bluecalm Posted February 18, 2011 Report Share Posted February 18, 2011 I had some fun doing this so I will share :)I compiled hands from major events from 1995 to 2009 (unfortunately I wasn't able to get hands from 2010 yet), calculated dd results for them (thanks to awesome double dummy solver written by Piotr Beling, http://bcalc.w8.pl/index.php?lang=en&topic=abou) and made various things with them. First thing is to see which player outscores double result when declaring. This is done like this: for every hand dd result in actual contract bid is subtracted from result at the table. This is then converted to IMPs then summed for all hands.What are major events ? My completely arbitrary choice was : world teamc championship, Olympiad, WMSG, Rosemblum, Spingold, Vanderbilt, USBC, European Team Championships and European Champions Cup as well as Cap Gemini and Forbo tournaments. The results (first column is total imps won, second total hands played, third average per hand). I didn't make much attempt to remove hands from senior tournaments. They shouldn't be there but if somehow they were thrown into vugraph folders for major events then the stats will include them. Ok, have fun:(the data only contains players who declared at least 250 hands on vugraph) Elinescu : 248 , 285 , avg: 0.87Gitelman : 341 , 464 , avg: 0.73FANTONI : 546 , 767 , avg: 0.71Levin : 312 , 441 , avg: 0.71pszczola : 170 , 250 , avg: 0.68Cronier : 229 , 343 , avg: 0.67VERSACE : 1047 , 1607 , avg: 0.65WLADOW : 235 , 374 , avg: 0.63helgemo : 424 , 712 , avg: 0.60ROSENBERG : 418 , 739 , avg: 0.57bertens : 218 , 407 , avg: 0.54BOCCHI : 630 , 1190 , avg: 0.53Levy : 213 , 409 , avg: 0.52Mari : 144 , 278 , avg: 0.52Multon : 159 , 309 , avg: 0.51WEINSTEIN : 468 , 916 , avg: 0.51Hampson : 191 , 375 , avg: 0.51bakkeren : 144 , 285 , avg: 0.51Sontag : 216 , 436 , avg: 0.50sementa : 206 , 420 , avg: 0.49Cohen : 303 , 618 , avg: 0.49BALICKI : 269 , 557 , avg: 0.48DUBOIN : 617 , 1278 , avg: 0.48HAMMAN : 428 , 919 , avg: 0.47RODWELL : 569 , 1293 , avg: 0.44Meckstroth : 593 , 1380 , avg: 0.43FALLENIUS : 157 , 367 , avg: 0.43MARTENS : 112 , 268 , avg: 0.42Jansma : 129 , 309 , avg: 0.42Garner : 170 , 410 , avg: 0.41Zia : 409 , 1008 , avg: 0.41lauria : 629 , 1562 , avg: 0.40Branco : 180 , 449 , avg: 0.40piekarek : 142 , 359 , avg: 0.40Lanzarott : 125 , 319 , avg: 0.39SEAMON : 134 , 343 , avg: 0.39Brogelan : 191 , 495 , avg: 0.39Muller : 181 , 472 , avg: 0.38Nystrom : 127 , 334 , avg: 0.38Soloway : 206 , 549 , avg: 0.38Lindkvis : 110 , 298 , avg: 0.37WEICHSEL : 119 , 355 , avg: 0.34BOMPIS : 148 , 446 , avg: 0.33Jonsson : 135 , 407 , avg: 0.33Helness : 249 , 769 , avg: 0.32FREDIN : 141 , 443 , avg: 0.32Stansby : 183 , 577 , avg: 0.32nickell : 176 , 568 , avg: 0.31Ramondt : 92 , 302 , avg: 0.30PASSELL : 91 , 301 , avg: 0.30WESTRA : 109 , 371 , avg: 0.29KOWALSKI : 81 , 286 , avg: 0.28Ekeblad : 111 , 404 , avg: 0.27Dubinin : 96 , 352 , avg: 0.27Nunes : 107 , 400 , avg: 0.27Gromov : 146 , 553 , avg: 0.26HACKETT : 225 , 862 , avg: 0.26GAWRYS : 72 , 305 , avg: 0.24CHAGAS : 112 , 480 , avg: 0.23freeman : 125 , 538 , avg: 0.23TUSZYNSKI : 76 , 337 , avg: 0.23Martel : 97 , 446 , avg: 0.22Moss : 83 , 406 , avg: 0.20Schaltz : 52 , 272 , avg: 0.19JASSEM : 53 , 314 , avg: 0.17Quantin : 51 , 304 , avg: 0.17Mouiel : 58 , 363 , avg: 0.16Drijver : 43 , 281 , avg: 0.15Berkowitz : 93 , 664 , avg: 0.14bertheau : 49 , 415 , avg: 0.12Wang : 33 , 395 , avg: 0.08Chemla : 22 , 406 , avg: 0.05ZMUDZINSKI : -24 , 406 , avg: -0.06WELLAND : -33 , 401 , avg: -0.08Jacobs : -40 , 408 , avg: -0.10buratti : -42 , 348 , avg: -0.12SAELENSMI : -48 , 330 , avg: -0.15 This is just for fun purposes. Obviously some players face tougher opponents and sample size is small enough to make variance very significant, so I don't think any serious conclusions could be drew from the data :) 3 Quote Link to comment Share on other sites More sharing options...
