller rarely has only AJxxx/Q for his overcall and it's more likely that it's his partner who raised on xxx + Q ? I find this hand interesting because to me it's straightforward analysis to play low but somehow everybody finds playing a Q correct.

Restricted choice always applies. It just means that combinations which had other plays available are now less likely. In this hand in particular AKJ is twice less likely that it was a priori. I don't think W would bid 2H with xxx of hearts and withotu Ac (at least not too often while if E has AJxxx the location of Ac is imo more or less a flip (with W slightly more likely to have the card still). Anyway, a priori there are 10 ways to deal Kxx-AJxxx and 10 ways to deal xxx-AKJxx of which only 5 survived to the moment of our decision. It's 10-5 for Kxx now but out of those 10 cases you lose some when Ac is with E. I estimated it as less than 50% above so I am playing low expecting to make 3NT about 55-60% of the time :-)

I managed to get a script rotating the deals working. One last question: old BBO client displays overall score of the match. As I am rotating contracts/vulnerabilities as well it would be nice to have this score correct (for example to quickly see how chosen pair perform in deals the .lin was constructed from). Could you guys help me with how the overall score is calculated ? I have imp scores for individual deals displayed correctly (in scoreboard) but not the overall one (the one on the top). I would like to have score for Team one (light blue) as sum of NS scores and the other one as sum of WE's scores. The header of sample .lin I constructed: vg|,,I,1,50,,,,| rs|2SN+2,2SN+1,4SS=,4SS=,3NS+4,3NS+2,4SN-3,4SNx-3,3SNx-1,2SN-1,2DN-1,2SN-1,3NS=,3NN+1,3NS+2,3NN+1,3SS-2,1NS-2,3NN+1,3NS+1,1NS-1,1NS=,4SN-1,4SS-1,1NN+2,1NN+3,3DS+2,5DS+1,4HS=,2HS+2,4SS-1,4SS-1,2SN+1,2SN=,2SN+3,2SN+2,3NN+3,4HN+1,2SS+2,2SS+3,4SN-2,,3NN=,4SS-1,1NN=,5CN-1,2SN-2,2SN+1,2SS-1,2SS=,4SN=,4SN+1,4SS=,,3DS-1,2SS-1,4SS=,4SS=,1NS+1,3CE-1,6HN=,6HN=,2HE-2,2HE-3,4HN=,,4SS=,4SS-1,2SS-2,2SS-2,2SS-2,2HN-1,3CN=,2HN-2,4HS+1,4HS+1,1NS-1,1NS-1,5DN-1,4SS+1,4SS=,4SS=,4SS+1,4SSx+1,1NN-1,1NN-1,4SN=,4SN=,3HS=,3HS=,3SS=,3SS=,2SN=,2SN-1,2SS-1,2SS-2,2SS+1,3SS-1,1NN-1,2DE+1| Again, it's enough to fill the scoreboard with scores of individual hands, but the sum doesn't match. What I am doing is searching a database and construct .lins from hands fulfilling given criteria to produce .lins of 50hands (to use as bidding tests for example). Rotating chosen ones by hand is not an option but thanks for pointing the option out :) Thanks for help!

I was hoping for easier way but the instruction you gave is more than sufficient and I will find a way to automate the process. Thanks much for help :)

I like preparing bidding tests from vugraph archives. What I get are .lins for given situation with my favorite players playing slam/game contracts etc. Unfortunately they are sometimes on NS and sometimes on WE. I prepare two lins: one with my favorite pair on NS and the other when they are on WE. While I can use NS file easily by choosing "deal from source" I can't figure out how to use WE one. Is it possible to rotate the deals by 90 degrees or just sit on WE at partnership bidding table ? The most useful optino for me would be "use side with more hcp for our hand" option but anythin allowing me to use WE hands would be enough.

