Word Freak : Heartbreak, Triumph, Genius, and Obsession in the World of Competitive Scrabble Players (9780547524313)

by Fatsis, Stefan

  The Sherman house is a brick two-family on Laconia Avenue, with a pink awning, a chain-link fence, and a decrepit yew tree out front. A sign on the door says, MISSIONARIES AND OTHER SALESPEOPLE PLEASE USE ENTRANCE IN NEW JERSEY. Joel has always lived at home. His mother, Gertrude, worked in the family court system, then the registrar’s office at Lehman College, then in the Bronx district attorney’s office as a trial separation assistant. She died of pancreatic cancer in 1988. Joel’s father, Mike, is a retired accountant for the city of New York. His brother, Larry, ten years Joel’s senior, went to college for three years and works as a freelance proofreader. Larry also lives at home—in the basement with his two cats, who aren’t allowed upstairs because of Joel’s allergies—though he has lived on his own in Manhattan and nearby in the Bronx. Larry is an expert-level Scrabble player too, but suffers no debilitating ailments like Joel. Neither brother has ever married, and Joel rarely if ever dates, a subject, like all others, that he will discuss gladly.

  Joel was six years old when he began playing Scrabble with his family. He was fascinated by the idea that a combination of letters could make more than one word, and that he could find them. It was also a tolerable way of spending time with his mother, whom he describes as difficult and bitter. “Through most of the last ten years of her life I played hours and hours of Scrabble with her,” Joel says. “Basically it was the only thing I enjoyed doing with her.”

  In high school, Joel played chess constantly, but never really liked the game; it was just a way of killing time. His peak rating was 2051, expert level but “squat on a scale where a master is 2200 and up, senior master is 2400 and up, and grand masters are 2600 and up. I had no chance of achieving a draw against any of those, that’s how bad I was.” In his twenties, when he worked downtown, Joel would stop at the Manhattan Chess Club on the tenth floor of the Carnegie Hall Studios most nights because he hated riding the crowded rush-hour subway home. Half the time he would read the paper, schmooze, or even play an occasional game of Scrabble, though only one of the chess masters, Asa Hoffmann, a hustler who had played hundreds of games against Bobby Fischer, was also a bona fide Scrabble player.

  By then, Joel had played in a Scrabble club—in Phoenix, while visiting his parents, who had moved there briefly—and won most of his games against, he was told, expert players. When he tired of playing chess for money against Hoffmann, they switched to Scrabble. Asa crushed him, but Joel gradually improved. “When I beat him badly two sessions in a row I finally said, ‘Okay, it’s time to look up the New York club,’” Joel says.

  Joel inhaled the two- and three-letter words, and received a 1570 rating after his first tournament. He qualified for the expert division in the 1989 Nationals in New York, finished 14–13, and his rating leaped to 1774. When his roiling gut forced him to quit working the next year, Joel turned to Scrabble. He really didn’t need to work. There were no mortgage payments on his father’s house, and a small inheritance from his mother allowed him, his father, and his brother to live comfortably. So Scrabble would be Joel’s focus. Why not? “There basically wasn’t anything else I was going to do in this life that well.”

  Joel leaves home a couple of times a week, usually for errands in the neighborhood or to direct the Manhattan Scrabble Club on Thursday nights. The rest of the time? “Sleeping, eating, playing Scrabble on-line, wasting time, procrastinating,” Joel tells me.

  He works at a computer (Ashley Judd is the screen saver) on a fold-down desk next to the kitchen. What used to be the dining room table is covered with unopened junk mail, stacks of paper, floppy disks, and tattered Scrabble dictionaries. There is an old portable radio, rolls of pennies, bags of plastic bags, and sheaths of takeout menus in the old, cluttered kitchen. (Except for breakfast, the Shermans have most of their meals delivered.) In the similarly cluttered living room, Joel shows me a charcoal drawing of him done by a woman at the Manhattan Chess Club and a sketch of his father made in the Philippines during World War II; they look alike—big ears, wispy hair, a large nose that makes an almost ninety-degree turn at the bridge.