Mbodell Posted February 19, 2011 Report Share Posted February 19, 2011 This is just for fun purposes. Obviously some players face tougher opponents and sample size is small enough to make variance very significant, so I don't think any serious conclusions could be drew from the data :) One very obvious conclusion is that at the level involved people are better declarers then defenders in that they tend to score more tricks than double dummy suggests. I wonder to what degree the opening lead effects things. I.e., if you calculated their score versus double dummy only after the opening lead (so double dummy needs to make the same lead), does that negate much of the declarer's advantage or not? Quote Link to comment Share on other sites More sharing options...
bluecalm Posted February 19, 2011 Author Report Share Posted February 19, 2011 I wonder to what degree the opening lead effects things. I.e., if you calculated their score versus double dummy only after the opening lead (so double dummy needs to make the same lead), does that negate much of the declarer's advantage or not? This is one idea I had alredy. Both calculating opening lead "success" and declaring after opening lead. I will do that once I am back home :)The conclusion that declarers have advantage over defenders couldn't be drawn from this sample because I deleted all the people who declared less than 250 hands on vugraph. I will calculate what advantage (if any) declarer has for whole sample though. Quote Link to comment Share on other sites More sharing options...
daveharty Posted February 19, 2011 Report Share Posted February 19, 2011 Fascinating, thanks for this. I was curious to see if the results aligned well with conventional wisdom. I would have guessed Michael Rosenberg, and he did very well. Quote Link to comment Share on other sites More sharing options...
bluecalm Posted February 19, 2011 Author Report Share Posted February 19, 2011 I will have to double check the calculations but quick script shows me that declarers won 27610imps over dd expectation during 97706 hands which gives them 0.282 imps advantage per hand.This is also 0.067 tricks per hand. Quote Link to comment Share on other sites More sharing options...
hrothgar Posted February 19, 2011 Report Share Posted February 19, 2011 I really don't find results like this particularly convincing...It's impossible to decouple bidding and play. If you really want to make valid comparisons regarding a single facet of bridge, you need to design a test that isolates this dimension.There have been par contests in living memory...I think that those results are a lot more compelling. I've occasionally wondered whether it would be possible to writing a computer program to generate hands for contests.My understanding is that one of the main factors limiting par contests is the expense involved to create hands. 2 Quote Link to comment Share on other sites More sharing options...
