Currently, there are more than 120 playable characters in this game. They are distributed between five tiers for now. For each tier, we can use different types of tokens to recruit them, even some characters have been removed. I think each and every character should be recruitable with tokens.
I think it is time to readjust the odds in the different tokens types in a way to take account of the increasing roster of the game.
Standard tokens
Those are more useful for beginners. Actually, they can provide characters of the first three tiers of the game. I think, it will be better off they provide the player with only the first two tiers I think the odds should be rebalance as follow:
- 60% of chance to pull any

character (so 7.5% to pull a specific character). - 40% of chance to pull a


(so 2.85% to pull a specific character).
Heroic tokens
Actually, the provide characters from three tiers(

to



). It could be interesting to reduce it to 2 tiers. And the distribution could be : - 60% of chance for any


(so 4.285% for a specific character) - 40% of chance for any



(so 0.95% for a sepcific character.
Elite tokens
Now, they give

and


. It could be interesting if they only provide


and



with the following odds: - 75% of chance to pull a



(so 1.785% to pull a specific character). - 25% of chance to pull a




(so 0.568% to pull a specific character).
Legendary tokens
Currently, they can be used to pull the three latest




and any



. The new odds could be : - 60% of chance to pull a




(so 1.33% for a specific one) - 20% of chance to pull a cover for the 3 latest





(so 6.67% for a specific one) - 20% of chance for the oldest ones





(2.22% for a specific one).
I think those changes in the tokens could help each type of players to strengthen their roster more easily and be more inclined to invest in roster slots also.
An interesting feature could be an exchange system between tokens for an exchange rate. It could be as follows: - 5

to exchange 5
for 1
- 10

to exchange 5
for 1
- 25

to exchange 5
for 1
.