Iām not convinced of thisā¦but as I have neither data nor experience. ā¦
What is an ideal war defense to use this against? I will start trying it once I get skadi if sheās offered.
Do you use a mix of fast and avg healers in addition to kiril?
From what I noticed itās more rewarding to gian the points from feeding these 20x5* heroes into one you already have than waste them on one of these offered here.
I think if your going to blow 20 5* heroes and throw the money you may have spent getting them (plus the mats if your going to use a levelled hero) away then the least they could offer in return is a choice of the 5 top/strongest heroes in the game.
Noone know for sure what will be the offered heroes, but I think there is about 90-95% chance of these will be the offered heroes, but we will see in about 3 weeks.
HA10 is a 5% chance of a non-s1 hero which is pretty much all anyone cares about. Thatās 1 in 20 pulls on average which is going to be ~2.5 pulls a year or 5 every 2 years. If you want to include costumes of S1 heroes as useful that does increase it but it isnāt quite the same thing and also has a limited pool and higher chances for dupes pretty quickly.
Soul Exchange actually ends up being pretty close to that when you work out the math for 3 TC20ās. ~5% chance of a 5* hero from TC20 from what people have gathered. Same 1 in 20 chance and a 2 day training time so 40 days of training time per 5* hero on average. With three TC20ās that means a 5* every 13.3 days or roughly every 2 weeks. That means ~26 5* heroes a year which is basically the same rate of heroes at the 10 to 1 trade in level. You can opt for less frequent trades if you want a hero from a higher tier and might get some 5*s from other sources to help out with more frequent trade ins or higher cost trade ins.
As someone who is pretty C2P this should be a pretty noticeable improvement on the rate I get 5*'s. Being able to pick is also very appealing. Iād love a Titan killer better than Wu and a 5* healer besides Vivica and Iād certainly be willing to pay more for that if I have it to spend at the time.
Personally I am running 3x RT20, 1x HA08, and 1x HA10
Advanced House Level 10, and Advanced Farms, easily support this when you add in AR each month
HA10 AND 3x RT20
The biggest advantage with HA10 is no 5* are consumed
Even if you just care about HA10 limited availability 5* heroes, that is 1:1, with an average of 3 per 420 days of HA10 while Soul Exchange and 3x RT20 is also an average of 3.15 ( lower tier ) to 1.575 ( upper tier ) limited availability heroes per 420 days
If you consider it as success (non plain S1) and failure (plain S1) in a given number of attempts, you could calculate possible HA10 results as a binomial distribution.
There is not one answer, but many as you have a distribution of possible outcomes. A person could hypothetically complete 52 HA10 summons and get all successes, all failures, or any outcome between the extremes.
For example, 52 failures is roughly 0.4%, or 4 out of 1000 would fail on every attempt (if the distribution holds true). While those 4 people are not āeverybodyā, i wouldnāt be surprised if they completely devalue HA10.
From a personal standpoint, I ran HA10 for a year resulting in one S2 and one costume S1, both low-value heroes. Similar to the alliance that āwants my $1.99 backā, I āwant my year backā.
Soul exchange WILL give me a guaranteed hero of my choosing at the cost of feeders. I canāt begin to place a value on that, as it pertains to me.
The biggest advantage of Soul Exchange is you can choose a hero you want (with a pool of from 10 to 20), while in HA you get a random hero (Iāve got Grimble twice in for HA10 during last 1.5 year and one more non-S1 hero) . I think for F2P/C2P Soul Exchange is a perfect place to exchange S1 heroes dupes, that you probably never ascend for something more useful and releasing space in hero roster. My TC20 is producing 5* heroes (dupes) faster than HA10 can retrain it into something useful. I was thinking about stopping TC20 for some time, changed my mind lowing Soul Exchange.