Riddler - Solutions to Castles Puzzle: castle-solutions-2.csv
Data license: CC Attribution 4.0 License · Data source: fivethirtyeight/data on GitHub · About: simonw/fivethirtyeight-datasette
902 rows sorted by Castle 6
This data as json, copyable, CSV (advanced)
Suggested facets: Castle 1, Castle 2, Castle 3, Castle 4
Link | rowid | Castle 1 | Castle 2 | Castle 3 | Castle 4 | Castle 5 | Castle 6 ▼ | Castle 7 | Castle 8 | Castle 9 | Castle 10 | Why did you choose your troop deployment? |
---|---|---|---|---|---|---|---|---|---|---|---|---|
5 | 5 | 0 | 0 | 0 | 16 | 21 | 0 | 0 | 0 | 36 | 27 | Near optimal integer program vs previous round: beats 1068 of them. |
7 | 7 | 0 | 0 | 0 | 16 | 21 | 0 | 0 | 0 | 31 | 32 | 28 to win. Looked like castles 4,5,9,10 got less troops allocated to them per value than other spots last go around. Didn't bother putting troops anywhere else. Also wanted to be one greater than round numbers like 15 or 30. |
18 | 18 | 0 | 0 | 0 | 17 | 17 | 0 | 0 | 0 | 30 | 36 | Variation on the heavily commit to undervalued top castles, try to steal two smaller ones, and ignore everywhere else. Went for 4 and 5 rather than 6 and 3 or 7 and 2, because people during last battle really committed to 6, 7, and 8 |
19 | 19 | 0 | 8 | 0 | 0 | 0 | 0 | 28 | 0 | 32 | 32 | Gambit strategy that preys on anyone who uses balanced troop distribution. This would have failed in the first iteration of the game, but I predict the metagame shifts towards more normal-looking strategies which will get beaten by this one. |
20 | 20 | 6 | 0 | 0 | 0 | 0 | 0 | 0 | 38 | 32 | 24 | Out of 55 total points, you only need 28 to win, so let's go all in and see what happens! The way to do this with the fewest number of castles is by winning castles 10, 9, 8,and 1. We'll start by doubling the mean allocation from the previous battle, giving 22 soldiers to castle #10, 32 to #9, 38 to #8, and 6 to #1. This leaves 2 soldiers left, which I'll additionally allocate to castle #10 (because I randomly feel people will be more aggressive on that number based on past results). |
24 | 24 | 9 | 0 | 0 | 0 | 0 | 0 | 0 | 32 | 32 | 27 | 10+9+8+1=28 |
61 | 61 | 6 | 5 | 0 | 0 | 0 | 0 | 0 | 37 | 32 | 20 | heavy investment in most valuable positions, with some investment in least competitive battlefields |
103 | 103 | 10 | 0 | 10 | 0 | 20 | 0 | 0 | 0 | 30 | 30 | Intuitiveness |
126 | 126 | 7 | 0 | 0 | 0 | 0 | 0 | 0 | 31 | 31 | 31 | It's a simple all or nothing assault. The goal is to directly seize the 28 points needed to win. The 10, 9, 8, and 1 castles do just this. Contesting any other fortress distracts from this goal. The strategy is designed to overwhelm balanced assaults on the various castles. |
127 | 127 | 7 | 0 | 0 | 0 | 0 | 0 | 0 | 31 | 31 | 31 | The total number of points is 55, so a player needs more than 27.5 points to win. From there, I decided to minimize the number of castles that must be conquered (although that strategy runs contrary to what the previous winner did) in order to maximize the number of troops that can be sent to each one. Using the previous contest's distribution, I (not very rigorously) determined that I would only send 7 troops to Castle 1. The resulting occurrence of sending 31 troops to each remaining castle was a happy accident (although, I wanted to divide them up as evenly as possible; if I lose one castle, I almost definitely lose, so in a sense they should all be weighted equally. However, the opponent might choose to send troops based more strictly on the proportion of points that each castle offers, in which case I would have to re-evaluate my divisions). |