bluecalm Posted February 24, 2011 Author Report Share Posted February 24, 2011 Some more statistics: Best declarer after opening lead (actual result is compared to double dummy result after first lead is made):(at least 250 hands in results) Levin : -92 , 580 , avg: -0.16FANTONI : -127 , 767 , avg: -0.17bertens : -77 , 406 , avg: -0.19pszczola : -47 , 240 , avg: -0.20helgemo : -157 , 709 , avg: -0.22VERSACE : -381 , 1600 , avg: -0.24Nunes : -191 , 752 , avg: -0.25SEAMON : -88 , 338 , avg: -0.26WLADOW : -100 , 374 , avg: -0.27WEINSTEIN : -246 , 909 , avg: -0.27Sontag : -119 , 435 , avg: -0.27BOCCHI : -328 , 1184 , avg: -0.28DUBOIN : -356 , 1264 , avg: -0.28Nystrom : -94 , 333 , avg: -0.28piekarek : -103 , 356 , avg: -0.29Mari : -84 , 275 , avg: -0.31ROSENBERG : -224 , 730 , avg: -0.31BOMPIS : -138 , 446 , avg: -0.31Brogelan : -155 , 489 , avg: -0.32bakkeren : -92 , 285 , avg: -0.32sementa : -143 , 419 , avg: -0.34Hampson : -128 , 375 , avg: -0.34Helness : -266 , 764 , avg: -0.35BALICKI : -190 , 545 , avg: -0.35Levy : -196 , 562 , avg: -0.35JASSEM : -108 , 308 , avg: -0.35Meckstroth : -488 , 1369 , avg: -0.36Muller : -166 , 462 , avg: -0.36Zia : -188 , 523 , avg: -0.36Lanzarott : -116 , 318 , avg: -0.36HAMMAN : -333 , 908 , avg: -0.37WESTRA : -136 , 369 , avg: -0.37Mahmood : -177 , 478 , avg: -0.37Multon : -115 , 309 , avg: -0.37RODWELL : -485 , 1286 , avg: -0.38nickell : -219 , 565 , avg: -0.39Gitelman : -183 , 463 , avg: -0.40WEICHSEL : -140 , 353 , avg: -0.40lauria : -613 , 1543 , avg: -0.40Katz : -121 , 303 , avg: -0.40Branco : -183 , 449 , avg: -0.41FREDIN : -181 , 441 , avg: -0.41Stansby : -237 , 574 , avg: -0.41Moss : -227 , 539 , avg: -0.42Wang : -282 , 667 , avg: -0.42Cohen : -327 , 771 , avg: -0.42Soloway : -232 , 543 , avg: -0.43Jonsson : -180 , 406 , avg: -0.44Drijver : -126 , 281 , avg: -0.45freeman : -239 , 531 , avg: -0.45BALDURSS : -109 , 242 , avg: -0.45HACKETT : -393 , 859 , avg: -0.46KOWALSKI : -131 , 286 , avg: -0.46FALLENIUS : -166 , 362 , avg: -0.46TUSZYNSKI : -152 , 329 , avg: -0.46Greco : -168 , 360 , avg: -0.47Ekeblad : -189 , 404 , avg: -0.47Elinescu : -137 , 284 , avg: -0.48Gromov : -270 , 552 , avg: -0.49MOLLER : -118 , 240 , avg: -0.49CHAGAS : -235 , 476 , avg: -0.49MARTENS : -133 , 267 , avg: -0.50Berkowitz : -341 , 653 , avg: -0.52Brink : -161 , 301 , avg: -0.53Ramondt : -157 , 293 , avg: -0.54Jansma : -167 , 306 , avg: -0.55Dubinin : -192 , 351 , avg: -0.55Lindkvis : -168 , 293 , avg: -0.57Garner : -232 , 403 , avg: -0.58GAWRYS : -177 , 304 , avg: -0.58Chemla : -235 , 393 , avg: -0.60Martel : -273 , 436 , avg: -0.63PASSELL : -189 , 300 , avg: -0.63bertheau : -270 , 412 , avg: -0.66Schaltz : -179 , 272 , avg: -0.66Quantin : -212 , 304 , avg: -0.70ZMUDZINSKI : -308 , 395 , avg: -0.78Mouiel : -292 , 362 , avg: -0.81WELLAND : -331 , 400 , avg: -0.83SAELENSMI : -275 , 325 , avg: -0.85buratti : -292 , 344 , avg: -0.85Jacobs : -356 , 403 , avg: -0.88 Average: -0.49 Overall the best defenders after first lead is made: (at least 500 hands): BOCCHI : 2491 , 2335 , avg: 1.07FANTONI : 1519 , 1525 , avg: 1.00Nunes : 1519 , 1525 , avg: 1.00DUBOIN : 2650 , 2666 , avg: 0.99Mahmood : 1014 , 1045 , avg: 0.97buratti : 589 , 681 , avg: 0.86Lanzarott : 579 , 683 , avg: 0.85GAWRYS : 470 , 563 , avg: 0.83CHAGAS : 788 , 953 , avg: 0.83Branco : 750 , 909 , avg: 0.83ZMUDZINSKI : 769 , 945 , avg: 0.81VERSACE : 2605 , 3217 , avg: 0.81BALICKI : 832 , 1028 , avg: 0.81lauria : 2451 , 3059 , avg: 0.80JASSEM : 452 , 594 , avg: 0.76Kirmse : 373 , 