The Tao of Gaming

Boardgames and lesser pursuits

Recent Bridge-ness

With the local sectional happening this weekend (and Veteran’s day as a holiday) I got in 4 games of club & sectional bridge during the week. I think I’m finally clear of the rust I had, and kind of a middle-of-the-pack flight B player. I make some poor judgement calls and could use more polish on my card play, but my mistakes are beyond what flight C players make, and I sometimes play quite well.

It doesn’t hurt that I’ve got a partnership with another solid-B player who thinks roughly the same way about bidding as I do.  We’ve had 1-2 undiscussed auctions pop up in a session and we’re invariably on the same wavelength. Again, now all I need is better judgement and technique. (We’re slowly working through the grey areas of system).

I don’t have any particular examples of great plays or boneheaded ones (which isn’t to say they don’t exist, just that they aren’t particularly blog-worthy), but in general I’m quite pleased with this recent turn of events.

Advertisements

Written by taogaming

November 14, 2009 at 2:10 pm

Posted in Bridge

17 Responses

Subscribe to comments with RSS.

  1. What ACBL district do you play in? You can check out an attempt at a realistic bridge ranking system here:

    http://www.coloradospringsbridge.com/pr.htm

    It only measures performance in MP events in the past 2 years so improvement can really change your rating.

    Eugene

    November 14, 2009 at 3:45 pm

  2. Interesting. I couldn’t find a methodology, which I’d like to see… according to the power ratings for my unit, I’m #67 out of 167. My most frequent partner is mid-90s, my partner for this sectional was around 40.

    By the averages, it rates me as a Bronze Life Master (500-1000 MPs), which is low to middle Flight B.

    taogaming

    November 14, 2009 at 5:30 pm

      • Well, that’s slightly below how you’re rating yourself, but remember, it takes in performance over last 2 years. Since you think you’ve been improving lately, it looks like a reasonable approximation. Note that :

        http://www.coloradospringsbridge.com/PR_FILES/WEBPAGES/U172WEB/U172PR2C.HTM

        lists the comparable rank for MP. You’re #329 out of #543 in your unit (with 213 masterpoints). Which do you think did a better job in assessing your current skill?

        If you’d like to help make these ratings more accurate, have your local clubs e-mail ckchampion@netzero.net with their ACBLscore files from the past 2 years and you’ll get many more rated players. (El Paso in your area and Santa Clara Valley in my area are already doing it.)

        Eugene

        November 15, 2009 at 2:45 am

      • Oh the Power Rating is reasonably accurate; the discrepancy could just be arrogance on my part (or the fact that I was pretty rusty for the first 1/3rd of my playing time). Pretty much everyone recognizes that MP measure a lot more than skill (which makes their skill usefulness a lot less). I get lots of offers for “Pro-Am” events. Anyway, this gives me something to watch, and I’ll likely mention the site to the two local club owners.

        taogaming

        November 15, 2009 at 10:22 am

  3. It might be reasonable for you, Brian, though I have no idea, but it’s not even remotely close to accurate near the top. It’s ridiculously off for both pairs and individuals, at least as an overall rating. Even considering that it only applies to matchpoint pairs, it’s still absurd. Take it with a huge grain of salt, or better yet, just ignore it. I think masterpoints does a far better job as a rating system than that does. And I think the ACBL masterpoint system is pretty silly.

    JeffG

    November 15, 2009 at 3:37 pm

  4. How is it ridiculously off? Many great players (Rodwell, Cohen/Berkowitz, Martel) are not there because they haven’t played enough matchpoint pairs games outside their main partnership to qualify. And some “great players” are coasting on their reputation. And I think this system is far better at noticing recent improvement than masterpoints.

    I think most people lack the scope to properly assess skill on a national level, simply because they don’t play often enough against the whole crowd. But, I think they can get a pretty good idea of the relative skill of the players in their area. Jeff, I notice your unit doesn’t have many rated players — maybe you should arrange for the LA clubs to send him their ACBLscore files. Your area may feel “better” once you get more data.

    Eugene

    November 16, 2009 at 2:36 am

    • Look at the top 10 players. At least half of them do not belong in the top 100. I’ve never heard of three of them.

