Welcome to the Candy Crush Saga Community
Do you have a question or need help with your game? Ask the Community.
Now this will be a weekend full of testing and number crunching!
christinewupp
Posts: 10,868 
in Discussions
What a beautiful opportunity the game has given us to find an answer to these questions:
- Where are players of Candy Crush from?
- What level ranges are players playing at?
- Which level ranges give the most purple candies?
- Are players at specific level ranges at an advantage or disadvantage?
I am referring to the Valentine event of course: 75 random players per leaderboard. All players of the game are invited, all had an easy chance to qualify for this final, in fact the qualifying round was difficult for anyone not to achieve. So essentially it would appear that every single player of the game is taking part in these finals.
Please could anyone reading this just confirm if they have NOT got this event in their game?
1
Comments
-
Don't worry I won't be asking anyone else to take notes of all those other players in their games this time. I have four finalists myself, and there are a whopping 75 per leaderboard, so that will be enough data to get some idea I should think. But I might ask you to help with one of your graphs @MannyFae .
So far I'm starting with Cornflower as its Candy Royale is about to expire. The other three will start later. This event is running over 71 hours.
1 -
Well, I don't have this in my main game 🙂 Probably this is because I stopped updating it a while ago, but that's for the better. I don't want to play non-stop all Sunday like I did last time.
75 per leaderboard is a lot. It's going to be a tough battle for the not-so-great reward. Good luck to Cornflower and other finalists!
1 -
Here is another question I've been trying to work at and first results are in on that one:
What is the best strategy (for me) to score the most purple candies?
I have carried out a controlled experiment on two strategies. I put the "for me" in brackets, since we all play the game differently and have differing strategies that suit us best. So my best strategy is not for everyone. I play with hints turned on, for example and many players prefer not to. A lot of players also prefer to pay gold or watch ads for extra moves instead of using boosters.
Both my games are new players on the top win streak. Both games used starting boosters for all levels. Both played exactly the same two episodes: levels 1221 - 1250.
Strategy Marshmallow game: Economy (no super colour bombs)
Use the minimum boosters required to pass each level and to keep win streak.
I used the strategy I generally recommend to others in my posts.
Using this strategy for me is to second guess what the algorithm demands of me, including the boosters needed near the end to pass. Therefore I always start with the first move the game recommends and play the recommended moves if I see no better choice.
Results:
Episode 1221-1236: 60,540
Episode 1236-1250: 60,630
Time taken: 25 minutes per episode. (Score per hour: approx. 144,000)
Boosters spent: UFOx4, LPx2 over 30 levels (note this was an easy level range so relatively cheap on boosters, boosters needed to pass change from episode to episode, and will be higher on higher levels)
Strategy for Piggy game: Speed (while super colour bomb enabled)
Make first move recommended by game. (This will often cause an initial cascade with the super colour bombs)
Use a party popper on second move. (This may finish the level or leave a few simple moves to finish it off)
Finish the level off as soon as the opportunity arises.
Episode 1221-1236: 50,760
Episode 1236-1250: 53,430
Time taken: 13 mins first episode, 15 mins second episode. The discrepancy in the episodes was mainly due to level 1240, where the dragon does not come out until halfway through the level after some good cascades that cannot be rushed
Score per hour: approximately 210,000
Conclusion:
This would make strategy "Economy" a good strategy to use in earlier qualifying rounds for me. To win the finals rounds strategy Speed is probably essential.
In practice of course I would use a mix of the two strategies, depending on how much gold I have left and how strong the opposition is.
My findings would also suggest that if you use strategy Speed any player could win a competition just by using a party popper on each level and bashing their way through it. The game no longer requires thinking skills, so competition will be fierce if there is money involved.
3 -
In addition to the above I have now played one episode at levels 5376-5390.
New levels, on top win streak, playing all starting boosters, no super colour bomb. (Cornflower)
Score for the episode: 65,610
Method was variable, involved lengthy advert watching to minimise costs, which would of course not be feasible in a leaderboard contest. Legendary levels were very booster hungry so I took advantage of ads where I could. It took me 37 minutes with watching the ads, but I have no doubt had I used the Economy method described above I could have completed the episode in 25 minutes without ads and reached a similar score.
I also played the Star Tournament. (Poppy)
No win streak boosters and no super colour bomb. Playing all starting boosters.