So I digged some more into Meckwell structure and apparently it's: 1N - 2N 3C - ? 3D = (13)(54) then relay for shortness 3H = 4spades-3hearts 3S = 3spades-4hearts 4C = clubs with 3card major, slammish Hands if you feel like browsing yourself (or maybe compare other methods to): https://www.dropbox.com/s/ha7u31a39ucfowm/puppetpart0.lin https://www.dropbox.com/s/96jslclbqnzin00/puppetpart1.lin

So I digged some more into Meckwell structure and apparently it's: 1N - 2N 3C - ? 3D = (13)(54) then relay for shortness 3H = 4spades-3hearts 3S = 3spades-4hearts 4C = clubs with 3card major, slammish Hands if you feel like browsing yourself: https://www.dropbox.com/s/ha7u31a39ucfowm/puppetpart0.lin https://www.dropbox.com/s/96jslclbqnzin00/puppetpart1.lin

I took a look at vugraph hands of which they are many now and it looks like standardish Italian system with weak NT mixed in some vulnerabilities. So they open 5 card majors and 1D keeps changing but as of now I believe is 5+ (unless 4-4-4D-1). After 2/1 there is ton of relays, there is Gazilli with again relays to the end (similar as Bocchi played with Duboin I believe). Then they put the trouble hand 18-19bal in 2C gaining a step over Lauria-Versace's 2D but it's still over 1NT :-) They hope GF hand don't come around often and they allocate 2D opening for that. 2H/2S are some novel preempts if not vul. As to 1C opening they bid suits up the line: that is 1D is hearts, 1H is spades and 1S is diamonds while 1NT is rare but in general means 3334 or (233)5C Then as there is no 18-19 in 1C they are able to have all he usual multi reverses available and don't need to jump to 2NT with 18bal opposite possible bust. If you have any particular question I can try to dig up the sequences but they change stuff often so dunno how the things are as of right now. As to general style they open a lot of crap at 3level non-vul but are disciplined vul. They bid any game that possibly could be bid. They open 1M with 5M-3-3-2 15-17 and as other Italian pairs default to game if that hits any 8PC in partner's hand.

While I think it's not the best idea in the world it could actually happen. Funnily enough I had this conversation with my now pro player friend 2 days ago in which he whined to me that 3D a puppet they play never gains and lost several times already to which I replied with my standard "yeah, it sucks what else is new". They play 1N with 5M332 additionally to Gazilli after 1M-1N... I fail to see the logic in that but w/e. They say those 5M-3-3-2's screw their fancy relay structure after 1M-2C because then some particular 6-4-3-0 wouldn't be biddable. Besides they won the Spingold and not me so what do I know :( Btw, to add somethign constructive to this post link to Pavlicek's stats: http://www.rpbridge.net/9x14.htm http://www.rpbridge.net/9x13.htm Not many hands, still very significant difference! As to puppet, yeah if you must, play it as 2NT and then after 3C (no 5cdM) add some bids to show (1-3)-(5-4) as you want that in your system. Similarly 3361 and 3316 is probably a good idea to puppet with, so that's some hands already :-) I don't know what exactly Meckwell play but: 1N 2N 3C 3H/3S = stiff, 3N = to play, 3D = 33(61) looks reasonable. If you do that then you can even puppet with those 3M-(4-4-2)'s as they won't double your 3C so 0.6imps once every 11 boards is worth collecting! Then there is 3-3-(5-2) as well which may actually be significant gain as with 3-3 in majors you hit the jackpot twice as often. When I look at this structure I am actually getting excited about the puppet thing! I don't exactly follow the structure from OP, but you if you do that with 5431 (5-3M) then you need something different with 5M-3M-(3-2) as you really don't want to end up in 4M with that opposite 4-3-3-3! This one sucks hard as these: http://www.rpbridge.net/8z17.htm stats show.

Right I missed it. It's not very likely they missed one particular hole but it's very likely there are some holes they missed. Anyway as to Meckwell they've changed 3C puppet to 2NT fixing the biggest problem of the convention (lead directing double) and put 5431's hands (as well as some more slammish hands) there which is important hand type to be able to show opposite 1NT bid. I like their version of the convention and if I every played a system which forces me to open 1N with 5M332 I would consider it.