128 | 128 | 7 | 0 | 0 | 0 | 0 | 0 | 0 | 31 | 31 | 31 | Win all my castles with troops. |
129 | 129 | 7 | 0 | 0 | 0 | 0 | 0 | 0 | 31 | 31 | 31 | Win the fewest number of castles needed by loading them up with troops. |
144 | 144 | 6 | 0 | 0 | 0 | 0 | 0 | 0 | 31 | 31 | 32 | This is sparta |
145 | 145 | 6 | 0 | 0 | 0 | 0 | 0 | 0 | 31 | 31 | 32 | Only need a few victories. |
196 | 196 | 15 | 0 | 10 | 0 | 20 | 0 | 0 | 0 | 30 | 25 | You only need 28 points to win, so I tried to focus on getting specific castles and not bothering to protect other castles. |
223 | 223 | 5 | 1 | 1 | 0 | 0 | 0 | 0 | 31 | 31 | 31 | One needs at least 28 points to win. My first thought was to focus on the middle range -- castles 4 through 8, but then realized this could easily fall to a strategy that focused only on castles 10, 9, 8, and 1. The goal isn't to maximize your expected score, it's to maximize the number of times you score 28 or more. Looking at the overall distribution, this distribution looks like it will win a good portion of the time. I throw a soldier to 2 and 3 in case somebody beats me out for castle 1. |
242 | 242 | 10 | 0 | 0 | 0 | 0 | 0 | 0 | 30 | 30 | 30 | I don't need big wins. All I need is 28 points. I figured that I would be able to win the 1-point castle most of the time with 10 troops there and then hope that most people won't be sending more than 30 troops anywhere. |
243 | 243 | 10 | 0 | 0 | 0 | 0 | 0 | 0 | 30 | 30 | 30 | With the caveat that deploying 30 troops for the biggest three is very unlikely, I should guarantee myself 27 points (which is just under half available). I only need to win just one more point to triumph hence deploy the remaining to castle 1 (although there may be some game theory that in the event of others deploying this strategy I should deploy to castle 2 or 3 to take the win over them also). |
246 | 246 | 5 | 0 | 0 | 0 | 0 | 0 | 0 | 28 | 32 | 35 | I am trying to use the most efficient way to 28 points (minimum needed to win) assuming that most players will distribute their troops to more castles. The fastest way is to win castles 10, 9, 8, and 1. I've distributed my troops proportionally to their value. |
265 | 265 | 12 | 0 | 0 | 0 | 0 | 0 | 0 | 30 | 29 | 29 | Need 27 points to win. Target the fancy castles hoping people follow winners strategy from last time. |
275 | 275 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 32 | 32 | 33 | I need to win 28 points, and I'm anticipating heavier resistance at the higher numbered castles. |
277 | 277 | 7 | 12 | 11 | 13 | 11 | 0 | 13 | 8 | 12 | 13 | Ran a quick simulation in R studio |
282 | 282 | 12 | 0 | 0 | 0 | 0 | 0 | 0 | 29 | 29 | 30 | The top 3 castle account for 49% of the points so I decided to hit them hard. The 12 troops to castle 1 should be an easy win and put the total beyond 50%. |
309 | 309 | 0 | 10 | 7 | 10 | 0 | 0 | 10 | 21 | 21 | 21 | Get top 3 + 1 other for 28. |
336 | 336 | 0 | 0 | 14 | 0 | 0 | 0 | 26 | 30 | 0 | 30 | picked the easiest looking quartet worth a majority |
351 | 351 | 0 | 4 | 6 | 9 | 12 | 0 | 27 | 32 | 5 | 5 | Crush enemies. |
366 | 366 | 5 | 5 | 0 | 0 | 0 | 0 | 0 | 30 | 30 | 30 | Banking on people neglecting the highest point castles |
377 | 377 | 6 | 0 | 0 | 0 | 0 | 0 | 0 | 51 | 24 | 19 | Based exclusively off the results in the prior round. |
382 | 382 | 5 | 7 | 0 | 0 | 0 | 0 | 22 | 22 | 22 | 22 | Top 4 castle get all the troops, higher than 20 deployment of the higher points castles to beat anyone else using my system, and another one added to beat those following my system with only one iteration. No point wasting troops on lower point castles, leftovers given to them to maybe snag a few points |
