How do y'all complete puzzles so fast?

[uigrad]

An incorrect submission's 120 second penalty is pretty steep, but for most puzzle sizes it might be better points-wise to take that hit and correct the answers, even if it means starting from scratch (and it often won't). If one generally gets the solutions right and averages say 85% of the max score, I think the better gamble is to accept scores of, say, 30% for the error and 85% for the next one instead of 0% and 85%. If it takes longer to correct the error than would have been spent solving a whole new puzzle (in which case the comparison is between 30+85 and 0+85+85), that probably will be a loss of the gamble for points, but might help one learn something about the error made and help reduce the odds of a recurrence.

I agree with this all 100%.

There are a lot of possible goals that people have when playing:

• Score enough point to be in the top 20 for the month
• Get a few top times
• Keep "success rate" at 100%
• Keep points/time ratio on the monthly chart as high as possible
• Learn logic
• Have fun

Points 3 and 4 seem to be in opposition to each other. Personally I lean towards 3 and don't care about 4 at all, but some people care a lot about 4 and not at all about 3. Those people are the ones likely to "throw away" a puzzle that has errors.

Generally, I focus on #1 and #2. Since the site has become more popular, I can't really achieve #1 any more, but for the first few years, that was usually my top goal. Now I play more for #2.

Really, #5 and #6 should be the top goals. If I get stuck on a puzzle (not very common, but it does happen), I prefer to use the "hint" button and slow down and try to learn (for goal #5), and completely ignore the points. Some players prefer to never log in, because keeping track of times would ruin goal #6 for them. These are all valid ways to play.

[redhill]

@ contrary : no need to apologize, enjoy your tea! (are you perhaps British like me, although I drink more coffee these days!)

[contrary]

@Fromalabama: That makes sense to me if you were intending to maximize points, where the backtrack is more point-efficient. I find it's very rare that I actually get to the end of a puzzle and get an incorrect submit, because usually if I messed something up earlier in the puzzle most often one of the other clues I have left doesn't make sense.

@Redhill: I apologize - maximizing points on the competitions never even occurred to me as a point on this thread, since I was answering this question in a context of individual puzzle by individual puzzle. I play because it's a great way to give my brain a nice reset, and three games on 4x5 is between a 5-10 minute break with 5 minutes left for me to get a new cup of tea.

[fromalabama]

I don't want to criticize anyone's preferred strategies, but if one is trying to maximize points, error-checking may be too much of a gamble, at least if one is generally getting times less than the median time. The penalty is so harsh that making/removing more than a single error basically assures getting only one point for the puzzle if it is fixed and resubmitted. An incorrect submission's 120 second penalty is pretty steep, but for most puzzle sizes it might be better points-wise to take that hit and correct the answers, even if it means starting from scratch (and it often won't). If one generally gets the solutions right and averages say 85% of the max score, I think the better gamble is to accept scores of, say, 30% for the error and 85% for the next one instead of 0% and 85%. If it takes longer to correct the error than would have been spent solving a whole new puzzle (in which case the comparison is between 30+85 and 0+85+85), that probably will be a loss of the gamble for points, but might help one learn something about the error made and help reduce the odds of a recurrence.
Of course, if one generally gets the solutions right, the error-checking will rarely matter at all, costing maybe a couple of points per game at worst, not enough to seriously affect any of the monthly rankings. So in that view it's not much of a gamble. But by the same logic, the outcome of any single puzzle, whether a zero from giving up or a relatively low score from an incorrect submission, will also not have much effect. Achieving one of the monthly trophies may arguably have more to do with how much time in total can be devoted to solving the puzzles than how efficiently one wracks up the points.

[redhill]

Yes, I care about maximizing points, for isn't this the point of competing each month ? Since 2013 when I joined as 'curryqueen' I was in the top 20 with 100% success rate and gained 'trophies' more than seven times. I used to be placed around 48 in the 'top 100 of all time' too. Having taken a long break and refined my game tactics I now play as 'redhill' but compete less frequently. Playing the odd game keeps my brain sharp! So, my point is that assuming you are playing 'to win', it is essential to speed-read the clues if you can. Checking for errors before submitting a puzzle enables you to choose whether you submit only your highest scoring correct puzzles and move up the leader board faster. As a result I no longer have a high 'success rate' but this doesn't affect my actual games as far as competing is concerned. Of course you must take time as well to backtrack or go more slowly sometimes. Most of the top players have increased their speed over the years by constant practice and familiarity with the types of clues too. As it is said, "it takes dedication and skill" !