So what do you do with say 3-5-1-4 hand and no slam interest ? The thing is there is a lot of equity in playing in 3NT if one have 5M-3-3-2 opposite 4-3-3-3 so if transfer and jump to 3NT just asks to convert with fit you lose this opportunity and if opener is allowed to leave it then you play silly 3nt with singleton to 3 card suit. In general looking for 5 card majors after 1NT is -EV play at imps. With something like 2-3-4-4 you have 1 in 11 chance of finding your fit and then you make ~6% more often which gives you ~0.6imp once every 11 tries. In remaining hands they have 10 chances to hurt you by making lead directional double (or from inference from one not appearing) and defending better in middle hand. One overtrick is enough to kill your profit and your loses are going to be more than that. The only way you can make puppet work is if you somehow include other crucial hands there and discover 5-3 fits by accident. Again it's not worth much even if you succeed at not leaking information and it's better to just forget about it and focus on more profitable situations (like playing in 3nt with 5M-3-3-2 opposite 4-3-3-3).

So somehow I lost the oldest version of the client I had and in the current one when I open .lins using it the GIB button is inactive :( On the other hand if I just open the NetBridgeVu.exe and select open bridge movie from there then GIB button magically works. I like browsing .lins having double dummy button available. Any way to fix it ? (or maybe someone could send me some older version of NetBridgeVu).

Just wanted to say that my regular BBO partner with whom I played maybe 10k hands together just won Spingold main event. He is Wojtek Gawel (redds on bbo). Congrats to him! I feel great as well as we practically learnt to play bridge together before he took a shot at going pro with one of the big Polish names (Rafal Jagniewski). I knew they have world class potential but winning Spingold and showing it to the world is whole different story. Amount of work they did in last 1.5 year is staggering (especially system work). Winning their first big US event is a dream scenario coming true :) Oh, and they had most difficult draw possible beating Bermuda Bowl winning Netherland pairs (Zagorin), legendary Nickell, Bocchi-Madala and Buras-Narkiewicz (Gromov) in the semi and Grue in the final.

I think opening 1NT with a singleton is about always a bad idea (the only exception is 3rd hand non vs vul for tactical reasons). I see the reasons but those are usually caused by fixable holes in rest of the system. I went through many hands and imo people who regularly open 1NT with stiff (many American pairs, not so many European) hurt themselves by doing this. As to specific problems: 1D - 1H 1S - 1NT 2C = 16+, any hand here is one good idea played by Lauria Versace for example. 1D - 2C many ideas here, one is that you just bid 2NT with 12-14 and 4 diamonds; one other was describe by Mikeh. Some others require 2H as multimeaning rebid. I don't think those problems are especially hard to solve nor frequent comapre to disasters opening 1NT with stiff yield.

If you have good enough understanding about what youare showing Smith Echo (aka substite attitude) is exactly the same thing as s/p just expressed differently.

Does it happen though ? I mean it seems it me it's rare to be able to benefit. I for example have few people against who I am determined to play for dropping Qx with Axxx - KJxx if T doesn't appear (cause if they have Tx they would always trying to "fool") but I am yet to have chances of executing :)

This is the most basic signal and every player here play it ranging from novices to experts. I think it's good idea to combine it with OS'ish idea - that is to have some rules establishing what the most important information is and trump signal (or first side suit we don't give count in) should convey it. Usually that information is high honor in a suit which is most likely to be switched to. Notice that smith echo is exactly the same idea just called differently. It doesn't matter if you use s/p or substitute attitude for given suit as those are the same thing. If you choose s/p (like in standard trump signal) you give false s/p if you are afraid to mislead partner when it comes to most important suit. If you choose substitute attitude (like in smith echo) you just discourage. If you have any experience with careful signalling you will quickly start expecting specific information once the dummy i placed. It will "well, ok declarer will take this and it looks like we might want to open clubs, let's see if my pd confirms/denies an honor there once declarer touches trumps".

When I played a lot of online bridge I discussed it with my pd (after some disaster in 1H p 4H p p d sequence). We estabslihed that dbl is t/o if they bid and jump raise hearts and penalty if their suit is spades. My opinion is that it doesn't matter much :)

They are playing very well. Impressive.

So how do you manage to ruff a club if opening leader has AKxx and declarer xxx ?