410 | 410 | 2 | 0 | 3 | 0 | 0 | 0 | 0 | 33 | 31 | 31 | Go for broke. Win 8,9,10 and either 1 or 3. |
419 | 419 | 0 | 0 | 0 | 19 | 0 | 0 | 27 | 27 | 27 | 0 | arad.mor@gmail.com |
424 | 424 | 4 | 4 | 0 | 0 | 0 | 0 | 0 | 27 | 31 | 34 | Win 8, 9, and 10 outright and either 1 or 2. This wins me 28 or 29 out of 55. Hope that others put their troops in the middle. |
437 | 437 | 2 | 9 | 1 | 15 | 5 | 0 | 19 | 18 | 18 | 13 | Quick simulation in Rstudio. |
451 | 451 | 0 | 0 | 0 | 12 | 0 | 0 | 26 | 28 | 30 | 4 | San Jose |
460 | 460 | 7 | 8 | 0 | 0 | 0 | 0 | 0 | 25 | 30 | 30 | 28 points wins |
463 | 463 | 0 | 22 | 11 | 11 | 11 | 0 | 11 | 11 | 12 | 11 | Putting 10 across the board beats the current champ. Sacrifice the first castle to beat the other people who think of this. sacrifice 6 to double up on 2 because 8,9,10 needs one other castle to win. |
482 | 482 | 1 | 8 | 7 | 9 | 13 | 0 | 23 | 30 | 3 | 6 | This randomly generated strategy had the best win rate I could find against the previous dataset (of the previous 1387 strategies, this wins 1192 of the matches)... I think. |
499 | 499 | 0 | 0 | 0 | 14 | 2 | 0 | 26 | 26 | 31 | 1 | I already submitted one entry based on "gut". I thought I should do something more method-oriented. This time, I wrote a genetic algorithm as follows: I chose 1000 "random" configurations, each constructed by placing troops 1-by-1, with the chance that a troop goes to a castle proportional to 1+n with n the number already chosen to go to that castle ("Bose stimulation" so the occupancies behave as in a bosonic system of 100 particles in 10 wells). Then I repeatedly held a tournament between my 1000 configurations, recording the best one and keeping the top half. Each of the top half was kept once exactly and once "mutated" by randomly removing 1 soldier and putting him back in with the same (1+n) method, 100 times. These were then the configurations used in the next round of the algorithm. After 1000 tournaments, I had 1000 tournament winners. I played a final tournament between these winning strategies, and submit the one which won that tournament. The winners of "normal" tournaments are mostly of the form, with a few castles heavily fortified and several with less fortification. But the winner of the "tournament of champions" is always of the form, with 28 points worth of castles heavily attacked and a few stray troops sent to other castles. So this seems to be a strategy to use when the other strategies have been "battle tested" to at least some extent. |
538 | 538 | 0 | 0 | 0 | 15 | 0 | 0 | 20 | 32 | 33 | 0 | Forces marshaled on castles in hopes of winning 28 points |
554 | 554 | 0 | 0 | 0 | 11 | 0 | 0 | 31 | 32 | 26 | 0 | There are 55 possible points, so you only need 28 to win. I put a bunch of soldiers at 7, 8, and 9 to total 24 points. I put the remaining 11 soldiers at 4, because I think my opponent won't put many soldiers there. I also made sure to put 1 or 2 more than a round number everywhere I put a soldier. |
598 | 598 | 0 | 0 | 0 | 0 | 0 | 0 | 25 | 25 | 25 | 25 | In round 1, the higher castles were taken by much lower #s of troops. I'm going for the big ones. |
615 | 615 | 0 | 0 | 0 | 11 | 0 | 0 | 25 | 31 | 32 | 1 | Figure #10 is overvalued and #7 is undervalued, enough in #4 to beat even distributions, and 1 in #10 to beat those that abandon it. |
616 | 616 | 2 | 1 | 8 | 8 | 9 | 0 | 0 | 22 | 20 | 30 | Fight for the weak points compared to last time |