  Upstairs, Mike Sherman is sitting in his bedroom at a card table with his back facing the window, an aqua transistor radio and a bag of lollipops close at hand. He is playing Master Monty, the primitive electronic Scrabble game that dates to the early 1980s. On an old cardboard set so worn that he has had to pencil in the values of some of the premium squares, Mike plays a half dozen games a day by himself. He punches his word into Monty, which three minutes later spits out its counterplay and updates Mike’s rack. Mike occasionally joins his sons at tournaments. It keeps him active. Mike is eighty-five years old.

  Joel hands Mike his hearing aid. Mike cleans off the wax.

  “I just want you to know that the house looked different when my wife was alive,” Mike says in his gravelly voice. “I do what I can. I clean some. But I can’t cope with that mess. With all that paper lying around.”

  Mike occasionally pressures his sons to get nine-to-five work, but seems resigned to their choice. “I worked for sixty years,” he notes. “That didn’t come from me.”

  Joel and I head back downstairs, and I ask him about his “job,” full-time game player, which is how he asked to be described in press material at the 1997 Worlds. Surely Joel studies all the time, I had thought. But he has no collection of word books, or any sort of books, for that matter. “I don’t read,” he tells me back at his computer. “The only reading I do is on this damned screen. Because looking at printed matter puts me to sleep. It’s been doing this for years and years. If I open a book and sit there and try to read it, my eyes shut on me.”

  Even if it’s a word-list book?

  “Faster.”

  The words were never enough. Joel slogged through the early stages of studying—the pages and pages of high-probability bingos in Mike Baron’s book. Joel hated memorizing, but he was a natural anagrammer. He could just see the words. When he got bored reading word lists, a computer program called Lexability revived his interest. For a couple of years, he studied on the computer for an hour or two a day, four days a week.

  “I would put it at five-tenths rack preparation, two-tenths board awareness, and three-tenths board preparation,” he says of the game. “And what I mean by board preparation is making sure that the board is always amenable to my having a good play.”

  “It would seem fundamental, but not everybody can do it,” I say. “What’s the difference?”

  “Possibly my concentration on those factors when other people are concentrating on word finding. To me the word finding is the easy part. My brain just does it.”

  “And the memory part?”

  “That’s the hardest part.”

  I laugh.

  “I just don’t feel like I retain well,” Joel insists.

  “You must, though. You’ve memorized eighty, ninety thousand words.”

  “I don’t exaggerate my shortcomings. I’m painfully aware of them.”

  It’s true, Joel’s strength doesn’t derive from a love of the words. (It’s also true that he doesn’t exaggerate his shortcomings.) He wasn’t an innovator. He didn’t create his own word lists. “I’m an opportunist,” he says. “I piggybacked on the work other people did. I don’t deny that for a moment.” But he did have an affinity for patterns, and word patterns held more meaning for him than chess patterns. Joel can rattle off words of Greek origin which end in OS and pluralize with OI instead of I. He knows which words ending in US take an I plural and which take II, and which ones ending in UM pluralize with A. He knows whether words are botanical or anatomical, minerals or mathematical terms.

  But the strategic conflicts of the game are what drive him. Joel spends hours, usually after midnight, analyzing board positions and evaluating play sequences, and then posting lengthy analyses on CGP about specific racks. Here’s one example. Your opponent opens the game by playing the word FOU (an adjective meaning “drunk”). A question is posed: From a rack of EOOPVWZ, what do you do
? Play VOW above FOU at 7F, forming OF and WO, or POOVE (an offensive term for a gay man) below it at 9I, forming UP? Here are the choices:

  or

  Via its simulation feature, which assesses the strength of plays, the computer program Maven determines that VOW is superior. Joel disagrees, in a manner that demonstrates the depths to which a few letters on a Scrabble board can be analyzed:

  POOVE and VOW score equally. POOVE draws two more tiles with plenty of valuable things in the pool. POOVE does not leave hooks on easily lucrative lines that accept tiles other than S. I is a very likely draw, so at the very least, the WZ leave from POOVE threatens WIZ 8M for 50 points. The V on column L means opponent is not very likely to take that spot across row 8 for a three-tile play of his own. Opponent might play thru the E of POOVE but if he does, ZOWIE K8 is still a minimum 34 points for a highly likely IE draw out of five chances with 9 I’s and 11 E’s unseen. Worst case scenario is opponent plays thru the V, and somehow fails to give back any big Z play, and that’s really hard to do. While EPOZ is clearly a superior leave to WZ, on a board that requires hooks on every line that are not part of this leave, drawing three tiles instead of five simply is not supportable.