[contrary]

Speed-reading and scanning of the clues is the only way to play fast and get high points. Over-analysing is tedious and slow. Hit the error check button when puzzle is complete before submitting it-(you only lose a fraction of a second ). If puzzle has errors, ditch it !

That assumes you care about maximizing points. Is there something about incorrect submits that lowers your points for puzzles you ditch? Otherwise I would say skip the error check button since it's an extra step and you get the same information ("something in your solution is incorrect") by clicking the submit button, and you don't lose that fraction of a second.

I also wouldn't do this if you're still working on getting faster. Instead, either backtrack until you figure out where you read the clue wrong, or start over and work the puzzle more slowly until you find where you went wrong. That's how you'll catch reading errors and traps. For example, the one on the crossword puzzle story, where one of the clues usually reads "[Person X] got [word#] right". Speed reading that when the puzzle first came out I would see [Person X] [positive association] [word#] and associate it the two positively and always get the puzzle wrong. Now I know I need to look for that trap.

[redhill]

Speed-reading and scanning of the clues is the only way to play fast and get high points. Over-analysing is tedious and slow. Hit the error check button when puzzle is complete before submitting it-(you only lose a fraction of a second ). If puzzle has errors, ditch it !

[tviolet]

I'm trying to teach myself to only use the top boxes and not fill in the bottom ones at all

I've been working on this for the last few days and it's definitely faster - I've gotten two top scores today! And if you see the attached, previously my scores were pretty much in line with forests's scores, now I'm 20 to 40 seconds faster.

However, if I get stuck, I go back to my old method of using the lower boxes and tend to wind up with a score that is much slower than I would have been. I also think using the upper boxes only is only possible because I've been doing these so long and would have been impossible when I was newer. So I guess my advice is practice practice practice and eventually transition to only using the top boxes.

[tviolet]

Finally got one! I'm at this point in the puzzle, I know the following. Each line is a discrete compiled clue in my head. The clues I didn't include in the screenshot are no longer relevant because any possible remaining solutions satisfy them.

Order: ( River Kings) - (blank) - (12) - (25). Corrollary: River Kings is in position 1/2, number 12 is in position 3/4, number 25 is in position 4/5.
25 is (4th position/Comets) OR (5th position/Monsters)
Underwood can't be 12
Underwood or Guerra are (3rd position OR Wolverines (1st OR 5th))
There is no more information about Underwood or Guerra in the puzzle.


If I put 12 in 4th position, that puts (25/Monsters) in 5th position, leaving (Wolverines / (Guerra OR Underwood) in position 1 and (Guerra OR Underwood) in position 3, and satisfies the clue that Underwood can't be number 12, since 12 is in position 4. There is no more information to tell us whether Underwood should be 1 or 3, and there is no clue referencing Guerra.
Therefore, 12 must be in position 3/Guerra, River Kings must be position 1, Underwood must be position 5/ Wolverines.

It's definitely not a shortcut when you have to type it all out.

Gosh, that's a lot to hold in your head, but I can see if you can do it, it's very fast.

After reading this thread, I'm trying to teach myself to only use the top boxes and not fill in the bottom ones at all - I was using them a little but concentrating mostly on the top and now I can see the extra time to do even that much will limit how fast I can get. It's interesting, it's either faster or I get hopelessly lost and it's much, much slower. I'll keep working on it.

[contrary]

Oh, that's a tricky one. For me, I'd probably rely on Clues 2 and 9 to tell me that #2 is the player from the Monsters. (Clue 9 refers to Owen and Duran, and based on Clue 2, neither of them are from the Monsters, and since Clue 9 itself says #2 isn't on the River Kings, the only team left is the Monsters.)

From there, Clue 5 gives #25 as the player for the Comets, so 9 goals. #12 has 8 goals, and based on Clue 6, isn't Underwood so must be Guerra. Clue 8 says River Kings scored 6 goals. Wolverines scored 10 goals, and is Underwood, based on Clue 1, and the puzzle is solved.

That's the proper way to do it; there's always a solution that doesn't rely on assuming solvability. For me personally, that's really slow in this case because I don't have #2 in my mental index anywhere because I have no order reference for it, so I would have to go back to the clues in the screenshot and figure out which one has a full exclusion criteria from scratch.

[boxeeboxeebox]

Finally got one! I'm at this point in the puzzle, I know the following. Each line is a discrete compiled clue in my head. The clues I didn't include in the screenshot are no longer relevant because any possible remaining solutions satisfy them.