640 | 640 | 0 | 0 | 0 | 11 | 0 | 0 | 31 | 31 | 26 | 1 | There are 55 available points, so you only need 28 to win. I loaded up 7, 8, and 9 to get 24 then put the rest on 4 to total 28 (as well as 1 on 10 just in case I lose 7, 8, or 9). I also made sure to put 1 above a round number to beat anyone who put said round number. For example, I put 31 on 7 and 8 so I beat anyone that puts 30 on either. |
647 | 647 | 25 | 0 | 0 | 0 | 0 | 0 | 0 | 30 | 25 | 20 | Last time I put a TON of thougth into it. But so did everyone, leading a lot of people to come up with clever strategies and many people not bothering to fight very hard for the highest value castles. So this time I flipped that on it's head. Nice and simple. Go for the highest value castles (and castle 1) so that my point total, if I win them all, is 28, the minimum necessary to win. |
650 | 650 | 0 | 0 | 0 | 10 | 0 | 0 | 0 | 30 | 30 | 30 | Have to win 28 VP, so go all in on the top 3 and then go for #4 as a random guess. |
670 | 670 | 0 | 4 | 7 | 9 | 10 | 0 | 0 | 0 | 30 | 40 | High risk- high reward. Gotta lock in those big points then have enough of a chance to win the smaller castles to move past the 50% of available points needed to win. |
692 | 692 | 0 | 0 | 0 | 12 | 3 | 0 | 19 | 33 | 33 | 0 | Need 28 points- overwhelm 4 castles to achieve 28 points |
702 | 702 | 0 | 0 | 0 | 0 | 0 | 0 | 20 | 23 | 27 | 30 | because no-one did it last time and I am curious if people will repeat that |
727 | 727 | 0 | 0 | 0 | 10 | 0 | 0 | 30 | 30 | 30 | 0 | Get 28 |
737 | 737 | 0 | 0 | 0 | 0 | 0 | 0 | 20 | 20 | 30 | 30 | Big Baller Brand only goes for Big Points ( I know it's a terrible strategy... just work with me on this one...) |
865 | 865 | 4 | 6 | 8 | 2 | 16 | 0 | 28 | 32 | 2 | 2 | to win. rp |
880 | 880 | 0 | 0 | 35 | 0 | 6 | 0 | 33 | 3 | 2 | 21 | Random solution meant to help my initial submission. |
885 | 885 | 0 | 4 | 31 | 27 | 14 | 0 | 12 | 8 | 3 | 1 | Trying to defeat the winner from last round |
888 | 888 | 0 | 0 | 0 | 0 | 0 | 0 | 7 | 0 | 32 | 61 | Game theory is hard. |
896 | 896 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100 | Because I am hoping nobody else would send 100 troops to castle ten, because they want to have stake in everything, or something else. They also wouldn't be stpid enough to take this calculated risk, like me. It is also hard to amass 10 victory points by a combination. |
897 | 897 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100 | Just to see what happens |
898 | 898 | 34 | 30 | 30 | 0 | 0 | 0 | 0 | 0 | 0 | 6 | The top 3 castles and any other castle will win it. This strategy allows me to big bid on the high value castle. |
899 | 899 | 32 | 26 | 23 | 0 | 19 | 0 | 0 | 0 | 0 | 0 | The deployment aims to get 3 out of four of castles 10,9,8,6, which always gives you over 23 points. I believe most people will spread their troops more evenly. |
900 | 900 | 35 | 30 | 30 | 0 | 0 | 0 | 0 | 0 | 0 | 5 | You only need to win the top 3 castles and the last castle to claim victory (~51% of total points) and since these castles were way underdeployed last time, a big shot in the arm should be enough to take each of them. Since I am completely abandoning the rest, I should be able to over deploy the rest and win the castles that matter. |
901 | 901 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100 | 0 | Someone will try going for 10, just sending all their troops there. Heck, many people may try that. I want to guarantee to get castle 9, and hopefully split it among fewer people. |
902 | 902 | 100 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | i am guaranteed one point |