507 , avg: 0.74Quantin : 439 , 597 , avg: 0.74Gitelman : 681 , 934 , avg: 0.73RODWELL : 1687 , 2334 , avg: 0.72Meckstroth : 1678 , 2331 , avg: 0.72CHMURSKI : 353 , 505 , avg: 0.70Helness : 918 , 1324 , avg: 0.69pszczola : 366 , 532 , avg: 0.69Muller : 585 , 858 , avg: 0.68Moss : 660 , 970 , avg: 0.68BOMPIS : 594 , 884 , avg: 0.67Soloway : 714 , 1079 , avg: 0.66helgemo : 871 , 1338 , avg: 0.65gromoeller : 332 , 512 , avg: 0.65MARTENS : 365 , 571 , avg: 0.64ROSENBERG : 943 , 1496 , avg: 0.63TUSZYNSKI : 420 , 682 , avg: 0.62HAMMAN : 1083 , 1773 , avg: 0.61Chemla : 532 , 876 , avg: 0.61HACKETT : 1048 , 1737 , avg: 0.60bertheau : 526 , 873 , avg: 0.60WLADOW : 362 , 605 , avg: 0.60Levy : 679 , 1178 , avg: 0.58Elinescu : 356 , 621 , avg: 0.57Schaltz : 289 , 533 , avg: 0.54Sontag : 520 , 963 , avg: 0.54nickell : 657 , 1229 , avg: 0.53Zia : 521 , 985 , avg: 0.53freeman : 587 , 1119 , avg: 0.52Mouiel : 345 , 662 , avg: 0.52Dubinin : 391 , 756 , avg: 0.52Nystrom : 372 , 726 , avg: 0.51WEICHSEL : 378 , 741 , avg: 0.51Cohen : 779 , 1533 , avg: 0.51Berkowitz : 696 , 1408 , avg: 0.49Levin : 557 , 1138 , avg: 0.49Garner : 406 , 870 , avg: 0.47KOWALSKI : 277 , 605 , avg: 0.46Brogelan : 443 , 972 , avg: 0.46Drijver : 257 , 577 , avg: 0.45Hampson : 311 , 713 , avg: 0.44Greco : 311 , 713 , avg: 0.44bertens : 339 , 778 , avg: 0.44Stansby : 508 , 1180 , avg: 0.43WESTRA : 275 , 641 , avg: 0.43Brink : 268 , 636 , avg: 0.42Verhees : 229 , 548 , avg: 0.42Jansma : 252 , 608 , avg: 0.41Jacobs : 319 , 775 , avg: 0.41Katz : 266 , 648 , avg: 0.41piekarek : 303 , 744 , avg: 0.41Jonsson : 282 , 706 , avg: 0.40Gromov : 410 , 1027 , avg: 0.40Ramondt : 226 , 572 , avg: 0.40SAELENSMI : 247 , 632 , avg: 0.39Wellan : 308 , 806 , avg: 0.38Martel : 356 , 960 , avg: 0.37Wang : 501 , 1353 , avg: 0.37WEINSTEIN : 627 , 1705 , avg: 0.37bakkeren : 213 , 623 , avg: 0.34Ekeblad : 291 , 864 , avg: 0.34FREDIN : 250 , 763 , avg: 0.33Multon : 200 , 619 , avg: 0.32Lindkvis : 159 , 505 , avg: 0.31Schwartz : 156 , 505 , avg: 0.31FALLENIUS : 219 , 795 , avg: 0.28sementa : 209 , 871 , avg: 0.24PASSELL : 115 , 513 , avg: 0.22SEAMON : 78 , 570 , avg: 0.14 Average: +0.49 Have fun :) (btw anybody knows how to edit so every space will have exact same size as every letter ?) 2 Quote Link to comment Share on other sites More sharing options...
awm Posted February 24, 2011 Report Share Posted February 24, 2011 I think it's interesting that almost all the declarers are positive prior to the opening lead, but they're all negative after the opening lead. This reinforces my opinion that opening lead is hard. :) Quote Link to comment Share on other sites More sharing options...
Vampyr Posted February 24, 2011 Report Share Posted February 24, 2011 Which HACKETT is the one in the list? Quote Link to comment Share on other sites More sharing options...
Bbradley62 Posted February 24, 2011 Report Share Posted February 24, 2011 btw anybody knows how to edit so every space will have exact same size as every letter?Change your font to "Courier New"Levin : -92 , 580 , avg: -0.16FANTONI : -127 , 767 , avg: -0.17bertens : -77 , 406 , avg: -0.19Well, that didn't work... Courier New makes every letter (and space) the same width, like an old-fashioned typewriter, so "FANTONI" and "bertens" are the same width, unlike in the original post. Unfortunately, it appears that the Forum removes double-blanks, so you'd have to use _ or something else instead of blanks. Quote Link to comment Share on other sites More sharing options...