      Look at the top 10 pairs. Three of them are clearly among the best. Three more can be argued are near the best. Maybe. Look at the next 10. Two are clearly top pairs. There are a few other quite good ones, but most of those guys, even if you asked them, would laugh at being claimed among the top 20 pairs in the country. It’s just clearly not true.

      Also, look at the raw numbers. Meckwell averages 69%? In what, a club game? Even they can’t average 69%. It’s just not possible. No one has half their games at 70%. Even Crane didn’t do that. How many 75% games does it take to recover from the annual 43% disaster night?

      Yes, there are obviously missing players: Steve Weinstein, Martel, Woolsey. No, I see Steve: #235. Yeah, right. Speaking of Steve…looking for Robinson. Close. Craig Robinson is at #53. Amusing coincidence.

      It’s true that “most people lack the scope to properly assess skill on a national level.” Of course, most people don’t play bridge. Let’s accept the possibility that I don’t know the folks on the East Coast. It’s not true, but sure. Let’s just list the folks who live in California in descending order.

      1. Mark Itabashi
      2. Chris Larsen
      3. Jim Munday
      4. Lew Stansby
      5. Ron Smith
      6. John Jones
      7. Ifti Baqui
      8. Mark Ralph
      9. Bob Bratcher
      10. Mitch Dunitz
      11. Stephen Goldstein
      12. Mark Perlmutter
      13. John Swanson
      14. Howard Einberg
      15. Bob Kerr
      16. Bill Schreiber
      17. Rick Roeder
      18. Marshall Miles
      19. JoAnna Stansby
      20. Jordon Choderow
      21. Grant Vance

      OK, I’m tired of digging through it. I’m sure I missed a few, because I probably have never heard of them. Let’s just go with this list. Can you argue with a straight face that those are roughly the 21 best players in California? Of course not. Continued in another post; it’s too irritating to have to scroll to the bottom every few seconds.

      JeffG

      November 17, 2009 at 4:30 pm

      • The list is not meant to measure the best 21 players in California. The list is meant to measure the players who have performed the best in matchpoint events in the last two years, who we have enough data on. Obviously there are some big misses (Martel, Woolsey, Moss are obvious names that come to mind) but that’s not because the system thinks they are poor players — it is because they do not have enough data for an accurate rating. If you look at the provisional rating for some of the great missing players, you’ll see they would also be on the list. Also, remember that performance in team games / BAM is not factored in. It’s strictly the subset of pairs games.

        Meckwell is rated 69 because they would average 69 in a field full of 25 PR (average) players. Average = around 600 MP, a Flight B field. You know what, I think they very well might be able to do that! They already average 58% in NABC-level fields.

        Eugene Hung

        November 17, 2009 at 4:50 pm

      • Here are the top in California by lifetime monsterpoints.

        1. Itabashi (one match!)
        2. Weischel
        3. Simpson
        4. Shuman
        5. Ferguson (probably a GNT screwiness; he lives in Idaho)
        6. Mary Jane
        7. Smith
        8. Lew Stansby
        9. Kyle Larsen
        10. Hamish Bennett
        11. Alan Bell
        12. Woolsey
        13. Chip Martel
        14. Steve Gross
        15. Ronnie McMurdie
        16. Rhoda Walsh
        17. Brenda Keller
        18. Jill Meyers
        19. Rose Meltzer
        20. Beverly Rosenberg
        21. Harvey Brody

        That’s not a list of the top 21 players in California, either, but it’s a whole lot closer than the previous list. And that’s lifetime. You want to measure recent? How about this year? Let’s look at the player of the year race.

        1. Brad Moss
        2. Rose Meltzer
        3. Kyle Larsen
        4. Jill Meyers
        5. Lou Ann O’Rourke
        6. Chip Martel
        7. Lew Stansby
        8. Kit Woolsey
        9. Joanna Stansby

        Player of the year only has 9 Califorians. Excluding the two clients, and I’ll buy that list as being reasonable. Imperfect, but not bad.

        How about overall monsterpoints for 2009?