3 | 3 | 0 | 0 | 0 | 15 | 19 | 1 | 1 | 1 | 32 | 31 | Previous winner won 84%. Took the 90%ile of the previous distribution and subtracted the optimal even distribution of 100 soldiers/28 points. Found best values of 4/5/9/10, and matched those number. Added a couple to the lower numbers. Used the rest to spread between the others with 1 soldier |
8 | 8 | 0 | 8 | 0 | 0 | 0 | 1 | 28 | 1 | 33 | 29 | My approach: let `S` be all the strategies available, initialized to the strategies posted on github. Use simulated annealing to find the strategy that ~maximises `P(winning | S)`, and then add that strategy to `S` and repeat. Eventually we will find a strategy that is "good" against the empirical strategies and other optimal strategies. |
10 | 10 | 6 | 0 | 0 | 0 | 0 | 1 | 2 | 33 | 33 | 25 | Against most opponents, I am trying to win the 10/9/8/1 castles. But there are some strategies that try to do the same, and I attack them on a different front. I don't compete against them for the 10, but trump their assumed zeros on the 7 and 6 (also trumping the guy with my idea with a 2 on the 7). Even if I lose the 9 vs such a strategy I get 28 points if I win the 876 and 1 (tying the rest with 0). |
15 | 15 | 1 | 1 | 1 | 15 | 20 | 1 | 1 | 1 | 30 | 29 | I only need 28 points to win. All I need to do is divide my resources so that I am able to try to get that many points. So I focus most of my forces on the castles that have the most value to me and then get two other mid-range castles to supplement the points. I devote 1 to each of the castles that are not main targets because I figure that if someone is going to beat me on 9 or 10 that they may not have any covering 8, 7 etc. Instead of dividing the points, I think I can win them in these situations. |
26 | 26 | 9 | 1 | 1 | 1 | 1 | 1 | 1 | 35 | 25 | 25 | Ideally, this will beat out many more balanced strategies, as it captures exactly 28 points if all 1,8,9, and 10 (with more than one troop) are captured. The values are all well above the mean and winning strategy for these castles, so people trying to mimic that will lose to me as well. The other castles have one each to pick up any open points in case I lose a main castle. |
28 | 28 | 0 | 1 | 8 | 14 | 16 | 1 | 22 | 3 | 25 | 10 | https://pastebin.com/LSXrjJJV |
29 | 29 | 0 | 0 | 0 | 15 | 18 | 1 | 1 | 1 | 26 | 38 | Mostly random |
52 | 52 | 1 | 8 | 9 | 16 | 1 | 1 | 1 | 1 | 31 | 31 | Nothing flashy. Tried to assemble 28 from the castles that didn't get enough love last time around (2+3+4+9+10). Left a lone straggler at the others to punish the fools that leave castles naked. This strategy is incapable of winning big, but it wins by a small margin an impressive amount of the time. |
99 | 99 | 8 | 10 | 1 | 1 | 1 | 1 | 1 | 33 | 22 | 22 | The goal of the game is not to win as many castle as possible, its to get 28 points. The easy way (looking at the graphs) to do this it to win castles 10,9,8,and 1. The hardest of these is 8 so I invest enough to beat most people on 8 first. I then invest enough to win 1 and 2 as a back up for 1. This leaves me with 49 points. I place one on the castles I'm not going for because the ROI is ridiculously high. (Around 8% chance of getting the points for 1 troop) and split the final 22 between 10 and 9 which should be enough to win most of the time. |
192 | 192 | 5 | 8 | 9 | 1 | 1 | 1 | 1 | 34 | 25 | 15 | Technically, this could perform well if the opponent goes for the middle. I genuinely have no idea if this will work, but if it does, that'll be pretty cool. |
205 | 205 | 1 | 0 | 3 | 10 | 5 | 1 | 1 | 34 | 34 | 11 | I think this beats any major strategy. |
239 | 239 | 2 | 15 | 15 | 15 | 2 | 1 | 1 | 11 | 16 | 22 | Attempt to capture 10 and 9 for majority of points 8 and either 4,3,2 or 4 and 3 and 2, the rest to capture if 0 sent |
