| Player 1 |
|
#player1 Joshua_Sokol Joshua Sokol
#player2 Nigel_Richards Nigel Richards
>Joshua_Sokol: BDDQSTU -BDD +0 0
#note From the get-go things were interesting in this underdog vs overdog showdown at the Scrabble chokefest in Buffalo! Many options look close. The only option that scores more than 0 is SUQ. It's quite defensive, and if my opponent does not have an I, I'm alright despite my leave. Unfortunately he doesn't have an I less than half of the time.
>Nigel_Richards: ?HIKNRT H6 ReTHINK +86 86
#note It's an interesting choice here between 8C THINKeR and this. The advantages Nigel may have perceived for his play is the idea that I will usually have kept 4 one-pointers on my exchange, and that therefore I can't take advantage of his slotting of the E next to two double letter squares, and obviously also the extra two points. I think it's really close though, since the (RETHINKING) possibility adds some variance to the game, when Nigel probably doesn't really want that.
>Joshua_Sokol: OOOQSTU 10F QU.T +33 33
#note The sim has this close to plays that keep the QUS combination. Despite my agreement that playing something like I8 OOT is not bad, I want to get rid of the Q and score as much as possible here, cause I'm already in big trouble.
>Nigel_Richards: BBEOPRT J6 PROBE +33 119
#note This play seems right, better than the other option, J9 BEBOP, due to this play's superior defense and score.
>Joshua_Sokol: OOOSUUZ 8J .UZO +23 56
#note I8 OU(T) is the play that simulates the best here. That play is not going to win me the game, despite its benefits: scoring decently while retaining a great scoring tile, not increasing Nigel's opportunities to seal the game early, etc.
>Nigel_Richards: BDIJOTX 7M JIB +37 156
#note When Nigel played this, I imagined that the most likely tile that he kept on his rack was an X. I was right. This is better than JOB because 1) the leaves are similar with the X on his rack, 2) there are many I's to come, 3) some of the time I will give him 50 points for his X already, and 4), it prepares for a very strong next play, a play that is, in my opinion, very much up for debate.
>Joshua_Sokol: EELOOSU O7 .OO +15 71
#note O(B)OE is better than this.
>Nigel_Richards: DIOORTX M5 DO.. +12 168
#note JIB was fine, it was better than JOB on many levels. One of the reasons to play JIB over JOB was to keep this opportunity available for next turn if need be. Now the question remains: does Nigel need to play this? Or, differently put, is doing this worth it? Nigel makes these plays extremely often. Often enough that someone like myself knows just about exactly what he's doing. So this setup is in no way stealthy, which is one of the biggest benefits of a setup. Perhaps just that fact makes DO(JO) wrong, because plays like these stem not from optimality, but rather from the exploitative side of strategy. Let me explain: Quackle plays (or at least tries to play) optimally. It does not rely on inference or any other exploitative procedure to decide what plays to make. It instead relies on heuristics, statistics, probabilistics, etc. This is why Quackle is *exploitable*.If somebody understands Quackle enough, it is relatively easy to make very strong inferences based on Quackle's plays, and by those inferences, modify one's gameplan based on those inferences. Against real players, it is still possible to do this, but it's less reliable simply due to the fact that players have imperfect board vision, word knowledge, strategy and optimality. Sometimes a player will make a play that will make the other player's exploitative strategy backfire, whether intentionally or unintentionally, and this is where exploitativeness is less useful. Nigel could very well have made his play of DO(JO) without having the X in hand, simply because he was trying to out-exploit me. He knows (or could know) that after (B)OO I do not under virtually any circumstance hold an X, and that therefore my chance of having it on my next rack is close to never. In that case, depending on Nigel's rack, DO(JO) may have been a fantastic play. But alas, Nigel did not go that far. He went as far as knowing I have one-pointers and am fishing from my last play, but he indeed has the X and is setting it up, which I can be pretty certain of myself, having seen this same strategy applied in many different situations before, and never a fake setup that would be this obvious. So, knowing I can't normally respond to this setup even though I know what's going on, is this the right play? Well, what other options does he have?N9 TOXOID is a play that springs to mind here. The obvious problem with TOXOID is that it gives me at least 7 more points if I have an S bingo, which is the most likely scenario in Nigel's eye after my first few plays. Although my range is something along the lines of S, 3 vowels and another one-point consonant or S, two other consonants and 3 less strong vowels, it would look like me playing an S bingo next turn is still pretty likely. With all of what my normal range would have been post-BOO, I would have had a bingo last turn anyway, such as ORIOLES, OOLITES, TOONIES, etc., which is why my range isn't as strong as it would normally be. TOXOID scores 34 and avgs 39 the turn after with a partial inference of EILNS, something a tad stronger than normal for me. DO(JO) is pretty much a guaranteed 50 points next turn, which is still 11 points less than the two-turn average of TOXOID, but DO(JO) has a bit more long-term scoring potential than the other plays. (Z)OOID, on the other hand, scores 68 on average after 2 turns and doesn't give me a better S lane, and it also retains a tile with long-term scoring/rack balancing potential; the X. Those seem to be the two plays that would compete with DO(JO) for 1st place here. But I want to get to the biggest problem that I see with Nigel's play of DO(JO), but it's not necessarily with the play itself. I just introduced the concept of "long term scoring potential" as a benefit of keeping the X. But the X, especially in this position of the game, is a double-edged blade. It has good setup potential; DO(JO) is proof of this. However, the strongest setups are those that a) are not obvious and/or b) don't necessarily need to be used next turn to be worth doing. The triple-letter score at N6 does not need to be used next turn, and doesn't help me much other than to fish if I draw an H or something and I don't have a bingo. Nevertheless, here's the thing: Nigel wants to score as much as possible so as to put the game out of reach to me, and it seems that DO(JO) may not accomplish this in the way it intends to. I can exploit Nigel's setup by creating something that he didn't want to create this turn: something quite dangerous that he needs to attend to, and leave him with the choice of either continuing his game of exploitation, or rather cashing his setup. More on this next move.
>Joshua_Sokol: EELLSUV F10 .UELL +16 87
#note Now I can counterplay his setup with volatility and he is left with the choice to either counter my counterplay by playing some defense, if he can, or otherwise just play his own game and cash his setup.
>Nigel_Richards: IORTWWX 14B TWIR. +16 184
#note Yeah, he drew WW!
Now Nigel either cashes his setup for 50, taking a 131-point lead while leaving a rack that he'll have to deal with next turn, while leaving the board virtually uncontrollable for at least 2 turns, or forgo his 12-point setup and make an equity play such as WOW at G5, or lastly, defensively counter my play of QUELL with something like TWIR(L), which he does. I might agree with his play, as it also gives him a chance at 40+ with his X in the near future, but this is basically the culmination of what I was talking about earlier. He scored 12 and then 16 points, and his lead is slowly closing, whereas there were things he could have done differently to prevent this from happening.
>Joshua_Sokol: AENPSSV 6L S.P +21 108
#note Apparently, the bingo percentages post-S(O)P and post-(O)P are similar. I thought SOP bingoed more than OP by a decent margin, but I was wrong. This makes OP a better play, since keeping a second S is better here so that I can score next turn and still keep one. I very much dislike the choice Quackle would make here without the X inference, 15F SAVE, as I feel like Nigel will have a much easier time winning with most of the bingo potential stuck on the bottom left side of the board, which he can block relatively easily in two to three turns. He's scoring 35 next turn anyways though, so maybe I should play SAVE just so that he maybe keeps the X longer and doesn't bingo, and still tries to counterplay my counterplay as he did last turn. But anyways, (O)P was better here. At least he almost challenged S(O)P, but he released after 10 seconds, within which my entire life flashed before my eyes and I questioned my implication in tournament Scrabble!
>Nigel_Richards: DNOWX 11J DOWN +27 211
#note I no longer have Nigel's racks starting here. Forgot to ask. At least here he was scoring a bit better and playing some defense at the same time, but his options with the x were getting less useful.
>Joshua_Sokol: AEGNSVY N10 GYVES +44 152
#note Despite the better leave, GNS, playing L11 (W)AVEY isn't a good option here, since it gives Nigel a great scoring spot and reduces my own options. This is also a reason that S(O)P was wrong, since I now have to hope for another S. The good thing is that I'm now very near striking distance after scoring quite well and countering Nigel's plays.
>Nigel_Richards: AAX 13C AXA. +43 254
#note He finally cashes his X, and is now winning by around the same amount more as he was after my play of (B)OO, with likely a worse rack and many more threats to address. Perhaps in the scheme of things this was a bad case scenario for him, but I think that he committed himself to this possibility when playing DO(JO) and didn't have to. But he's still a very clear favorite to win.
>Joshua_Sokol: AACGNNV L1 CANVA. +31 183
#note The only reasonable option here.
>Nigel_Richards: AFIMOR 1G FORMI.A +42 296
>Joshua_Sokol: ?DEEGNR 2B ENRaGED +82 265
#note The type of draw I was hoping for. Now the board is still very dynamic and I'm approaching Nigel's lead. I just need to draw something else.
>Nigel_Richards: AHLU 3G HAUL +29 325
>Joshua_Sokol: ACEFINS 8A FANCIES. +98 363
#note There it is! Just like that I am now in command! What a clutch draw!
>Nigel_Richards: ADIINRT D11 DI..T +28 353
#note I asked Nigel what his leave here was and each leave he gave me meant that he would have missed an easy bingo this turn, so I determined on my own that it was AINR.
>Joshua_Sokol: AEIIOST 15F STOAI +30 393
#note Now that I draw the case S, I'm in a better position than ever. At this point I wanted a bingo to be the only thing that beat me. I thought that maybe FORMICATE* was a word, but I still felt I could outrun even that. I also wanted to block (ReTHINK)ING, since it's very likely Nigel had two of ING on his leave, although DI(XI)T would be played most of the time anyway if no bingo were available, I think. I had hoped that Nigel was in E trouble due to there being 5 E's in the pool. Quackle likes fishing, but that's not saying much. I think I either need to block bingo lanes or score here, I don't want to rely on bingoing with the pool there was and with my lead.
>Nigel_Richards: AEEILNR 6A NEARLIE. +62 415
#note turns out FORMICARY is a word. I'm sure Nigel knew it. He took forever to play this, since he went through every possible endgame, but when he finally laid it down I said "what took you so long?". :(
>Joshua_Sokol: EEEGIMY 14H GYM +33 426
#note So yeah
>Nigel_Richards: ET 2J ET. +14 429
#note I got Nigel'd. ;)
>Nigel_Richards: (EEEI) +8 437
|