Strategy similar to Speed method: open up the board while there are opportunities to create special candies and combos. When all special candies are gone, play a party popper. Skip levels that are too slow, e.g. skull levels.
Score: 187,000 per hour. I'm sure this can be improved upon with experience and getting to know these repeating levels levels better.
That's enough strategy testing for me now. Time to calculate the costs next.
3 -
Results are in:
Question 1. Where are players from?
I tried to stagger my own start times over a 24 hour period with my four games but due to sleep requirement could not spread them out ideally. UK Start times were: 9am, 13.30h, 21.00h and 07.00h Saturday morning. These are my results:
Western Europe: 67 (23%)
North America (USA/Canada) : 33 (11%)
Australia : 2 (1%)
Middle East (including Turkey): 30 (10%)
Africa : 18 (6%)
South Central America/Latin America (including Mexico): 29 (10%)
Asia: 114 (39%)
Now if we think my data are skewed due me not starting a game in the middle of the night these are the data for only the 9am and 9pm games:
Asia: 45 (31%)
Europe: 37 (25%)
Middle East: 19 (13%)
Latin + South America: 18 (12%)
USA and Canada: 18 (12%)
Africa: 9 (6%)
Australia: 1 (1%)
So basically about a third of all active players are from Asia. Most of course from India. (There were none in my games from China or Japan)
With All Stars in mind, this would mean that if there is to be a live final same as we had the past two years, the majority of players would have to be excluded due to visa requirements. I wonder if King have something up their sleeves to include the large number of players in Asia this year.
3 -
Questions 2 and 3:
What levels are players playing at? and
Are any level ranges at an advantage over others?
I won't bore you all with the exact data at this point but can provide a more detailed picture if anyone asks. This is the summary from four leaderboards, so a total of 296 random players:
Early levels: 1-3000: 40%
mid range: 3000-10,000: 38%
over 10,000: 22 %
Now to answer the question if any level range is at an advantage:
The top ten players from each of the 75 were:
Early levels: 38%
mid range: 35%
>10,000: 27%
So actually if anything the early levels did worse than expected and the highest players did better than expected. This is not surprising since experience matters!
I noted that in the past we calculated that players in the range 1-2000 had to crush through a lot more levels to score as much purple as players in the 5000-9000 group making us think those levels are more lucrative. I'm not so sure now if this is not just due to lack of experience @MannyFae . I managed to score between 50,000 and 65,000 per episode at my best speed in the 1200 + range which was very similar to what I scored with the game above level 5000. Other players on my boards in the 1000-2000 range played a lot more levels than I did just to reach the same score. I suspect that all ranges are pretty similar scoring using the starting boosters plus win streak strategy, the only difference being the amount of boosters and gold required to crush through the levels, which of course is higher the higher you go.
By the way I won three of my groups. I was surprised this was possible, but I was playing faster than my opponents and managed to overtake them from behind even while they were playing, which made them then give up and allow me to switch to my other game. My Poppy game won almost entirely using the Star Tournament.
3 -
@christinewupp You might be right according to your test it does seem the majority of the players will not be able to join the all stars which is so wrong. Maybe King has been working this year on how to include more players. The all stars has to be the best way to grab money. Best to try and include as many countries as possible.
1 -
Congrats to all three winners! 🙂 I'm especially glad for Poppy conquering the Star Tournament.
I won my group in my new game as well, which is still below level 1,000 due to being mostly on hold. I averaged 53k per episode.
Are scores similar across all level ranges? If I remember correctly, we calculated average scores for blue candies previously and concluded that middle range levels yield more. I'm very curious about the scores data for purple candies in the 7,000-9,000 range. Sadly, I didn't participate in any purple collection leaderboard events while at those level ranges, so I don't have my own data. The only data I have is from All Stars, which was 85k per episode in round 4 and 72k per episode in round 5 (I was in the 10k range at that time). I still tend to think that midrange to higher levels yield more, but this is offset by the longer time needed to pass them, when it comes to hourly speed.
3 -
Congratulations on your win!
I'll be curious how boring you'll find those repetitive levels devoid of any conveyor belts or other interesting features. I can't remember when they end. Good luck with your game!
2 -
So basically about a third of all active players are from Asia. Most of course from India. (There were none in my games from China or Japan)
If I am not mistaken, CCS is not available in China.
2