240 | 240 | 3 | 3 | 13 | 2 | 4 | 1 | 27 | 30 | 10 | 7 | This strategy was found after simulating tournaments with an agent-based model, where warlords who succeeded enough would eventually reproduce (with the possibility of a mutation in the offspring) and those who were not successful died. On Monday, May 29, 2017, there will appear a blog post describing my strategy on https://ntguardian.wordpress.com |
291 | 291 | 1 | 2 | 5 | 7 | 10 | 1 | 27 | 35 | 6 | 6 | Started with last round's winning deployment. Moved one troop from each of the low yield castles (1-5) each of the high yield castles (6-10). Then assumed that most contestants will continue the strategy of winning castles (1-7) over (7-10) and moved 2 soldiers from each of 1-6 and distributed evenly (8-10). This left castles 1 and 6 unguarded, so i moved one soldier each from 9 and 10 to cover these. |
296 | 296 | 5 | 14 | 6 | 17 | 13 | 1 | 4 | 2 | 30 | 8 | Random solution meant to help my initial submission. |
323 | 323 | 9 | 2 | 3 | 3 | 4 | 1 | 3 | 33 | 36 | 6 | optimised against previous results |
326 | 326 | 2 | 3 | 8 | 11 | 1 | 1 | 31 | 31 | 6 | 6 | Looking at the distributions from the previous attempt castle's 9 and 10 were either the most contested or most thrown so I'm sending only 6 to each to get easy victories if they're thrown, but not wasting too many troops if they're contested. Instead I'm sending more troops to 7 and 8 to secure what should be more guaranteed victories for the troops I'm expending. I'm essentially throwing 5 and 6 and instead putting the troops towards 3 and 4 where I feel I have more guaranteed points. 1 and 2 are low troop counts for the low points, but still putting points towards them to snag easy points if other people throw them. I also chose to use odd/not perfectly rounded numbers (e.g. 6 instead of 5, 31 instead of 30) in a bid to dodge anyone who had similar thoughts to me. |
371 | 371 | 0 | 5 | 6 | 9 | 12 | 1 | 26 | 31 | 5 | 5 | I built a tool that could test any deployment against all the attempts from round 1, plus repeats of the previous top 5 and minor variations, as I guessed some people would just make very minor tweaks to these this time round. The tool would also then test all 90 movements of a single soldier from the first deployment, take the highest scorer, and start again with that deployment until it reached a (local) maximum score. For starting points, I took the previous top 5, plus some random deployments to see if I could improve on them by chance - I couldn't. A couple of deployments stood out as the best, but I suspected other players might find them (who knows), so I fed them back into the test set and re-ran the process. This is where I ended up. |
372 | 372 | 12 | 0 | 1 | 1 | 1 | 1 | 1 | 25 | 28 | 30 | Last time, nobody went for the highest castles, including the winner. If I do, I should beat many of them. |
393 | 393 | 1 | 3 | 1 | 1 | 11 | 1 | 31 | 21 | 24 | 6 | Ceding most 10 matches, but ideally beating a chunk of people who had same thought! |
414 | 414 | 0 | 5 | 7 | 9 | 12 | 1 | 26 | 31 | 4 | 5 | St. Louis, MO |
470 | 470 | 4 | 1 | 6 | 4 | 1 | 1 | 19 | 30 | 19 | 15 | I just want to win... and be victorious... and have my name live in GLORY ON THE 538 WEBSITE!! ARE YOU WITH ME?!?!!... AHHHHHHHHHHHH!!!! ~|---------------> |
488 | 488 | 2 | 1 | 7 | 1 | 8 | 1 | 1 | 26 | 26 | 27 | Shut up don't talk to me I won. |
