| Player 1 |
|
#character-encoding UTF-8
#player1 PH Paul Holser
#player2 JJB John J. Bulten
>PH: AEFHIOW 8H WHOA +20 20
#note 1:59 [23:01] (howf 8g 26 +4.6) After PH led the round robin 6-1 and JJB swept the swiss 4-0, with each player finding the other half unduly challenging, they shake hands and prepare for a best-man-wins final game, recognizing that the loser will likely take third place behind Asif Ali, and a significantly smaller prize. All systems are go for a dramatic finish. JJB offers PH the shuffle, briefly forgetting PH goes first. From a powerful opening rack, Quackle static values recommend cashing the power first with howf 8g 26 and accepting the leave of AEI worth -6.3 on placement: that's because other top leaves are also negative, such as EFI at -2.1 with whoa 8f (or -4.9 with this placement). PH is going for the unhelpful (as his next play will make clear): no hooks, no easy doubles, only a light quadruple risk. Waif and wife also have better leaves with the H than whoa with the F. General rule, the 6 points upfront are likely better than the couple of points arguing over two-vowel leaves (and even AEI), unless one argues that preventing easy hooks is a goal. Simulation puts only a few plays at 50%-52% win percentage sometimes, such as howf 8e or even at 8f, waif 8f or 8g, or who 8g; also some exchanges like keeping AE or AEW. This seems to favor keeping short words in the middle (and keeping hooks like howff out of double lines) more than keeping vowels away from doubles; tentative preference for the whole play might go to howf 8e (probably not 8g).
>JJB: EFIINQU 7K FIQUE +32 32
#note 0:38 [24:22] Limited options make it unimportant to check alternatives after finding fique/fa. JJB missed quoin 36, but he really wanted the IN leave with the nonuple lane to create the necessary volatility, and to not eat much time early, and EFI with quoin is behind by .9.
>PH: CEFIIOP N6 F.CI +17 37
#note 1:19 [21:42] (coif 9g 29 +15.6) PH rejects the open nonuple decisively: now the only word that will play on 8o is ich! Yet the EIOP leave is a big deficit compared to EIP after coif 29. Piece 27 was also available and probably the better short-term block, as its leave is not so scary considering its immediate points. Epic 24 (blocking effectively upward and downward) holds its own in sims against fuci, coif, piece, pfui 17, and cuif 15.
>JJB: BINORSS 9I ORB +22 54
#note 2:39 [21:43] (bison m9 27 +4.5) PH has helpfully blocked boursins. JJB had rearranged to try to make the nonuple work, but is now desirous to make a subordinate flashy play like insofar 20, not too bad. He takes no notice of m9-13, where 3 words can score 27, and reasons that INSS is probably higher than INS by more than a point (the second S actually only adds a half-point, so orbs should have been preferred). Also synergistic: bos/ow/sh 20 with INRS +15.1.
>PH: EIINOPV M9 POI +15 52
#note 0:27 [21:15] (ovine 10f 23 +4.0) EINV is the best positive leave at +1.3, but Quackle is greedy again and argues for ovine 23 with the not-too-shabby negative leave IP.
>JJB: DIKNSSU O9 SUNDISK +97 151
>JJB: DIKNSSU -- -97 54
#note 0:56 [20:47] (dusk 12k 24 +35.6) Though dusk 24 gives everything he needs, and kinds 10f 34 gives the upfront points, JJB may not have learned his lesson as he decides to risk the game on a speculative term. It "ought" to be a word, and PH is likely to dislike it anyway, and it creates significant exposure besides the loss of turn score, but if the long shot of sundisk* 97 is good then it seals up much of the game, and if not there is still power in the KSS combo. JJB will admit later he is willing to risk the tournament over this term being important enough for the dictionary. PH is again having none of it and has the word removed impressively in the first second of his turn.
>PH: BEINOVZ 10D BIZONE +43 95
#note 2:23 [18:52] (zineb 10f 59 +12.7) Mental note: always connect bizone and zineb. In this case zineb would bring in a cool 59 in the same place. However, knowing one's opponent's rack creates certain problems! If the static values are applied to JJB's turn, both top plays permit a reply of nudnik 39 with SS worth 12.3; so zineb's value of 50.1 becomes -1.2 on next play. Bovine l10 30 is even worse, allowing unkissed! Bovine 10e 26 (more than 10d!) rates 31.1 and does not allow a killer return, but dusk is still 35.9, netting -4.8. That is, the other plays are so far behind that allowing nudnik is actually best, but it should be done with zineb to get two big triples.
>JJB: DIKNSSU H10 .UDNIK +39 93
#note 0:21 [20:26] JJB plays rapidly, almost punitively. His escapade takes what could have been a good start at a lead and makes it instead a very even game, but he is consoled that there is no great loss from it otherwise. PH admitted needing to upend his strategy over the next several turns knowing that SS remained on the rack. But the straight equity of 95-93 is not what either player needs right now.
