Mastery Points

[MVNLA2]

@greatleo — Very interesting. Did you only take one bird into the battle? Was it 2191 mastery per bird or 2191 mastery for 3?
Would be really interesting to know if you get the same results for a castle that has harder opponents in the first wave. : D (I find it amazingly easy to think up time-consuming questions for others to answer.)

[MVNLA2]

@greatleo — Well I answered one question anyway. Ran CPC 20 times with level 9 SeaDog (only one attacking) and level 9 Paladin and Princess. Your estimate was 2190 mastery per bird per hour.
My numbers agree pretty well with yours. I noticed that I also got a wealthy rogue 2/20 times. He replaced one of the stick pigs, and was not killed.

[ToothGnasher]

@greatleo – Thanks for doing the research. I’ve been running automation, but at this point most of my birds are at or near level 10, so my long tests are not as useful as yours. That said, I’ll see what results I get with 200 runs. (Level 41, Berserker9, Cleric9, Guardian9 – As noted in my video above, I’ll attack the upper character with Guardian because I’m greedy and 101,000 snoutlings doesn’t seem to be enough. Also, I’ll be using the 10% bonus since it just appeared before I started my test.)

@mvnla2 – My results for a 10.24 hour run in CPC was 1442 mph per bird (4326 mph total).

[MVNLA2]

I tried CPC with same combo as above for 20 times, but this time finishing the first wave, which included a Big Boss. This time I got 0.33 probability of mastery drop per opponent per restart, which implies that the Big Boss is more likely to drop mastery than the smaller opponents. I also got 2495 mastery per bird per hour, but I’ve got to say that doing this manually would be torture.

[MVNLA2]

I also tried SRC, finishing the 1st wave, which had a Pyro Pig, 2 rogues, and 1 stick pig. The results were actually worse than you’d expect from @dr_ishmael’s prediction a few pages ago. It also took about twice as much time for less mastery than CPC.
The probability of mastery drop per pig per restart was 0.28.

[ToothGnasher]

@mvnla2 @greatleo

My run of 200 just finished the CPC test. Here is the raw data. It was an amazing outcome in terms of total mastery gained. I am not sure why I got such a high mph result. Perhaps it is the efficiency of NOT trying to kill the BOSS which, as you noted drops more mastery.

The upper left number is the elapsed time, and that includes pausing long enough to go to the nest (on my IOS device) to get the current snoutlings and determine the “remaining mastery” for each bird.

The leftmost columns are the totals gained. The highlighted column are the snoutlings and then mastery per bird per run on average. That is, 12 per bird.

I ran the experiment in four sets of 50. There was some variation in mastery gained in each set… The smallest gain was 567 and the largest 665 per bird.

In none of the runs was the boss killed, so that left 5 birds available to drop mastery. At level 41 they drop 7 per medal. So… after all that, my results were the same as @greatleo: 34.3% chance that any of the birds would drop mastery.

(Note I misspelled Berserker)

[greatleo]

@mvnla2 @toothgnasher Sorry, I’ve been away from the Interwebs for most of today.

I ran my two tests with 3 birds each: Paladin, Spies and Berserker. Regater’s 3 birds each gained 1022 mastery, Gerta’s 3 birds each gained 1169. So the total mastery dropped is 1022 + 1169.

I’ve run another test where Paladin and Spies attack the Big Boss before Berserker explodes, taking out the whole side. I’ll post results tomorrow morning.

Note that you really need to run a lot of tests to get significant digits. My 5/16 number is a guess.

[Nytmair]

@toothgnasher- Random off topic question… How do you deal with ads during your auto-play using your emulator? Just curious as it seems like they are unpredictable and would throw off the repeating mouse clicks. :)

[dr_ishmael]

@greatleo @toothgnasher @mvnla2 Great research here, guys! Summarizing your results, it looks like the chance of a mastery drop is 1 in 3, or 33.3%. And since MVNLA2’s result with the Big Boss is actually lower than Tooth’s result without him, I don’t think the type of pig makes any difference.