491 | 491 | 0 | 5 | 7 | 9 | 12 | 1 | 26 | 31 | 5 | 4 | This is my 2nd submission. My 1st submission scored 1176 out of 1313, this one does slightly better, with 1181 out of 1313 (1178 W, 6 T, 129 L). The 1313 are the GitHub plans from battle #1, minus some "clear losers" (i.e. couldn't get to 28 points, used <100 soldiers, etc.). As before, this may not be the global best vs. those 1313. If I find improvements, I'll submit them. I'm sure I'm "over fitting the data", but oh well. :) |
492 | 492 | 0 | 5 | 7 | 9 | 12 | 1 | 26 | 31 | 5 | 4 | trial and error modification on previous winner |
493 | 493 | 0 | 5 | 7 | 9 | 12 | 1 | 26 | 31 | 5 | 4 | Use simulated annealing to try to find the optimal deployment given the list of deployments from the last competition. Then append that deployment to the list (simulating someone who did the same thing that I did) and repeat for a while. This assumes that people will take the last battle's results into account. (Simulated annealing: take a deployment, make a random switch, and test the results against all other deployments. If it's better, continue from the new deployment. Occasionally continue from the new deployment if it's worse Š—– don't want to get stuck in local minima. ) I also made sure it beats last time's winner, in case many people play that. This is not the optimal deployment for the last battle's results (> 91% is possible, though it still would have won) but it's the best one I found that has good performance for both the original list and the list containing good players. |
494 | 494 | 0 | 5 | 7 | 9 | 12 | 1 | 26 | 31 | 5 | 4 | Optimal under previous deployment distributions |
495 | 495 | 0 | 5 | 7 | 9 | 12 | 1 | 26 | 31 | 5 | 4 | Took previous winner's spread and tweaked it to get ~90% win rate using the data provided. Hoping that history repeats itself. |
496 | 496 | 0 | 5 | 7 | 9 | 12 | 1 | 26 | 31 | 5 | 4 | My boyfriend said to. |
528 | 528 | 1 | 2 | 11 | 13 | 16 | 1 | 20 | 20 | 1 | 15 | I designed several strategies that seemed good to me and then designed this one specifically to beat all of them. |
532 | 532 | 0 | 4 | 8 | 10 | 13 | 1 | 26 | 30 | 3 | 5 | Computer Program on the last data set |
555 | 555 | 0 | 5 | 7 | 10 | 12 | 1 | 26 | 31 | 4 | 4 | Alteration of the best solution from the 1st time that would have performed better and won. Additionally, it beats most of the other top solutions. |
571 | 571 | 3 | 4 | 4 | 12 | 11 | 1 | 26 | 30 | 3 | 6 | Randomly selected |
580 | 580 | 0 | 5 | 6 | 12 | 12 | 1 | 25 | 31 | 4 | 4 | ###OPTIMAL SOLUTION FROM ORIGINAL DATA SET### (1255 wins) |
587 | 587 | 5 | 1 | 10 | 15 | 17 | 1 | 23 | 1 | 26 | 1 | trying to get number 9 |
634 | 634 | 0 | 5 | 7 | 12 | 12 | 1 | 25 | 31 | 3 | 4 | I believe it was one of the best possible combinations for previous battle; it may not work if others change their tactics substantially. |
635 | 635 | 0 | 5 | 7 | 12 | 12 | 1 | 25 | 31 | 3 | 4 | This seems to maximize the number of wins with data from Round 1 |
672 | 672 | 0 | 5 | 6 | 9 | 15 | 1 | 26 | 31 | 4 | 3 | Downloaded GitHub data from battle #1, ignored plans that were "clear losers" (couldn't score 28 points, didn't use all 100 troops, etc.), and optimized over the remaining 1313 plans. This deployment scored 1176 out of 1313 (1169 Wins, 14 Ties, 130 Losses). Can't say this is the best vs. those 1313, could be a local maximum rather than a global, just the best I could come up with. |
Advanced export
JSON shape: default, array, newline-delimited
CREATE TABLE "riddler-castles/castle-solutions-2" ( "Castle 1" INTEGER, "Castle 2" INTEGER, "Castle 3" INTEGER, "Castle 4" INTEGER, "Castle 5" INTEGER, "Castle 6" INTEGER, "Castle 7" INTEGER, "Castle 8" INTEGER, "Castle 9" INTEGER, "Castle 10" INTEGER, "Why did you choose your troop deployment?" TEXT );