>PH: ?DEERRV 13B ReVERE.D +78 173
#note 2:12 [16:40] (overdrew h1 95 +17) PH is rewarded for his sharp play by drawing six to the V for a blank bingo, and his choice is a good find among available slots. JJB later suspected that reveried* would be the quadruple: turns out it was override 94. But even higher is the triple with the W, overdrew 95. Choices abound, but PH uses JJB's reveal to limit bingo opportunities significantly. Averred/pia 87 would also help, but PH resolves to block that spot later. Most choices rate around 80% win likelihood, with slight preference for verdured 12e 74.
>JJB: ADESSTT 11C SETA +27 120
#note 4:58 [15:28] (ors j8 25 +9.4) This rack only plays through EFI, and PH has correctly blocked both I's at once. IE might sometimes make a 9, but JJB sees that steadiest does not go (the anagram is stateside). JJB sees Quackle's non-bingo preference, bizones/ors keeping 6, but considers it both unlikely to top the list, and less than 100% certain (in his shell-shock). Seta is fine, though adust gets the same score for the same Z with the vastly preferable EST (not DST). But bizones is indeed the indicated play as there are still enough lanes not to block. JJB is also tactically skeptical of blocking the R line and what remains of the I line, but is craving points at this moment.
>PH: ADEEENO O9 AEON +10 183
#note 0:43 [15:57] (donee 12h 15 +6.7; aeon 9b 13 +3) PH continues considering himself driven to close down the S spots above balancing his rack, with good reason. Options on a vowel draw after a bingo are not high-value anyway. Donee 12h and aeon 9b would give good balance and acceptable score, and donee n11 would also block the key column, but the latter would allow too easy an opponent blowback. PH finds the best column block that also allows minimal reply.
>JJB: ADLSSTT E8 SL..T.D +8 128
#note 2:04 [13:24] (tads 12j 16 +6.7) JJB promptly reduplicates the very letters he had just unduplicated. Insisting on a "traditionally" balanced vowel-consonant leave, he finds a cute consonant dump without premiums. Aside from the fact that bizones 25 is still available and highly ranked, with the leave not a big problem, the bigger issue is that leaves like LST are also in the 10-point range just like AST. So no reason not to pluralize something now (rather than letting opponent do so next). JJB had been lamenting there being nothing clear to pluralize, but now there are four choices.
>PH: DEEELPW 8A PLEW. +42 225
#note 0:27 [15:30] (weeps 8a 39 +3.8) JJB, Quackle, and the principle of minimizing opponent options all favored weeps 39.
>JJB: ?AASTTY A8 .ATTY +14 142
#note 1:39 [11:45] (ya 7c 23 +8.5) JJB's hope for better things from the draw is now rewarded, but he still needs to rebalance. Ya 23 has best value but is rejected for closing lines. Satay/tasty are considered but their optimal placement at l11 is not noted. The play chosen is not far behind and JJB reasons that retaining the S is still preferable on this board.
>PH: ADEEEOR L11 ADORE +16 241
#note 0:55 [14:35] (beaver d10 31 +13.6; adored 12g 21 +5) PH continues to reason that JJB is probably also holding the third S, and closes up column L and row 15 for good. The need to block is wearing on him, though, and having the opportunity for beaver 31 is the kind of deal he does not like passing up. Yet more turnover is usually good for the leading player: get it over with.
>JJB: ?ARSTUV B1 VeSTURA. +74 216
#note 3:21 [8:24] (outraves c2 80 +6) JJB finally gets the synergy and lines he needs, and sees both outraves and vestural, the only two bingos available. Still shell-shocked, under pressure he tries for a minute to recall his very old mnemonic to determine if outraves is merely invented or legit. Unable to say definitively that it is good, he opts for the perversity of opening another nonuple, and the sac is only 6 points; but, if the mnemonic had been reviewed in tourney prep as was eminently possible, it would have been better for both points and further prospects. He has narrowed the gap to 25, and both players know the work necessary for an upset at this point without underestimation; JJB remains hopeful that more good things are coming. Both players are glad the final match turned out to be a well-balanced fight after their prior two games.
>PH: AEEGINX 1A A.ENGE +33 274
#note 1:09 [13:26] (dexie 14e 31 +5.7) On good leave, Quackle would cash the X now, with dexie 31 or ex 14b 30. The best X block is xi c2 30. But avenge makes the best block of the decent nonuple threat, even though there is volatility from the -rs extension.
>JJB: CELMNOT 2E OM +23 239
#note 3:32 [4:52] (cormel 14j 32 +6.8) JJB sees that cormlet doesn't play, but didn't know cormel, the top static choice; he writes out coelom, which is unavailable, but neglects that celom is available and is second. (Third is cleome, the two-E word he sought.) Further, the odd extension is avengeful, not avengement*. But merlon, merlot, cornet, morel, cornel all work, and confab would be a timely extension. Instead, JJB tests his ability to manage one-vowel leaves with C, and narrows the range of bingos available, due to his moving from row 2 to row 3 (neglecting the fact that mol 21 might succeed at those goals better). But breaking up the tempting leaves is preferable in any case.
>PH: EIIJLXY 13K J.Y +26 300
#note 1:55 [11:31] PH now imbalances his X with J and must compensate, limiting his ability to block the open lines. Both joey and joy score equally on top. The first decent blocking play, several points behind, is tix 4b 20, which PH actively considers; but it's an incomplete block.
>JJB: CELNOST 15K T.LCO +21 260
#note 0:43 [4:09] (confab k4 26 +12.0) JJB draws the bingo rack without the bingo but is still in striking distance with 9 in the bag. If the extension confab had been spotted, its powerful leave and new opening lines might have well commended it. JJB chooses the best row 15 filler and takes the risk of breaking up the synergy with a leave worth only +11.8. PH is still rightly concerned but the game is winding down and the bag is not promising.
>PH: AEGIILX 4B .IX +20 320
#note ~2:51 [~8:40] (taxi 4b 22 +3.6) Finally getting to block the upper half, PH opts carefully, with taxi a little ahead of tix. Trouble is that Quackle's top seven plays all allow JJB's actual rack to bingo into the lead; only the nonintuitive axe 3d 26 blocks it, but would allow other S bingos. PH is resigned to the possibility that the block might not work, but 60 is not enough lead to withstand a bingo cleanly.
>JJB: AEINNRS 5D INSANER +73 333
#note 0:16 [3:53] JJB instantly switches gears from his trannies, and plunks down the bingo as PH counts the tiles aloud, acting as if all his fears had come to pass. Hundreds of dollars are on the line. This play gives JJB a lead of 13 and three tiles to play out, and a reasonable hope of winning. Insnare rates 48% (slightly ahead of adding a line with ana 7b 45%); significantly, insaner is only 34%, probably because now column H works better with PH's likely tiles. But since both players made an equivalent number of mistakes through the rest of the tournament, JJB is basically throwing the game odds and recasting them as a final chance test, namely the usability of the 3 tiles remaining in the bag. At this point, though, it's important to note that the significant preference for insnare needs to be gaged not on its hooks but on its likelihood of -ing synergy: thus, one should oddly prefer the bingo that has less synergy with the bag tiles when emptying the bag.
>PH: AEGGHIL 4I HAGGLE +29 349
#note ~6:11 [2:29] (galing h1 27, ism 3a 18, eh 6e 28+2 +20) PH begins with some grumps about the 13-point deficit, looking at seven tiles to three unknowns, with a strong likelihood JJB has drawn into a win. Then PH asks himself what JJB actually holds, and spends valuable minutes completing and verifying his tracking. Both have concluded that JJB has only one out, namely muni. This consoles PH because it is near certain that he can block and go out in two. It remains to find the best way to do this, and in the end he opts to simplify by using six tiles. He had considered column H, but it is not clear the leave will have sufficient synergy, and it is clear with the one-tile leave after haggle. He also doubted haling 33, the highest score there, and turned down heal 6f 34 (recognizing it allows no second outplay). After the game, it is agreed that haling is a better play; but best of all is galing because it allows both eh 6e 28 and her 14j 24.
>JJB: IMU 14J MU. +21 354
>JJB: IMU -- -21 333
#note 3:05.0 [0:48.0] (ism 3a 18, ki 15h 6+2) JJB correctly recognizes that ism loses by 6 and accepts the handwriting on the wall. He goes for a perverse risk of a remaining play: assuming PH permits a phony, and assuming he doesn't spot one of the highest spots for his I (alternatively, assuming a recount finds more points, a recount JJB has already begun), then a tie or win might be squeaked out, swindled, if you will. But this is obviously a vain hope of salvaging a game lost on the last draw. Since the diff of 26 points makes little difference if there's an outside chance of winning the tournament, JJB goes with the confusion between murr and pur/purr in his phony attempt. Whether it was worth the shot is a valid question!
>PH: I 15H .I +6 355
#note 0:20 [2:09] PH is glad to challenge and take the tournament with a final-game spread of +32 rather than +6.
>PH: (IMU) +10 365
#note The players had agreed to let the best man win, and the gap between them was narrow. Judging by the arbitrary standards of Quackle, PH's sacrifices to maintain the lead were very close to JJB's flamboyant attempts to secure it; e.g., PH's release of 20 points with an inferior out-in-two was met with JJB's release of 26 points with a final desperation test of his opponent. Bingo chances and uses were very even. PH surrendered some points on overprotective closure, while JJB did so on poor leave choices. In the end, PH took first and JJB third (with Asif Ali capably winning a game to secure second place) primarily due to a large number of small preparations rather than any individual problematic plays. Yet (given both players' significant prep) JJB's erratic judgment can be questioned as a greater contributor to "drawing poorly" than any other factor. Known points available: JJB 32, PH 45. Overall points available: PH 107.3, JJB 115.5.
|