@nytmair Buying Lucky Coins will disable the forced ads, at least for a while. I purchased a big pack around a month ago (so I could unlock Sea Dog and Thunderbird before the Puzzle and Dragons event ended), and I haven’t seen any ads since then.

[MVNLA2]

@dr_ishmael — I think I may not have been clear enough, and didn’t post the details of my results. Will try to later. For my results alone, which are admittedly only for 20 runs, it did make a difference if you killed the Big Boss. More later.

[greatleo]

I ran two separate tests during which my birds took out all 6 pigs–in one turn–in the first wave. If possible, take out all 5 pigs in the second wave, again in one turn. In all tests I used a Berserker (mostly for its Explode feature, but it’s also very powerful when the Rage Chili is filled) either a Paladin or Samurai, and either a Spies or Mage.

Here’s the tactic I used:
Attack Big Boss with Paladin/Samurai and either Spies or Mage.

Scenario 1: If any of the weaker pigs are still standing, use Berserker’s Explode to finish off the wave. After wave 2 loads, restart to wave 1.

Scenario 2: All of the weaker pigs are killed. Finish off Big Boss with Berserker and proceed to wave two with the Rage Chili still filled. Attack Rogue Leader with Paladin/Samurai and either Spies or Mage. Finish off the wave using Berserker’s Explode. After wave 3 loads, restart to wave 1.

In each case I ran 50 Wave 1 attacks, and tallied the number of Wave 2s. That number depended on the randomness of chain attacks from Red and Mage/Spies. I assume that the Big Boss drops mastery with the same chance as the weaker pigs, but with twice the value.

Here’s my raw data (I’m pretty sure I spaced out undercounted a wave here):

Test 1, Dec. 13:
Regater Leo mastery before/after:
Paladin 4379 – 3385 = 994
Berserker 7725 – 6731 = 994
Spies 1718 – 724 = 994
Wave 2 scenarios: 9
Total Mastery drops = 994/7 = 142
Mastery opportunities: Wave 1 = 350 (i.e. 50 x 7 [Big Boss = 2, plus 5 weaker pigs])
Mastery opportunities: Wave 2 = 45 (i.e. 9 x 5 pigs)
Total mastery opportunities: 395
Mastery probability: 142/395 = 0.395

Gerta Leo mastery before/after:
Paladin 16932 – 16057 = 875
Berserker 36376 – 35501 = 875
Spies 5392 – 4517 = 875
Wave 2 scenarios: 10
Total Mastery drops = 875/7 = 125
Mastery opportunities: Wave 1 = 350 (i.e. 50 x 7 [Big Boss = 2, plus 5 weaker pigs])
Mastery opportunities: Wave 2 = 50 (i.e. 9 x 5 pigs)
Total mastery opportunities: 400
Mastery probability: 125/400 = 0.3125

Test 2, Dec. 14:
Regater Leo mastery before/after:
Samurai 1281 – 301 = 980
Mage 11612 – 10632 = 980
Berserker 1710 – 730 = 980
Wave 2 scenarios: 18
Total Mastery drops = 980/7 = 140
Mastery opportunities: Wave 1 = 350 (i.e. 50 x 7 [Big Boss = 2, plus 5 weaker pigs])
Mastery opportunities: Wave 2 = 90 (i.e. 18 x 5 pigs)
Total mastery opportunities: 440
Mastery probability: 140/440 = 0.318

Gerta Leo mastery before/after:
Samurai 9826 – 8937 = 889
Berserker 34595 – 33706 = 889
Spies 3111 – 2222 = 889
Wave 2 scenarios: 13
Total Mastery drops = 889/7 = 127
Mastery opportunities: Wave 1 = 350 (i.e. 50 x 7 [Big Boss = 2, plus 5 weaker pigs])
Mastery opportunities: Wave 2 = 65 (i.e. 13 x 5 pigs)
Total mastery opportunities: 415
Mastery probability: 127/415 = 0.306

Finally, we can combine all of these results to hopefully get a more accurate probability:

Total mastery dropped: 142 + 125 + 140 + 127 = 534
Total mastery opportunities: 395 + 400 + 440 + 415 = 1650
Mastery probability: 534 / 1650 = 0.323

Honestly, I’m tempted to throw out the very Test 1, since I’m pretty sure I under-counted waves, and it’s an outlier. Meanwhile, Test 2 gives numbers very close to my first test where I only took out the 5 weak pigs in wave 1:
Probability = Drops / Opportunities = (140 + 127) / (440 + 415) = 267 / 855 = 0.3123.