        1. Itabashi
        2. Ferguson
        3. Bennett
        4. Ellen Anten
        5. Simpson
        6. Smith
        7. Kyle
        8. Don Nemiro
        9. Looby
        10. Marjorie Michelin
        11. Bruce Noda
        12. Keller
        13. Ahmed Sorathia
        14. Rose Meltzer
        15. Becky Clough
        16. Greg House
        17. Bruce Horiguchi
        18. Steven Love
        19. Gross
        20. Diane Shannon
        21. Roger Clough

        I’ll buy that that list isn’t as good as the power ratings one.

        Player of the decade is probably pretty good. Let’s see.

        1. Lew Stansby
        2. Chip Martel
        3. Rose Meltzer
        4. Kyle Larsen
        5. Brad Moss
        6. Peter Weischel
        7. Kit Woolsey
        8. Jill Meyers
        9. Jo Anna Stansby
        10. Ron Smith
        11. Lou Ann O’Rourke
        12. Rita Shugart
        13. Renee Mancuso
        14. Reese Milner
        15. Ed Wojewoda
        16. Pam Wittes
        17. Itabashi
        18. Ferguson
        19. Dunitz
        20. Wafik Abdou
        21. Farid Assemi
        Just missing: Billy Eisenberg, Billy Cohen, Simpson, Chris Larsen. Since there are four clients on the list above, let’s plug those guys in instead. It’s not quite fair; the list emphasizes women and hurts young players (Jason Feldman is probably on the list if we take only the last 3 or 4 years), but that list isn’t bad at all. I wouldn’t argue emphatically that those are the 21 best players in California, but it’s not too bad, particularly given the three biases (clients, women’s events, and youth).

        So bottom line is that power rankings are roughly as good as yearly masterpoints in evaluating overall rankings, maybe a little worse. They are considerably worse than platinum masterpoints, which look to me pretty darn good; if there were a way to correct for the obvious biases, we’d have something.

        JeffG

        November 17, 2009 at 4:53 pm

  5. I agree something like player of the decade is far more appropriate for measuring top player performance, because it’s over a longer period of time and encapsulates all types of events, including KOs and BAM. But, you’re trying to apply power ratings to something it’s not meant to measure. It’s measuring recent matchpoint pairs performance, and that’s something that top players don’t really need to worry about as much — they already have the reputation and play mostly in KO/BAM. And look at the players on the POY list:

    1. Stansby : he’s on the power ratings list at #4. He’s still going strong, but much of his great performance in the past 2 years has come in BAM or KOs, not pairs, so he’s not as high as he could be.

    2. Martel — not rated by Power Ratings yet, not enough data. http://www.coloradospringsbridge.com/PR_FILES/WEBPAGES/U505WEB/S4174844.HTM shows he has never played with Stansby in a pairs game in the last 2 years (not even the Blue Ribbons!), and he has only played 6 with Zia and 4 with Fleisher. Definitely not enough data.

    3. Meltzer — a client, a clear problem with masterpoints — not rated due to lack of pairs games.

    4. K Larsen — not rated, only 3 mediocre games http://www.coloradospringsbridge.com/PR_FILES/WEBPAGES/U505WEB/S4174844.HTM

    5. Moss — not rated

    6. Weischel — not rated

    7. Woolsey — not rated, does play a lot of pairs games, but doesn’t play with enough people besides Stewart : http://www.coloradospringsbridge.com/PR_FILES/WEBPAGES/U497WEB/S2842084.HTM

    8. Meyers — not rated

    9. JoAnna Stansby — rated, and also on the power rankings at #19

    10. Ron Smith — rated, and also on the power rankings at #5.

    11. O’Rourke — client, and not rated

    12. Shugart — client, and not rated

    13. Mancuso — not rated

    14. Milner — not rated

    15. Wojewoda — not rated

    16. Wittes — not rated

    17. Itabashi — rated #1, and I always thought matchpoints suited him far better than other forms of scoring

    18. Ferguson — not rated

    19. Dunitz — rated #10

    20. Abdou — rated 30.85 (Grand Life Master)

    —————

    As you can see most of these players are not underrated, there is simply not enough data to rate them. I think most top players actually don’t play much in pairs games, but for those of us in the rank and file, we do, and I think these ratings provide better insight into how we’ve been performing lately.