CSGibson Posted February 24, 2011 Report Share Posted February 24, 2011 What I find interesting is the amount of advantage some declarers get from the opening lead. For example, Fred is one of the best declarers on this list before the opening lead, and about halfway down after the opening lead. I wonder how much of that is unrevealing auctions vs misleading auctions vs luck... Quote Link to comment Share on other sites More sharing options...
gnasher Posted February 24, 2011 Report Share Posted February 24, 2011 (btw anybody knows how to edit so every space will have exact same size as every letter ?) Put it in a code block like this Or use a fixed width font like Courier and replace all the spaces with the word "space" in square brackets like this Quote Link to comment Share on other sites More sharing options...
the hog Posted February 24, 2011 Report Share Posted February 24, 2011 Copy the data into a fixed width file rather than comma delimited. Quote Link to comment Share on other sites More sharing options...
BillHiggin Posted February 24, 2011 Report Share Posted February 24, 2011 Zia : -188 , 523 , avg: -0.36Mahmood : -177 , 478 , avg: -0.37 He does about as well under either name ;) Quote Link to comment Share on other sites More sharing options...
BillHiggin Posted February 24, 2011 Report Share Posted February 24, 2011 Which HACKETT is the one in the list?Since the data seems to have been pulled from vugraph records, the statistics in the list will be composite stats (i.e. both intermixed). Quote Link to comment Share on other sites More sharing options...
Vampyr Posted February 24, 2011 Report Share Posted February 24, 2011 Since the data seems to have been pulled from vugraph records, the statistics in the list will be composite stats (i.e. both intermixed).Both? Or all three? Quote Link to comment Share on other sites More sharing options...
bluecalm Posted February 24, 2011 Author Report Share Posted February 24, 2011 This reinforces my opinion that opening lead is hard. I think there is a lot of variance in the opening lead so proably that's why the results differ that much. On average opening leads gives away about 1 imp/hand (can't remember how much exactly now) which is more than play is "worth" (in double dummy terms) after it. Which HACKETT is the one in the list? Unfortunately the stats are limited by what vugraph operators put in to the files when filing up the names.I cleaned this mess a bit by joining all the 6+letter names into one category (all the stats go to the shortes one on the list and "eat" all the ones which are longer but contain the shorter) so all the:"Fantoni", "FANTONI", "Ful Fantoni", "Fantoni F." etc. becomes just "FANTONI" I also manually joined some of the shorter names ("Nunes", "Brink", "Levin", "Cohen", "Greco", "Moss", "Levy", "Wang").This has side effect of making stats for Cohen, Hackett, Bessiss and Lall (there are 80 hands of "Lall" in those ;) basically useless becaue there are more players with this exact name.(I didn't have this automatic names cleaner when I posted first part of stats that's why Levin has less hands there)I also forgot to manually join Zia and Mahmood as you can see on the list :)I will recompile later joning all the Zia's and Mahmood's. Or use a fixed width font like Courier and replace all the spaces with the word "space" in square brackets Thanks :) Quote Link to comment Share on other sites More sharing options...