  Not everyone agrees, and that’s classic Joel. Other top experts routinely bash his analysis, in this case for foolishly opening the board with POOVE and for keeping the rancid W.

  “Joel Sherman sees WIZ and ZOWIE and says, ‘Cool, I have synergy,’” Bob Felt rebuts. “But where are we going to play these nonbingos? Play off WZ next turn, and guess what? After three turns our rack is a random draw—we have done nothing toward building a coherent rack.... One of my advantages as an expert is I understand how to build racks. The notion of throwing away that advantage seems bizarre to me.”

  Jim Geary, whose rating is on par with Joel’s, e-mails me privately. “Joel’s latest piece of shit is so misguided I want to ralph.” It underscores, Geary says, one of Joel’s basic theoretical shortcomings: He worries too much about things not worth worrying about. In the case of VOW versus POOVE, Joel worries that his opponent will take advantage of hooks on the back of FOU—FOUL, FOUR—or the front of VOW—AVOW—that he, Joel, doesn’t possess. But he could get them easily himself because of their frequency in the bag.

  Geary addresses another position deconstructed by Joel. A 1300-level player named Scott posted to CGP that he held AINNRRT, a bingo-prone rack except for the duplicated N and R. Scott asked what the experts would do. He realized that he was five-sixths of the way to RETINA, the third-most fruitful six-letter bingo stem, with an open board early in the game. So his inclination was to make a small play, such as RAN, which would have left him with INRT, or to exchange an N and an R, leaving the even better AINRT. There are lots of E’s in the bag, which is good for Scott.

  Joel pooh-poohs the idea. He says that Scott needs to be concerned about the possibility that his opponent could hook a Y onto the word WITCH already on the board. WITCHY would net at least 38 points, because the Y would land on a triple-letter score that would count in two directions. “This enormous hot spot indicates turnover should be of great concern to you,” Joel writes. “If you don’t have a reasonable way to block that spot, and you don’t, you have to beat your opponent to the Y.” Joel suggests playing TRAIN or RIANT, improving Scott’s odds of drawing the Y by playing off more tiles.

  “Wrong. Wrong. Wrong,” Geary writes. “The most important factor, which 1300 Scott correctly guessed, is that he has TRAIN on his rack with a bunch of E’s left. How many ways is Joel an idiot? Let us count.” Jim counts to nine, starting with the fact that the Y hook is not that big a deal because there are only two Y’s in a bag that now contains seventy-three tiles, and it’s not that hot a spot anyway because the theoretical average play is 35 points, compared to WITCHY’s minimum of 38. Trying to block the spot is wrong.

  “Not only are we screwing ourselves to turn over good tiles for just two tiles in the bag that could even be used in this non-hot spot that’s not going to make much of a difference in our lives if we do in fact miraculously hit it, Joel has us wasting turns to block opponent from doing same,” Geary writes.

  Joel’s obsession with the potential Y hook has him eschewing a basic goal: building a bingo rack from great tiles, especially early in the game, when taking a lead can be decisive.

  “Fishing for bingos from high-probability racks is often correct,” Geary goes on. “Fishing for Y’s from high-probability racks is wrong on two counts. One, the Y ain’t that big a deal AND it’s hard to get. Two, that’s exactly the wrong thing to be doing with this potentially nice rack. And, most important, when you get five great synergistic letters together, you don’t want to burn them. Especially when the missing piece is one of 11(!) unseen E’s. Joel wants to shatter this combination in which even weak players see the potential energy.

  “In summary,” Jim writes, “Joel Sherman has very many bad ideas about Scrabble. If you write a chapter entitled Joel Sherman: Man or Myth?, please feel free to use this analysis in its entirety. Or maybe call it Jim Geary: Backbiting Also-Ran. Who knows which is right?”

