I definetely switch on 6-8 after @Revelate suggestion.
Same recruits, more heroes and more interesting items for less herbs and experience of 8-7.
Wish had discover it some updates ago.
3 Likes
Wouldn’t it be amusing for the developers reading this thread if they’ve coded the loot to be adaptive—over-hunted stages lose loot, under-hunted gain.
That’s how we coded the MMORPG where I was a dev—the treasure drops from monsters depended not only on their level but the relative frequency of their deaths. You could do very well by hunting in the backwaters of the game, provided you had the sense not to blab about it.
6 Likes
Probably you are right, but thats no other choice if you want to share.
8-7 was probably nerf because it was the best stage for everything. Experience, recruits, items.
Now while i see some preferences over the forum, probably there really isn’t a best stage for all the people everytime anymore.
So i hope the nerf is over 
3 Likes
After 100 runs I see a ratio of 2.5:1 for backpacks and swords in 5-8. Since the moving average is quite stable, I don’t think that this ratio will decrease to 2.0:1. But that doesn’t matter. For me it’s more important that I need 3 energy for 1 backpack in 5-8.
1 Like
100 runs is nowhere close to enough data but call it what you will, your statistical error rate is still very large at that count.
I didn’t even start to get any consistency till over 300 runs FWIW.
Been hitting 23-9 lately and it has been fun, nice raid, testing out what heroes manage that level, and been getting 10-11 recruits, 1-3 backpacks, 1-3 swords, roots, nuggets, dust, lots of items plus 1* 2* and hero troops, sometimes no heroes but many times 2 of them.
I just want heroes to feed lately, so every hero, sword or backpack is a feeder be it 1 or 2 stars. 8-7 sometimes gives me 2 heroes bonus tok but only sometimes, 23-9 gives more, spends more energy but is more entertaining
1 Like
high five got a Delilah 3 times in a row (did a 300 gems draw three times over the month, always a lousy 3* + bonus draw of Delilah). It’s really cool to farm with them on lower levels, when 15 minions do their attack with the “ching-ching-ching!” sound… pretty funny.
4 Likes
Exactly what I am talking about. It is like AK-47 against a sturdy shield.
1 Like
My problem with farming the higher stages is that you really don’t want to have to resort to battle items to prevent losing the level… that kinda defeats the object. The other criteria is auto-playability.
I also like to dip in to the higher (18+) levels from time to time… there is no denying the loot is good. I can only safely autofarm up to 19-9. After that I either have to at least keep an eye on it in case the AI makes it go pear-shaped, or help it along by selecting targets while auto-playing. (How many people are actually aware you can do that? 
)
But even then the sums I did in terms of recruit counts (which is one of my main focusses with my need to get 6 workable teams levelled up) and monster counts (to keep turning over my monster chest) makes the lower cost levels more worthwhile. I haven’t really taken the time to count the completed heroes received (of which I ‘feel’ there are more in higher levels) and factor that in to the recruits calculation, but there is still the need to turn over enough monsters to fill that chest in 1 batch of energy.
(P.S. Another 2 consecutive OrNu from 8-7 today - I’ve been looking out for 'em… and have had quite long runs (5 or 6 in a row) where I’m getting either MiRo or Grimoire Dust. I’m really still finding the loot in 8-7 quite worthwhile for my needs).
Thanks for the heads up about 6-8 @Elpis. I’ll give it a go.
1 Like
quick question. I am probably not there yet but what are the nugget and dust used for?
I have 73 nuggets and 185 gr dust. I am guess they will come in handy at some point.
Thanks.
Scott 
1 Like
Dust is for mana (regular and large) and for Dragon Banners (very handy for tough titans).
Nuggets for Tornado - Forge level 19 (very handy for recalcitrant boards against titans).
3 Likes
@Little_Infinity Thank you for responding. I am still working on all the strategy stuff, especially for titans. Need to get my hits up and carry the right items in with me.
Thanks again
1 Like
In my other F2P MMO “BoH”, every year the Devs added a new play area that incoporated some, but not all, of the forum comments from the previous year.
But also every year the Devs added new features and players discovered the utility of under used features.
It is interesting to go back through the different areas and remember what was the “hot new loot” area for each of the past six years.
Looking forward to Empires and Puzzles Season 2 loot surprises.
2 Likes
Been trying 6-8. You’re right in recruits… same if not slightly more, and you’re right about monsters… a little bit more. Same energy.
From what I’ve observed so far, however, the drops and Hero/Troop rates are not nearly as good as 8-7. Sure you still get surprisingly good drops from 6-8 for such a low level, but I feel 8-7 is still better value for energy.
(Maybe I’m just really lucky with 8-7… who knows).
Very good point. 23-9 I also find is much better for stuff. But I really don’t need stuff since AW started. I need recruits recruits recruits. It feels like a solent green factory at my base. : )
Tc19 is one of my favorites for growing feeder heroes. Recruits is often what I run out of. If I ever get done leveling my AW C and D lister heroes I will move back to 23-9
Glad to hear someone agrees with my gut feeling on 23-9
3 Likes
I’ve been collecting data on feeder materials while farming 8-7 for the past 452 runs. Below is a table of the results. I didn’t bother logging crafting mats b/c that would require an additional screenshot and make things a whole lot more annoying. The top row is the average value per run and the bottom is a 90% confidence interval of that average. The last column I call “recruit spread”. It is the amount of recruits you would gain per run, assuming you used all training camp materials every run (backpacks TC 2, swords TC 1, clothes TC 19). This data was going to be part of a separate post comparing it to another province (maybe 5?). But I figured there is enough here for anyone interested.
EDIT - The text beneath recruit spread is wrong. It should be Recruits-BP5-Swords5-Clothes2. The formula above it in the cell is correct though so the value should be right.
3 Likes
Depends on the spread. Even with huge data sets, you don’t always need hundreds of runs for some good conclusions. A common misunderstanding is that you need large amounts of samples. In general, a sample population can be much smaller than people think… unless the random factor is really big.
Considering the numbers thrown around, there’s randomness, but not so much that you need 1000 runs to make some solid conclusions (1000 samples is a LOT)
Think about it, if the margin of error is about 10% on 100 runs and 5% on 300 runs (depending on the dataset, but that’s a pretty standard level error) the difference would be somewhere between 0,1-0,2 backpack. With those error levels, going to 1000 samples would improve the error level towards 3%.
Of course, the margin of error could be larger if the data is really random, but considering the unanimous trend between everyone’s research it’s probably not so bad (especially not on lower level items with a larger drop rate).
That’s why polling can be done with pretty small samples. The problem there is that it sometimes hinges on those few percent swings, but generally the polls are just a few percent from each other (again, those few percents matter a lot in those cases).
Considering that it’s about farming a province and the differences between a specialized/non specialized province are so big, those 5-10% are probably not that important.
4 Likes
True if there is enough granularity, and the specialized vs. non-specialized I agree which was why I was pretty comfortable saying 2:1 for specialized at 500 rolls.
I did see in my data where there was some skew even above data count that for 8-7 but eventually it trended 1:1 for swords and backpacks.
Trying to get accuracy for percentages to figure out optimal farming locations, more data… for just showing if you really really want backpacks to the exclusion of everything else, just go to province 5 I agree.
1 Like
I really don’t know enough about this game yet to understand enough of what your talking about and yes I see many people do but my guess would be their know this game better than me! Thanks for your comment! Can the Dev. make a video on YouTube with tips that would be sweet if you so sooner than later thanks peace out!