This value is very close to the 0.313 probability from my very first test that took out only the 5 weak pigs in wave 1, 200 times.

The only way to really know if Big Boss drops mastery with a higher probability than the weaker pigs is to craft some birds that have no chain attacks, and then take out the Big Boss, and only the Big Boss, say, 1000 times. Small numbers of runs contain too much randomness.

[MVNLA2]

@greatleo — Very interesting data. I totally agree that my data falls in the category of “statistics” of small numbers. One question though — Do you know that the Big Boss drops twice the mastery?
[Edit]: Since GreatLeo collected so much data, I don’t think it is worthwhile to give more details about my few numbers of runs.
For the record, I have switched to CPC for farming mastery.

[MVNLA2]

@greatleo — Just checked — the Big Boss does drop twice the mastery. Above when I reported results with and without the Big Boss, I was assuming he only dropped one mastery. Consequently, my conclusion about his frequency of dropping was incorrect. The same frequency with twice the mastery is probably just as good or better. Sorry, but I’m not about to find out about his frequency. Checking the mastery was painful enough.

[MVNLA2]

@toothgnasher — Didn’t you somewhere give instructions on what software to use to run automatic play? I can’t remember where. I play on iPad and also have a MacBook Air.
Thanks for any help you can give.

[greatleo]

@mvnla2 You’ve already answered the question yourself, but anyway, I got the info about Big Boss dropping double mastery from the work of @dr_ishmael.

[MVNLA2]

@aquadude — Just for fun, or perversity, I tried Cave 9-5. Just once so really no statistics, but it does have 10 waves, so that should somewhat average out the probability of mastery drops. I’m level 10 and I played with my level 9 Paladin and Princess, and a borrowed level 10 SeaDog. I did get 150 mastery for each of my 2 birds, and it took 14 min, which gives 643 mastery per bird per hour. However, you only get mastery for 2 birds, so compared to Star Reef Castle, you aren’t getting as much mastery per hour, and Cobalt Pig’s Castle is even better. The down-side of the Castles compared to Cave 9-5, is that you have to pay pretty close attention to the Castles, because they don’t take that long. If you can’t pay very close attention, say only every 15 min or less, you will probably do better using Cave 9-5. However, my results in Cave 9-5 are based on borrowing a very strong bird, so without a borrowed strong bird, it will take much longer…

[ToothGnasher]

@mvnla2 I ran a test to see if the Boss in CPC wave 1 drops mastery at a different rate from the others… In 100 repetitions, my bird accrued 532 mastery which is 38 out of 100. Slightly higher than prior runs, but in the ball park.

I did the test by using Skulkers10, Thunderbird10 and Samurai9. Thunderbird attacks Big Boss, then Skulkers then Samurai, then restart. I made sure each bird had critical strike weapons, not chain attack so only the attacked pig (Big Boss) would die. Of course I didn’t watch EVERY one of them. It’s possible the rate is higher because if the Wealthy Rogue showed up, the Big Boss ends up in a different place.

Genymotion takes a bit of work to set up since you have to install Google Play manually, but here is the Mac video:

View it on YouTube where there are notes, too.

[MVNLA2]

@toothgnasher — Thanks! But why do you need to play an Android version? Granted that would be a good way to get the 100 LC that haven’t come to IOS yet, as far as I know. I guess I was hoping to be able to do something directly on my iPad, leaving my MacBook Air free, but if I run it at night, what difference does it make? If I run it during the day, I can play other AB games on my iPad. Still sounds like a lot of work, will consider whether I want to do it, but I’m getting close.
BTW — Interesting data on the Big Boss.

[ToothGnasher]

@mvnla2 asked:

@toothgnasher — Thanks! But why do you need to play an Android version?