  G.I. Joel’s peers can question his ability, world championship or not, and laugh at his physical tics, but it’s hard not to admire his quirky devotion. Other than Joe Edley, who works for the Scrabble hierarchy, Joel is the only player who devotes his life to the game. More than Edley or Cappelletto or Butts—whom I believe Joel will come to resemble physically later in life—Joel is Mr. Scrabble.

  Joel once played hundreds of games against Maven in which he allowed the computer to select his moves for him in an effort to determine whether Maven “cheats” in its tile assignments (by giving itself better letters, as some conspiracy theorists suspect). When Ron and Susi Tiekert moved to Atlanta, Joel took over the Manhattan club, and he e-mails thorough weekly summaries to members. His posttournament musings often include detailed hotel and restaurant reviews, like this one, about a porridge he ate in Singapore: “Whatever the heck it was, I liked it enormously, it was a very satisfying and nourishing breakfast, plainly and cheaply produced yet flavorful, and left my tummy quiet as a mouse, at least until lunchtime.” How can you not love that?

  Joel has won more than forty tournaments in his career. In the previous year, he lost about $2,000 playing Scrabble, meaning that after paying for entry fees, transportation, hotels, and food, he was two grand in the red. The year before that, when he won the Worlds, Joel netted $27,300. Despite the meager wages, when someone asks Joel what he does for a living, he answers, “professional Scrabble player.” (Not that many people do ask. Joel rarely encounters anyone who isn’t a Scrabble insider.)

  “The point is to be good enough at it that I can say seriously to somebody that I make a living at it,” Joel tells me. “And to get the game promoted enough that that can be a real possibility. It’s very frustrating to me that we have not yet managed to develop an audience for the game. I don’t mean participants. I mean literally people watching. People could watch Scrabble a lot more often than they would watch chess. Scrabble has more mainstream appeal. It is more easily understood than chess.”

  “Watch what?” I ask him. We’ve just played a game against a computer robot called ACBot over the Internet. “Watch you play TREHALA and then run to the dictionary?”

  “I’d sooner have them watch me make a five-letter play that has four overlaps,” Joel says. “The point is to see us use all of what we know and be inspired by that to use more of what they know. That is, the spectator should be inspired to use more of what the spectator knows.”

  “Which is?”

  “The maximization of resources. That’s why a five-letter word with the four overlaps is the interesting thing to see. Because that’s not necessarily going to involve something not everyone knows. It’s just going to involve using what you know and using all of what’s in front of you.”

  That may be one of the game’s intellectual beauties, but it doesn’t bring Joel’s dream any closer to reality. Hell
, I’d like to travel around playing Scrabble for a living, too, and have people watch me do it; that might even motivate me to take what little innate talent some experts sometimes tell me I seem to have (could I qualify that any more?) and develop it faster. Frustrated Scrabble experts love to compare their game to chess, the national organization of which boasts eight times as many members as the NSA. As Joel posted to CGP:

  I just don’t see why Garry Kasparov can make millions playing chess, and perhaps a dozen or two other grand masters can earn respectable livings as professionals at their game of choice and the same can’t happen in Scrabble. Those respectable livings are available because the game’s mass appeal is exploited by manufacturers and publishers of accessories and ancillary products used by hundreds of thousands or even millions of enthusiastic if far less talented players. This has happened with golf, tennis, bowling, and chess. The result is that the burden of supporting these professionals is spread out over so many consumers that it is barely noticeable to the individual consumer.

  But chess was a public phenomenon exactly once, when Fischer played Spassky. And the Scrabble played by Joel, even by me, is so radically different from what’s played at home as to render it even less accessible to the casual player than top-level chess. It’s a circus act—all those words! And it’s controlled by a single company (responsible for dozens of other products) which does very well, thank you, selling a million or two sets a year without spending hundreds of thousands more promoting a pro Scrabble tour that no one would watch. Hasbro isn’t likely to lend out its sacred trademark for sponsorship by other companies, as some players want. The company is more likely to license a World Championship Wrestling version of the game. The expert players just don’t matter that much.