Preface
Going into season 13, the vibes were pretty so-so. In season 12, I played decently for my standards, but still had a pretty rough season, where I finished 6-8. I managed to squeak into the safe zone on account of ending with a +184 spread after two blowout wins to end the season.
Game 1
One of my first interesting games came in my game against zzz. It actually started with a large board vision error.
With only 18 tiles in the bag left and trailing by 67 points, 301-368, I made the play of M6 JIG 16, trying to set up another lane where I can bingo starting with a blank S. I somehow didn’t see that this play actually fits placing the J on the double letter, shifted one space down at M5 for 30 points instead. Those 14 points are absolutely crucial here to make it more likely that a bingo I play will actually win the game.
Later in the preendgame, I was still in trouble. I had drawn the second blank as well, but was now 94 points behind after having passed on a low scoring bingo with both blanks on a previous turn.
In this position, I made the play of 10J W(OO)ING for 20, leaving one tile in the bag, trying to threaten a triple triple as well as a possible bingo to the C if I could draw one of the 2 remaining I’s from the pool. The problem with this play is that I’m still down a lot of points. Down 74 points and not on turn, there are way too many easy scoring plays through the G in the triple lane that outrun any bingo I can draw to the C. Also, the odds of drawing a bingo to the C is still not quite enough even with both blanks in hand. The engine’s favorite play is J9 TWa for 32, leaving GIN? behind. I somewhat doubt this actually wins 44.9% of the time because my opponent can try to block the more dangerous spot with a one tile play instead of emptying the bag, but this makes it a lot more likely that a bingo will win the game. There are still quite a few bingoes I can draw to the C, as well as high scoring bingoes to the right of BOO, and this option doesn’t offer the same cheap scoring plays in the triple lane.
In the actual game, zzz responded with KI(G)HT for 42, taking an insurmountable 116 point lead, and I drew too many clunky one pointers and ended up with no playable bingoes, even holding both blanks.
Game 2
My next interesting position came against gharp. It was a clunky, close game up to that point, and I was down 8 points with only 3 tiles left in the bag.
In this position, I made the play of 10D G(OW)F# for 15. My goal with this play was to preserve my only vowel (though there were vowels unseen) and get rid of the G, a poor scoring tile. In hindsight, I think I undervalued the power of the F as a scoring tile on this tight board. An adjustment I could’ve made was to play G(OW)D for a 2 point sacrifice. If I draw another vowel, then I can respond with high scoring plays to the triple if gharp makes a play down the N column. Also, I still save some bailout plays like 7B CAF hooking FLAG to eject high pointers.
Afterwards, we reached this endgame position. I’m trailing by 8 points and gharp is holding AEIMRU. In this position I made the play of N2 ANN# 14, setting up the outplay of 1L DICT# hooking CANN#. What I saw was that after he plays O1 MI for 14 to block my outplay and threaten the outplay of F4 AURE(I), I could block his outplay with F7 D(I) for 3 points, and he would be forced to play something like ER for 10 at O4 making BONE and I would win by just 2 after playing out with 7B TIC hooking CLAG. Instead, gharp surprised me with one of his winning plays after D(I) of URE(AL) [ARE(AL) is equivalent] for 15, leading to a 5 point loss. There are actually no winning plays in the original endgame position, but INN instead of ANN in the same spot performs 1 point better after a much more complicated sequence.
Game 3
Most of the way through my game with ralphi, I was in big trouble.
Down 131 points 2 turns after his 105 point bingo of WILDiSH, I barely manage to keep my hopes alive in this position with L8 HEKTaRE for 98 points, though I’m still only at 11.6% to win.
Two turns later, after I’ve been trying to keep the board open despite no S’s unplayed, I’m in this position with 4 in the bag (AACFLMNRTTU unseen), down by 48.
Here, I was debating between two plays, both at C12: either ZO# or ZIN for 34 or 36 points, respectively. ZO holds the tiles to play a potential sneaky DE- bingo hooking BETED#/RE and keeps some decent odds of drawing a bingo hooking (ZO)A if I can draw one of the two remaining A’s, or through the E in HEKTARE. My problem, I felt with this play, was that I was still only about 10-25% to bingo at best, not fully factoring in all the blocking plays he has, and I almost never bingo at all without the A to hook ZOA. I ultimately decided to go with ZIN. My idea was to set up the (ZIN)E hook on the triple, and the only other hook unseen is the lone C for ZINC. If I can draw the C or if it’s still in the bag, then I have some very difficult 40+ plays to block on the triple that could maybe win me the game without bingoing. DEOR could still also draw some bingoes that are difficult to block while also taking out the triple lane.
After ZIN, I drew AFT from the bag and was threatening plays like 15A TREFA 51, as well as the bingo FOREDAT(E) to the E in HEKTARE. ralphi responded with 15A MAC to go up by 48 points and I managed to bingo out, winning by 28.
There were some difficult 100% winning plays from ralphi’s perspective, holding ACLMNRU such as 14H NUCL(E)AR for 30, but they aren’t the most intuitive when the massive E hook is looming large on the triple.
Game 4
In this position against ferenc, I’m down 46 on a somewhat poor board, holding 5 vowels. My 7 letter bingoes of AUTOcUE# and COUTEAU don’t fit, and there are no 8 letter bingoes available (OUTAr(G)UE also doesn’t fit). In this position, I made my favorite play this season.
In this position, I got down the play (BE)AUTEOUs for 72.
There’s no other way of putting this, this was a lucky find. Though I’ve studied all the 5 vowel 7’s and 8’s, I never really looked at the 9’s, except that I had taken a random cursory glance at the short list of 6 vowel 9’s (though I didn’t study them) about 2 weeks before the season and had a small moment of recognition when looking for BEAU- plays.
Afterwards, I pulled AABEILN, and while ABELIAN didn’t fit on the board, even though I was only up 10 points after my opponent’s play, Macondo suggests I was already over 99.3% to win because of how light the unseen tiles were on vowels.
My MO after seeing this unseen pool was to play at most 1 vowel a turn while keeping the scoring pressure on.
Adjustments Needed
In my game against AnitaCH, I was in this position, leading by just 2 points.
I had some solid scoring plays such as 14H VOX for 40 points. I was extremely hesitant to play this because it had been 4 turns since they had last played an S and multiple of their recent plays were shorter, lower scoring plays dumping vowels, perhaps suggesting a decent rack only marred by too many vowels, but still good enough not to exchange. I ultimately decided against playing VOX for 40 because of the gigantic S hook it creates at H15, while still holding HNQV for next turn. I settled on the uninspiring 11B V(E)X for 22, but this is not Macondo’s favorite choice. Using a 5-ply sim with both inferences and with partial rack S, Macondo actually suggests exchanging down to HOX, holding the 40 point play in the same spot for next turn, where I may be able to use it with less risk if they have already used their S.
Interestingly, VOX still outperforms V(E)X even with the inference and partial rack, so I managed to both under and overadjust to AnitaCH’s play.
On the next turn, still in an about even game, I was in another challenging position.
Here, I relatively quickly played F2 QI for 34 points, holding some consonant dumping bailout plays like (DI)V(I)NG for next turn. This was a large board vision error on my part. I saw the word HOVING# on my rack, but somehow only saw it played at 14H for 32, which still sets up the massive S hook on the triple. If I had spent a little more time, I could’ve seen that it still plays at 11G for 38. Leaving the lone Q normally isn’t ideal but it’s a big asset here.
After HOVING#, I’m both scoring better and setting up a big spot to cash the Q on my next turn, with 2 I’s unseen for the simple QI for 64 (and F2 QI 34 as a backup) and also 4 A’s and 2 U’s that could be used to play QUA or other plays in different spots. This miss is actually a massive 17.7% win rate error.
Hey That’s My Spot!
In this position against ather, I had just bingoed with RADIANS to take an 86 point lead.
I drew ELOPUW? and saw I was threatening the massive triple triple of POWE(R)fUL for 158. Before I got a chance to play it, ather shattered my hopes with his own triple triple of UNDE(R)LIT for 131, leading to about a 225 point swing after I played UPfLOWE(D) on my next turn.
Blunder
Finally, in this position against VVB, I made my most obvious error. I simply did not look hard enough to find my common word bingo on my rack of DEHORNS, and just played 9G H(M) instead, for 70 less.
Summary
Overall, I think this is one of my best played seasons yet. I had my lowest average MiAOW ever at 2.5 and my highest average turn score at 36.3. 8-6 +506 is also my 2nd best performance in terms of wins I’ve had in division 1. I definitely need to do a lot more studying because I am certainly still at the bottom end in terms of word knowledge compared to other division 1 players.
Stats
Lost Endgame Points: 40 (8/14 optimal endgames, though this counts 1-move endgames)
Missed Bingo Turns: 3
Suboptimal Bingo Turns: 7