There are three reasons:

1. There are more hacking tools for Android while there is NO emulator for iOS,
2. The 100 LC is only available for Android; and, most significantly,
3. There does not seem to be a Mac version of ABE.

[dr_ishmael]

@toothgnasher re: Big Boss testing, your Samurai should be able to one-shot the Big Boss with Heroic Strike. At level 41, rank 9, and with a level 41 Titan’s Wrath, my Red’s Heroic Strike hits for 3431 damage. Level 41 Big Boss has 3289 health. So you could simplify the Big Boss testing to just dropping the chili on Red, and the position of the Big Boss won’t matter.

[dr_ishmael]

I just completed my own sets of trials on CPC-1, with similar results.

Scenario 1 – Use Bomb’s Explode rage ability to kill the 4 Stick Pigs and the Bird Catcher. 300 trials (performed as blocks of 100 at different times during the day).

Note: I performed these trials manually on my iPhone, so I was able to keep track of the Wealthy Rogue’s appearances and account for them in my calculations. The WR appeared 18 times.

Results:

• Bombs exploded: 300
• Pigs exploded: 1500
• Mastery-dropping pigs: 1482
• Mastery earned: 3731
• Mastery ribbons: 533
• Drop rate: 35.96%

Scenario 2 – Use Red’s Heroic Strike rage ability to kill the Big Boss. 100 trials.

Note: I automated these trials in Andy, but since Heroic Strike automatically targets the pig with the most health (on CPC-1, it’s the Big Boss), there was no possibility for variance between the trials.

Results:

• Big Bosses Heroically Struck: 100
• Mastery earned: 518
• Mastery ribbons: 37
• Drop rate: 37%

The results are close enough I think we can conclude that the Big Boss does not have a significantly higher chance of dropping mastery than the other pigs. Overall, the mastery drop rate is somewhere in the range of 32-37%.

[drobbinsjr]

So why the hell do you even play the game ?????

[Zack Voyager]

@drobbinsjr Since Rovio didn’t release any update since more than 1 month, this is all they have to do : find quick ways to upgrade all birds at lv 10 mastery…

[greatleo]

@dr_ishmael Huh… Interesting and confusing. In my tests I also counted Wealthy Rogues, but didn’t subtract them in my count of pigs dropping mastery. I forgot that they don’t give mastery.

Here are new numbers from two of my tests factoring in Wealthy Rogues:

Dec. 12:
Bombs exploded: 200
Pigs exploded: 1000
Mastery-dropping pigs: 989
Mastery earned: 2191
Mastery ribbons: 313
Drop rate: 31.65%

Dec. 14:
Pigs eliminated: 855
Mastery-dropping pigs: 844
Mastery earned: 1869
Mastery ribbons: 267
Drop rate: 31.63%

Even factoring in the Wealthy Rogues, I still get numbers that are lower than yours.

Additionally, given the number of tests you and I have both run, each of us has numbers that should produce at least two significant digits of accuracy. That is, my result should be trusted to at least 31%, and your result of 35.96%, i.e. 36%, should also be trusted.

At the moment, the only question I can think to ask is: do you run your tests on an Android device? I run my tests on an iPhone and an iPad. Those are definitely different code bases.

[dr_ishmael]

@greatleo I ran scenario 1 on my iPhone. Scenario 2 was automated in Andy. (I’ve added this to my earlier post.)

How did you track the number of trials? I “synchronized” my hands so that as my left hand hit “pause” and “reset” in the game, my right hand was typing “0” and “Enter” into Excel – when a WR showed up I typed “1” instead of “0”. I used the Count() function to determine the number of trials and the Sum() function to determine the number of WRs.

It’s possible that I under-counted the number of trials, but I would have had to have missed at least 40 trials to reduce my drop rate to ~31%. I’m pretty sure I didn’t miss that many.

How soon did you reset after dropping the chili? I waited at least until Bomb exploded and the damage numbers appeared. In previous tests, it appeared that resetting before the explosion was too soon, and I did not receive any mastery or snoutlings.