    Eugene Hung

    November 17, 2009 at 5:17 pm

    • I won’t argue that Power Ratings are a reasonable attempt to measure recent matchpoint performance of players with what data is available to the inventor. Whether it solves the problems of removing the impact of partnerships, I don’t really know, but it doesn’t look like it to me. But that’s just a guess. I do, however, claim that it is not particularly useful as a rating system. Mark Itabashi, no matter what his strengths, cannot be seriously considered as the third best player in the US, nor is he the best player in California, and any rating system which claims he is clearly severely flawed. Are the results it gets somewhat correlated with bridge skill? Sure. Itabashi is certainly a very good player. Jeff Meckstroth, who is rated #1 by power rankings, might well be the best player in the US.

      Basically, I think the proof is in the pudding. Its ranking of top players is simply wrong. No, I don’t know what is exactly right, but anyone can see that its results are not even moderately close. If you took a poll of everyone who has won a match in a Spingold or Vanderbilt in the last 5 years (is that a reasonable cross-country sampling?), asking them to name the other top 25 bridge players in the US, six of the top 11 power ranked players would get zero votes. (I say “other,” because we’d have to ask them for 26, right? Howard Schenken won a poll 50ish years ago. The question was, “who is the second best bridge player in the US?”) If popular conception and the ranking system are that far out of tune, either folks are badly misinformed (possible—Derek Jeter has won three Gold Gloves), or the ranking system is way off. My money is on the latter.

      JeffG

      November 18, 2009 at 8:19 pm

  6. […] a comment » After commenting that “my partner and I were on the same wavelength” I suppose this was inevitable. During a team game last night, LHO opened 2H (weak-ish) […]

  7. You’re trying to attribute my position as : “This ranking system determines who the very best players are.” By no means! When most of the best players don’t qualify for being rated, how can it possibly get a good fix on those players? I certainly agree Itabashi is not the best player in California. I am not arguing that he is, although I do think he does better in matchpoint events given his skillset.

    Most people’s idea of what makes a great bridge player is skewed towards knockout teams performance, and none of one’s achievements in that area matter here. Your argument is like saying the Gold Gloves are a bad system because the greatest hitters aren’t on there. But that’s the point, the Gold Gloves are designed to recognize recent defensive prowess, not all-around prowess or historical prowess. (And yes, I know that the Gold Gloves are a crappy award.)

    What the power ratings are saying that, given the limited pool of players that we *can* rank, Itabashi’s 2 year matchpoint pair performance has not been bettered within California. The small sample size doesn’t mean he IS the best — but he’s likely a very good player. For the most part, I don’t see top players rated as average players, nor do I see average players rated as great players.

    Eugene

    November 18, 2009 at 11:35 pm

    • I see above average players rated as great players routinely. I don’t see any average ratings, so I can’t argue about them.

      Your statement with which I’ve been disagreeing is one of your first: “You can check out an attempt at a realistic bridge ranking system here:” Power Ratings (PR) is not a realistic bridge ranking system. It is an attempt to measure recent matchpoint performance. If one is looking for a realistic bridge ranking system, one needs to look elsewhere.

      Is it successful at measuring recent matchpoint performance? The hard parts of doing that are (a) isolating player performance from that of the partnership, and (b) equalizing playing fields. We aren’t told the approach taken to accomplish these goals, but from the results, it seems pretty clear that it fails. Why? Wrto (a), the partnership rankings are much more reasonable than the individual rankings, which are laughable. That seems to me to imply that the player isolation method fails. (b) is much harder to tell. One quick and dirty approach is to assume that the fields in the open national matchpoint pair events are the strongest in the country by a fair bit. (With the exception of the Wehner, which I have omitted, this is pretty much true.) So the pairs who have done the best in national real pair events in the last two years should be pretty close to the top, or at least be reasonably correlated with the pair list, right? They aren’t. Totally unsurprisingly, the pair who has done the best in national pair events over the last two years is Berkowitz/Cohen. I don’t have easy access to percentages, so I’m going to just add up monsterpoints won, which isn’t a great answer, but the results are interesting. Here they are:

