Feeding Heroes: One at a Time vs. Ten at Once

Nobody cares about axes when you draw bell curves by hand.

1 Like

Oh look! An error! :grin:

Kidding, kidding…

I personally enjoyed your explanation. I’m still doing 1x 1* loading on my Magni. :slight_smile:

2 Likes

Mine hit max skill last night at 3-12. Such an improvement over Thorne.

2 Likes

It was desperation. I found myself trying to explain the concept verbally, and realised that a picture really is worth 1000 words, so thought I’d jot one down. Tried loose-hand drawing and the curves looked even worse than what I posted above so did a hacky compromise using line bending in Paint.

For a short while I really was considering just coding the automation and having Excel produce the curves, but thought that would just be a step too far for what is really just an academic point about a simple tablet game. Really, if I’m going to code, I should be doing it for my work. :smiley:

1 Like

I was not sure on the beginning but it works just fine.
I still do 10x 2* of the same color when i have them on 3* star heroes, but the rest of the time i use one by one.

The very difference is… You are not so disappointed if your special don’t goes up and very happy when it goes.
On the contrary, when you use 10x same color and the special just ignore you it really ■■■■ you off :rofl:

3 Likes

Well, that is an advantage I hadn’t previously considered. :joy:

1 Like

So I have been following this conversation on and off ever since the day @ScndStar originally brought this topic up in chat. I have found the whole debate rather amusing, but have now decided to add some content to the discussion.

I decided to do 100,000 simulations of a few of the basic scenarios.

Scenario #1: Feed 1* Heroes of the same color 1 at a time.

Scenario #2: Feed 1* Heroes of the same color 10 at a time.

Scenario #3: Feed 2* Heroes of the same color 1 at a time.

Scenario #4: Feed 2* Heroes of the same color 10 at a time.

Before the results of the scenarios here is some basic info:

1* Heroes of same color Fed 10 at a time = 20% chance of Special skill Increase

Feeding 10 1* Heroes of same color 1 at a time:
# of Skill Upgrades %Chance
0 81.71%
1 16.67%
2 1.53%
3 0.08%
4 0.003%
5 0.00007%
6 0.000001%
7 0.00000001%
8 0.0000000001%

(if you take the %chance times the # of skill upgrades and add them together guess what % it approaches…20%)

2* Heroes of same color Fed 10 at a time = 40% chance of Special skill Increase

Feeding 10 2* Heroes of same color 1 at a time:
# of Skill Upgrades %Chance
0 66.48%
1 27.70%
2 5.19%
3 0.58%
4 0.04%
5 0.002%
6 0.00007%
7 0.000001%
8 0.00000002%

(if you take the %chance times the # of skill upgrades and add them together guess what % it approaches…40%)

And with that I will now move on to the scenarios I listed above. The thing I measured in these simulations was the number of feeder troops required to upgrade a hero’s special skill 7 times (Max it). Skill increases due to ascensions were not factored in.

Scenario #1:
Minimum # of Feeders Needed: 31
Maximum # of Feeders Needed: 1196
Average # of Feeders Needed: 349.6
Mode: 286

Scenario #2:
Minimum # of Feeders Needed: 70
Maximum # of Feeders Needed: 1190
Average # of Feeders Needed: 350.4
Mode: 290

Scenario #3:
Minimum # of Feeders Needed: 18
Maximum # of Feeders Needed: 589
Average # of Feeders Needed: 174.71
Mode: 148

Scenario #4:
Minimum # of Feeders Needed: 70
Maximum # of Feeders Needed: 530
Average # of Feeders Needed: 175.4
Mode: 150

So what does all this show us? On average it is going to take the same number of feeder troops to get your hero to max regardless of if you do 1 at a time or 10 at a time. However, if you do it 1 at a time there are times when you can get a hero to max using significantly less feeders. This is because if you do them 10 at a time the best you can hope for is 7 tries or 70 feeders, however, feeding them 1 at a time the best you can hope for is 7 tries or 7 feeders (more likely to win the lottery, but still possible). So, I suppose if you want to be as efficient as possible with your feeders you would do them 1 at a time, but in reality it’s not going to make much of a difference and you should use them however works best for you.

The link below is to a Google sheet that contains some of the data from the simulations and charts.

https://docs.google.com/spreadsheets/d/1xidazrIQYdpHWK63GHcEqOB_MWAoQhQ9JEscdk6t_6o/edit?usp=sharing

15 Likes

Excellent data and calculations, KA. Thanks so much for contributing!

1 Like

If you try your simulation for when you need only one special skill increase (it will be at 7/8 at some point…), I think you’ll draw a different conclusion.

1 Like

@kahree would you care to expand on that line of thought? If I only had 10 feeders and I needed to assure myself of the best possible chance of making my hero’s special increase 1 time within those 10 feeders then yes, I would feed them all at the same time. Otherwise I am not so sure about it. I will run the Sims as you suggested when I get some free time.

1 Like

What about feeding 1x 1* rainbow (different colors?). That was what I was doing.

Also, I was using this on previously maxed 3* heroes, not unmaxed, so I’m not in a hurry to get these leveled. Am still testing in real time (though I paused to Level Magni). :wink:

1 Like

For increasing the special on a hero that is already max level. I would only feed 1 at a time of the same color. Unless, I had no other heroes to feed the others to.

1 Like

Did you run the simulation both ways? I’m genuinely curious.

So after the Sims I would still personally feed 1 at a time. For 1* feeders of the same color it is going to take on average 50 feeders to get that increase. In my 100k sims the actual results were 49.99 feeders on average for 1 at a time and 49.80 for 10x. So for 1* it really doesn’t matter. But for 2* feeders it does. The average drops to 25 feeders on average either way, but doing 10x you can’t get 25 so there an increased chance in wasting feeders. Now if you wanted to do 5 at a time the long term results would be pretty much the same without the potential waste.

Ultimately it really doesn’t matter which way you choose the results are going to end up pretty close to the same in the long run. In the short term you can always have outliers.

2 Likes

@rook I did not run the sim for your question.

2 Likes

This is inefficient. Before a hero hits max level, you’re getting 1% special-up odds for 150 xp; using matched color, you get 1% for 90 xp. (The goal is to maximize the %/xp.) you also use twice as much food per %.

Likewise after the hero is maxed, you are using the same food per card but only getting half the chance of a power up.

1 Like

Food cost is something that has been ignored in this debate and it would tip the scales in favor of 10x feeds a lot of the time. I believe the food cost per feeder is determined by the current level of the hero you are feeding them to. If you feed them 1 @ a time you will always use the max amount of food required. If you do 10x then you will skip the increase on food cost for some of the levels… especially early on.

3 Likes

I’m not worried about food; right now I permanently am overloaded with food.

My question is, does “half the [percent] chance matter?” Eg, is the chance itself per try the important part, not the percentage? Is 1% materially less than 2% in how the game calculates these chances? (Read back to Little_Infinity’s Post in this thread.)

1 Like

true, but the food cost at the beginning is minimal compared to the food cost at the end. so having an extra little bit in the first bunch of levels isn’t a big increase in overall food for the lifetime of a hero.

1 Like

I agree that it isn’t a big thing, but if you are looking for optimal usage there is a difference.

1 Like