Order: ( River Kings) - (blank) - (12) - (25). Corrollary: River Kings is in position 1/2, number 12 is in position 3/4, number 25 is in position 4/5.
25 is (4th position/Comets) OR (5th position/Monsters)
Underwood can't be 12
Underwood or Guerra are (3rd position OR Wolverines (1st OR 5th))
There is no more information about Underwood or Guerra in the puzzle.


If I put 12 in 4th position, that puts (25/Monsters) in 5th position, leaving (Wolverines / (Guerra OR Underwood) in position 1 and (Guerra OR Underwood) in position 3, and satisfies the clue that Underwood can't be number 12, since 12 is in position 4. There is no more information to tell us whether Underwood should be 1 or 3, and there is no clue referencing Guerra.
Therefore, 12 must be in position 3/Guerra, River Kings must be position 1, Underwood must be position 5/ Wolverines.

It's definitely not a shortcut when you have to type it all out.

Oh, that's a tricky one. For me, I'd probably rely on Clues 2 and 9 to tell me that #2 is the player from the Monsters. (Clue 9 refers to Owen and Duran, and based on Clue 2, neither of them are from the Monsters, and since Clue 9 itself says #2 isn't on the River Kings, the only team left is the Monsters.)

From there, Clue 5 gives #25 as the player for the Comets, so 9 goals. #12 has 8 goals, and based on Clue 6, isn't Underwood so must be Guerra. Clue 8 says River Kings scored 6 goals. Wolverines scored 10 goals, and is Underwood, based on Clue 1, and the puzzle is solved.

[contrary]

Finally got one! I'm at this point in the puzzle, I know the following. Each line is a discrete compiled clue in my head. The clues I didn't include in the screenshot are no longer relevant because any possible remaining solutions satisfy them.

Order: ( River Kings) - (blank) - (12) - (25). Corrollary: River Kings is in position 1/2, number 12 is in position 3/4, number 25 is in position 4/5.
25 is (4th position/Comets) OR (5th position/Monsters)
Underwood can't be 12
Underwood or Guerra are (3rd position OR Wolverines (1st OR 5th))
There is no more information about Underwood or Guerra in the puzzle.


If I put 12 in 4th position, that puts (25/Monsters) in 5th position, leaving (Wolverines / (Guerra OR Underwood) in position 1 and (Guerra OR Underwood) in position 3, and satisfies the clue that Underwood can't be number 12, since 12 is in position 4. There is no more information to tell us whether Underwood should be 1 or 3, and there is no clue referencing Guerra.
Therefore, 12 must be in position 3/Guerra, River Kings must be position 1, Underwood must be position 5/ Wolverines.

It's definitely not a shortcut when you have to type it all out.

[uigrad]

Now, if one of the last clues ...

For most examples I can think of, the key is that you are down to just a few remaining clues.

A situation that I often see is where I am down to my very last clue, and the very last marking in the chart. Keeping with your example, imagine the only unfilled boxes I have are Alice ($5 or $10) and Bob ($5 or $10). Neither one is $15.

I have one last clue, and I don't remember all of it, but I do know it was comparing Bob to something else in the chart. It says that Bob is higher or lower than some other thing.

Rather than going back and reading the clue, I deduce that it can't be a "Bob is higher" clue, because if so, I would have marked $5 as false for Bob. So it must be a "Bob is lower than ..." clue.

I don't need to read the rest of the clue to solve this. If that "other thing" is $15, then I know the puzzle is unsolvable, and all puzzles on the site are solvable. So, the clue must be comparing Bob to something else that is $10. Therefore, I can mark Bob as $5 and Alice at $10, without taking the time to actually read the clue.

Does it save a lot of time? No. I doubt it even saves a full second. But when I'm going for records, even half a second could make the difference.

[Laxgirl]

I cant believe how fast you can do it!! That's amazing!! Great job!!

[boxeeboxeebox]

I'd be interested in an example, I can't quite picture what you're describing.

I'm not sure if this is the exact situation contrary had in mind, but if you're in a situation where under Names you have Andrea, Bob, and Charlie, and where under Winnings Andrea and Bob could both have won $5, $10, or $15, while Charlie won either $5 or $10. And under Winning Hand, Straight flush could correspond to Winnings of $10 or $15.

Now, if one of the last clues you have says, "Bob won with a straight flush," it will likely be the case that Straight flush does not correspond to a Winning of $15, because if Bob won $15, then Andrea and Charlie will both have $5 and $10 in Winnings, and unless you have another clue to decide who won what, the puzzle is unsolvable.

However, this is a shortcut, and you can always get to the solution without making this kind of observation (though there are other types of logical eliminations based on remaining possibilities that aren't this "meta").