      323.05 David Berkowitz – Larry Cohen, Boca Raton FL
      291.66 Eric Greco, Philadelphia PA; Geoff Hampson, Las Vegas NV
      261.25 Eric Rodwell, Clearwater Bch FL; Jeff Meckstroth, Tampa FL
      241.94 Douglas Doub, W Hartford CT; Adam Wildavsky, New York NY
      191.88 John Diamond, Boca Raton FL; Brian Platnick, Evanston IL
      190 Stephen Landen, Ellicott City MD; Pratap Rajadhyaksha, Sarasota FL
      190 Michal Kwiecien, Lublin 20-732 Poland; Roald Ramer, Amsterdam 1071T Netherlands
      190 Jan Jansma, Malden Netherlands; Ricco Van Prooijen, Nicuw-Vennep Netherlands
      174.69 Ralph Katz, Hinsdale IL; Nikolay Demirev, Arlington Hts IL
      165.77 Martin De Knijff, Las Vegas NV; Frederic Wrang, Stockholm 1 Sweden
      159.41 Robert Lebi, Toronto ON; Dan Jacob, Vancouver BC
      142.5 Nikolay Demirev, Arlington Hts IL; Hemant Lall, Dallas TX
      142.5 Ernesto Muzzio – Alejandro Bianchedi, Buenos Aires Argentina
      141.35 Ralph Buchalter, South Orange NJ; Migry Zur Campanile, East Palo Alto CA
      130.11 Michael Prahin, Irvington NY; Alex Perlin, Metuchen NJ
      127.82 Chris Larsen, Costa Mesa CA; Wafik Abdou, Bakersfield CA
      125 Zia Mahmood – Bjorn Fallenius, New York NY
      109.7 Daniel Korbel, Waterloo ON; Jonathan Steinberg, Toronto ON
      109.28 Glenn Milgrim, Forest Hills NY; Robert Levin, Bronx NY
      108.68 Jerry Helms – Robert Bitterman, Charlotte NC
      106.88 Tom Townsend, London W9 1sd England; David Gold, Reading R Great Britain
      106.88 Ishmael Delmonte, Double Bay NSW Australia; Vincent Demuy, Laval QC
      100 Dick Bruno, Des Plaines IL; Peggy Kaplan, Minnetonka MN

      These data shouldn’t surprise anyone. (With the possible exception that Levin/Weinstein aren’t in the list, I suppose…but they haven’t played national pair events together in the last two years.) If I were to announce that the results of my careful system figuring out who were the most successful pairs in matchpoint events over the last two years were the above, would that seem reasonable? You bet it would. Most of those pairs are long-standing well-known pairs. Berkowitz/Cohen is the most successful matchpoint pair in the world; it makes complete sense that they are on top. Yeah, because this is only two years and 12 events, we see a few pairs there who just won one event, so the method isn’t very sound overall, but it did a heck of a lot better job than pair PR. So I conclude that PR doesn’t do a decent job of isolating context. Yeah, Itabashi wins a zillion sectional pair events against unbelievably weak opposition. He piles up 68% games in those events all the time. So what?

      Overall, I think we can conclude that not only is PR not a realistic bridge ranking system, it doesn’t even do a good job of measuring recent matchpoint performance. It’s cool that the inventor is trying; props to him for his effort. But the results, at least so far, don’t seem very accurate.

      JeffG

      November 19, 2009 at 3:33 pm

      • Whoops…I missed a line. Townsend/Gold get 20 more MPs. It’d be nice if there were a way to edit.

        JeffG

        November 19, 2009 at 3:38 pm

      • The approach taken is actually detailed on the website. I do think that the degree of difficulty seems to be off slightly — he thinks a NABC pairs is 10% harder than the average game. But, it’s possible the system is wrong due to lack of data from clubs.

        I definitely would put Berkowitz/Cohen or Meckwell (they don’t play often in pair events, but I imagine they might dominate if they did) as my personal top 2. You can see the stats he’s compiled for them here:

        http://www.coloradospringsbridge.com/PR_FILES/WEBPAGES/U128WEB/S2076578.HTM

        http://www.coloradospringsbridge.com/PR_FILES/WEBPAGES/U128WEB/S5482658.HTM

        Berkohen is averaging 55% in 11-strength fields, Meckwell is averaging 57% in similar fields. So Meckwell is rated higher by PR.

        Eugene

        November 19, 2009 at 3:41 pm


Comments are closed.

%d bloggers like this: