Round 4 was the only round where I couldn't finish all the puzzles. I was a bit behind after some of the intermediate-value classics threw me, so I found myself with a choice of trying for the 20 pointer (and maybe having enough time for just one of the two 12's), or simply going for 24 points with the easier 12s which I was sure I could get, and running out of time to do the 20. So I took the 24 points approach as opposed to higher variance, and left myself with the biggest fish and no time to catch it. But at least it gave me a few minutes to check the rest.
Round 5 was the first where my checking failed me. Being able to turn in with 9 minutes left, I spent 3 minutes checking before turning it in but didn't catch a set of fixable mistakes in the kropki that basically involved swapping a large set of digits in the last two rows. My main attention was on the super valuable puzzles (dots and levers), so I probably neglected doing enough on the Kropki; either column checks, or "missing dot checks" would have shown something was slightly off. If I'd caught it, I could have saved myself a 32 (14 puzzle + 18 time) point mistake, but I didn't.
Round 6 (where we still don't have results) went well for me, but I ran into the problem of having way too much time on the clock (19 minutes left when I was done). I took a full 4 minutes to check all 6 grids since losing the points would be very bad. But sometimes at the end of the day my eyes glaze over during this process so I hope I didn't miss anything blatant (I did spot two notes instead of numbers right at the start of this, but nothing else).
My favorite puzzles of the afternoon session were the Arrow and Property Sudoku in round 5, and the Blackout Skyscraper and Killer/Inequality of round 6.
Round 6 also had a typo error in the Little Killer that at the very least affected Ulrich. As the story has been explained to me, there was an error in the example for this puzzle where a 23 clue was given as a 33. Specifically the 5th arrow on the right side of the grid from the top. Someone was told to fix this in the example, and did so, but made a similar change on the competition puzzle where that exact arrow was a 33 and was not supposed to become a 23. That particular constraint was well removed from where I was using any sums at all, but as a small number with lots of bigs feeding into it, it could certainly have affected someone's solution.
With most of the first day of results, the surprise is the new name near the top, Kota Morinishi of Japan. I've seen his name before in some internet tests, typically around 4th place, but not within striking distance of my scores. He was by far the fastest in the first round, and blew through round 5 (5 minutes faster than when I could have turned in, but I took time to check and didn't get my error anyway). Combined with some poorer rounds from Jan, Kota and I are now the ones well in the lead, with Jan and Jakub in a second group, and then a closer bunch for the remaining spots.
Results are now posted here, as a computer screen cap it seems, but the quick excerpts after 5 of 7 are:
1 Kota Morinishi (JAP) - 764
2 Thomas Snyder (USA) - 752
3 Jan Mrozowski (POL) - 672
4 Jakub Ondrousek (CZE) - 658
ETA: As I predicted in my first post, where round 6 would be my best round, this indeed has turned out to be the case. I won three of the 4 variation rounds and have a nice gap on second. So cruise control in Round 7 will be plenty:
1 Thomas Snyder (USA) - 957
2 Kota Morinishi (JAP) - 895
3 Jakub Ondrousek (CZE) - 842
4 Jan Mzorowski (POL) - 802
5 Hideaki Jo (JAP) - 795