bluecalm Posted February 24, 2011 Author Report Share Posted February 24, 2011 I've moved overall stats here cause I was unable to make it work in the previous one (font formatting just collapsed near the end of the post) BOCCHI : 2163 , 3519 , avg: 0.61FANTONI : 1392 , 2292 , avg: 0.61DUBOIN : 2294 , 3930 , avg: 0.58Nunes : 1328 , 2277 , avg: 0.58Mahmood : 837 , 1523 , avg: 0.55Lanzarott : 463 , 1001 , avg: 0.46VERSACE : 2224 , 4817 , avg: 0.46Branco : 567 , 1358 , avg: 0.42pszczola : 319 , 772 , avg: 0.41BALICKI : 642 , 1573 , avg: 0.41lauria : 1838 , 4602 , avg: 0.40CHAGAS : 553 , 1429 , avg: 0.39JASSEM : 344 , 902 , avg: 0.38Gitelman : 498 , 1397 , avg: 0.36helgemo : 714 , 2047 , avg: 0.35ZMUDZINSKI : 461 , 1340 , avg: 0.34BOMPIS : 456 , 1330 , avg: 0.34GAWRYS : 293 , 867 , avg: 0.34RODWELL : 1202 , 3620 , avg: 0.33ROSENBERG : 719 , 2226 , avg: 0.32Meckstroth : 1190 , 3700 , avg: 0.32Muller : 419 , 1320 , avg: 0.32Helness : 652 , 2088 , avg: 0.31HACKETT : 709 , 2372 , avg: 0.30Soloway : 482 , 1622 , avg: 0.30buratti : 297 , 1025 , avg: 0.29Moss : 433 , 1509 , avg: 0.29Sontag : 401 , 1398 , avg: 0.29Nystrom : 293 , 1043 , avg: 0.28HAMMAN : 750 , 2681 , avg: 0.28Levy : 483 , 1740 , avg: 0.28MARTENS : 232 , 838 , avg: 0.28Levin : 465 , 1718 , avg: 0.27WLADOW : 262 , 979 , avg: 0.27TUSZYNSKI : 268 , 1011 , avg: 0.27Quantin : 227 , 901 , avg: 0.25Mari : 189 , 761 , avg: 0.25nickell : 438 , 1794 , avg: 0.24Elinescu : 219 , 905 , avg: 0.24Chemla : 297 , 1269 , avg: 0.23bertens : 262 , 1184 , avg: 0.22Zia : 333 , 1508 , avg: 0.22WEICHSEL : 238 , 1094 , avg: 0.22freeman : 348 , 1650 , avg: 0.21Verhees : 159 , 783 , avg: 0.20bertheau : 256 , 1285 , avg: 0.20Brogelan : 288 , 1461 , avg: 0.20Cohen : 452 , 2304 , avg: 0.20Berkowitz : 375 , 2051 , avg: 0.18piekarek : 200 , 1100 , avg: 0.18Dubinin : 199 , 1107 , avg: 0.18Hampson : 183 , 1088 , avg: 0.17KOWALSKI : 146 , 891 , avg: 0.16Stansby : 271 , 1754 , avg: 0.15Drijver : 131 , 858 , avg: 0.15Katz : 145 , 951 , avg: 0.15WEINSTEIN : 393 , 2599 , avg: 0.15Garner : 186 , 1258 , avg: 0.15WESTRA : 139 , 1010 , avg: 0.14Schaltz : 110 , 805 , avg: 0.14Greco : 143 , 1073 , avg: 0.13bakkeren : 121 , 908 , avg: 0.13Brink : 107 , 937 , avg: 0.11Wang : 219 , 2020 , avg: 0.11Jansma : 85 , 914 , avg: 0.09Jonsson : 102 , 1112 , avg: 0.09Multon : 85 , 928 , avg: 0.09Gromov : 140 , 1579 , avg: 0.09Ekeblad : 102 , 1268 , avg: 0.08Ramondt : 69 , 865 , avg: 0.08Martel : 83 , 1396 , avg: 0.06FREDIN : 69 , 1204 , avg: 0.06Mouiel : 53 , 1024 , avg: 0.05sementa : 66 , 1290 , avg: 0.05FALLENIUS : 37 , 1142 , avg: 0.03SEAMON : -10 , 908 , avg: -0.01Lindkvis : -9 , 798 , avg: -0.01SAELENSMI : -28 , 957 , avg: -0.03Jacobs : -37 , 1178 , avg: -0.03WELLAND : -39 , 1191 , avg: -0.03PASSELL : -74 , 813 , avg: -0.09 Average: 0.166imp/hand(Because defenders have advantage and every score is counted for only one declarer but 2 defenders, so average is above 0 here) 1 Quote Link to comment Share on other sites More sharing options...
bluecalm Posted February 24, 2011 Author Report Share Posted February 24, 2011 And the last one which probably doesn't mean much at all: The best bidders (assuming opening lead is part of the bidding, so the double dummy result is taken after opening lead is made and this is compared to par result for whole hand).Overall those stats are influenced by variance, opponent selection etc. bu they seem to confirm some things which I think are true (that Balicki is better declarer than Zmudzinski, that Italians are the best in defence and that Balicki-Zmudzinski's weak point is bidding)(at least 800 hands to include some new players) Forrester : 440 , 819 , avg: 0.54Drijver : 443 , 1132 , avg: 0.39Brink : 477 , 1222 , avg: 0.39Lanzarott : 511 , 1347 , avg: 0.38Kirmse : 330 , 951 , avg: 0.35buratti : 448 , 1341 , avg: 0.33BALDURSS : 322 , 970 , avg: 0.33Chemla : 539 , 1684 , avg: 0.32MARTENS : 351 , 1114 , avg: 0.32Mahmood : 624 , 2014 , avg: 0.31Nystrom : 421 , 1368 , avg: 0.31PASSELL : 323 , 1053 , avg: 0.31GAWRYS : 319 , 1112 , avg: 0.29FANTONI : 865 , 3044 , avg: 0.28Nunes : 865 , 3044 , avg: 0.28Hampson : 390 , 1448 , avg: 0.27Greco : 390 , 1448 , avg: 0.27gromoeller : 258 , 967 , avg: 0.27Ramondt : 299 , 1141 , avg: 0.26lauria : 1450 , 6134 , avg: 0.24Brogelan : 422 , 1922 , avg: 0.22Quantin : 254 , 1193 , avg: 0.21VERSACE : 1353 , 6437 , avg: 0.21CHAGAS : 372 , 1929 , avg: 0.19CHMURSKI : 183 , 1008 , avg: 0.18Lindkvis : 193 , 1087 , avg: 0.18RODWELL : 851 , 4986 , avg: 0.17Meckstroth : 814 , 4987 , avg: 0.16SAELENSMI : 207 , 1270 , avg: 0.16KOWALSKI : 189 , 1197 , avg: 0.16Multon : 194 , 1259 , avg: 0.15SEAMON : 186 , 1210 , avg: 0.15Becker : 132 , 859 , avg: 0.15FREDIN : 247 , 1625 , avg: 0.15Jonsson : 222 , 1476 , avg: 0.15Rubin : 138 , 933 , avg: 0.15WLADOW : 181 , 1250 , avg: 0.14Elinescu : 168 , 1286 , avg: 0.13Ekeblad : 215 , 1678 , avg: 0.13Mari : 216 , 1723 , avg: 0.13ROSENBERG : 373 , 2980 , avg: 0.13Zia : 221 , 1983 , avg: 0.11Bramley : 97 , 902 , avg: 0.11BOMPIS : 157 , 1787 , avg: 0.09Verhees : 90 , 1025 , avg: 0.09Branco : 155 , 1783 , avg: 0.09Stansby : 198 , 2309 , avg: 0.09Helness : 228 , 2782 , avg: 0.08helgemo : 192 , 2813 , avg: 0.07Skrzypczak : 61 , 902 , avg: 0.07Cohen : 201 , 3010 , avg: 0.07bertheau : 112 , 1716 , avg: 0.07Stewart : 44 , 805 , avg: 0.05WELLAND : 84 , 1548 , avg: 0.05Sontag : 89 , 1855 , avg: 0.05Martel : 88 , 1840 , avg: 0.05sementa : 78 , 1679 , avg: 0.05BOCCHI : 196 , 4637 , avg: 0.04Jansma : 46 , 1184 , avg: 0.04Manoppo : 31 , 836 , avg: 0.04BALICKI : 57 , 1998 , avg: 0.03DUBOIN : 129 , 5263 , avg: 0.02WEICHSEL : 32 , 1458 , avg: 0.02HAMMAN : 51 , 3552 , avg: 0.01Soloway : 21 , 2205 , avg: 0.01WEINSTEIN : 30 , 3430 , avg: 0.01Gierulski : 4 , 888 , avg: 0.00Berkowitz : 8 , 2797 , avg: 0.00ZMUDZINSKI : -10 , 1842 , avg: -0.01TUSZYNSKI : -9 , 1345 , avg: -0.01Garner : -21 , 1701 , avg: -0.01Robinson : -15 , 887 , avg: -0.02Levy : -64 , 2360 , avg: -0.03bertens : -43 , 1555 , avg: -0.03WESTRA : -45 , 1391 , avg: -0.03Wang : -103 , 2696 , avg: -0.04Kholomeev : -41 , 917 , avg: -0.04MOLLER : -42 , 934 , avg: -0.04bakkeren : -80 , 1226 , avg: -0.07Gitelman : -125 , 1889 , avg: -0.07Moss : -170 , 1993 , avg: -0.09HACKETT : -323 , 3509 , avg: -0.09Dubinin : -149 , 1514 , avg: -0.10Levin : -246 , 2341 , avg: -0.11Wolff : -90 , 824 , avg: -0.11FALLENIUS : -184 , 1518 , avg: -0.12Schaltz : -163 , 1082 , avg: -0.15Jacobs : -265 , 1542 , avg: -0.17JASSEM : -212 , 1194 , avg: -0.18Schwartz : -177 , 952 , avg: -0.19freeman : -407 , 2164 , avg: -0.19Katz : -256 , 1296 , avg: -0.20nickell : -476 , 2372 , avg: -0.20SMIRNOV : -199 , 911 , avg: -0.22Gromov : -467 , 2099 , avg: -0.22Zhuang : -202 , 812 , avg: -0.25pszczola : -259 , 1023 , avg: -0.25Muller : -488 , 1746 , avg: -0.28Mouiel : -395 , 1344 , avg: -0.29piekarek : -434 , 1436 , avg: -0.30Smith : -274 , 852 , avg: -0.32Sun : -434 , 839 , avg: -0.52 Average: -0.0009685442650556978 imp/hand so almost 0 (obviously :) ). By the way, if someone has vupgrah archives from 2010 on their hard drive please contact me. I don't want to kill bbo vu archives server with 100's of requests :) 1 Quote Link to comment Share on other sites More sharing options...
gwnn Posted February 24, 2011 Report Share Posted February 24, 2011 What if you discount hands where Buratti-Lanzarotti faced guesses in 9 card trump suits? :) Quote Link to comment Share on other sites More sharing options...
cherdano Posted February 24, 2011 Report Share Posted February 24, 2011 Comparing your two declarer play list, it looks like Wladow-Elinescu gain a HUDGE edge on the opponents' opening leads? Btw, it would be great if you could give error estimates for your numbers. Doesn't have to be precise, e.g. just the variance for, say, declarer play for the minimal number of hands would be fine. Quote Link to comment Share on other sites More sharing options...
bluecalm Posted February 24, 2011 Author Report Share Posted February 24, 2011 Standard dev is about 3.5imp/hand so for 250 hands sample it's 55imps (for play after 1st lead is made)For 1000 hands sample it's 110imps.For bidding (including 1st lead) it's 6imps/hand. As you can see variance is huge on those and hand samples from vugraph not that significant. It's still fun to see famous names winning on this one though ;) 1 Quote Link to comment Share on other sites More sharing options...
Fluffy Posted February 24, 2011 Report Share Posted February 24, 2011 Comparing your two declarer play list, it looks like Wladow-Elinescu gain a HUDGE edge on the opponents' opening leads? Btw, it would be great if you could give error estimates for your numbers. Doesn't have to be precise, e.g. just the variance for, say, declarer play for the minimal number of hands would be fine.I though the same, probably because they give away less info in the bidding? The fact that they win IMPs vs the average number of tricks taken doesn't mean that the contract they are playing is the best. they could win a lot of IMPs for saving vulnerable undertricks on a 4-1 fit althou I doubt this is the case. Quote Link to comment Share on other sites More sharing options...
kayin801 Posted February 24, 2011 Report Share Posted February 24, 2011 Can you clarify my understanding on this: On your list which shows how declarers do compared to double dummy after the opening lead, everyone is at least slightly in the negative in imps/hand. I find that to be pretty interesting since I would expect the defenders to still have an impact on the outcome of the hand. Which of these conclusions can I draw from this study? 1) After the opening lead, it's easier for the defenders to find a good defense than for declarer to find a good line.2) Declarer is in significantly more control than the defenders after the opening lead and the negative imps reflects his/her inability to see all 52 cards and the necessity to guess sometimes.3) The relative scoring for contracts adjusts those imps lost significantly. For example, a declarer not following the double dummy line or taking a safe line will frequently just miss an overtrick (missing a gain of 1 imp), while an error costing someone a contract or a vulnerable (doubled) undertrick usually costs more than an imp. Edit: though with #3, couldn't a defender just as easily err and allow a contract to make or chuck that undertrick? Based on that chart and the initial totals, can we draw any conclusions about the relative strength of the players' competition? Quote Link to comment Share on other sites More sharing options...
bluecalm Posted February 24, 2011 Author Report Share Posted February 24, 2011 1) After the opening lead, it's easier for the defenders to find a good defense than for declarer to find a good line. This in my opinion is true. 2) Declarer is in significantly more control than the defenders after the opening lead and the negative imps reflects his/her inability to see all 52 cards and the necessity to guess sometimes. I don't know how "control" is defined so I can't answer but basically defenders advantage comes from the fact that declarer needs to find all the missing honors and naturally often fails while very often next decision for defenders comes significantly later after opening lead and it's easier for them to get that right.Often defenders just wait and see with his Qxx of trumps but declarer will miss half the time... Comparing your two declarer play list, it looks like Wladow-Elinescu gain a HUDGE edge on the opponents' opening leads? My guess is that they just got lucky on small sample of vugraph hands. I don't know anything about their style but it looks like the most likely explanation to me. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.