In HS, it seems that effects/stats does not proportionately scales to mana cost.
Taking some examples:
Classic: Mad bomber (2/3/2) BC: deal 3 dmg randomly split between all other char.
WOTOG: Evolve (1) : Transform your minions into random minions that cost (1) more.
SoU: Puzzle Box of Yogg-Saron (10) : Cast 10 random spells (targets chosen randomly) .
However, as I look at some RNG focus cards, I wonder…
How to Dev decide how much RNG is allowed at each mana cost?
Its chosen at random I guess
You also have to take into account the total mana cost.
At low mana costs, the amount of randomness that can be introduced is pretty limited. But when you get close to or up to 10 Mana, the cards need to include a much higher potential upper bound to account for taking your entire turn and needing to survive long enough to play them (or draw something to cheat them out a bit earlier).
So, it’s not a linear effect deciding how big the random effect produced by each mana cost is.
EDIT:
To put it another way, the average or the mean expected effect from a random spell or battlecry is likely to increase in a fairly linear manner with mana cost. But the variance is going to increase a huge amount to make it worth it to players to use random cards with high mana costs. So, the closer you get to 10 mana, the higher the top end of the possible variance is likely to be.
The formula for RNG Stat Budget is:
1 Mana
==
1 Mana + (Patience * ((Luck + Karma) - (BackgroundVideo * Music)) * (Skill * .10))
RNG is simple:
You lag and you lose
Your game crash and you lose before reconnecting
Your mobile device runs out of battery and you lose
You hope for a fix about the above issues when patching the game and… gotta wait till “next” patch
You wait for weeks for an answer from tech support
You try to plan the odds of getting X-card in your deck on your next turn and you’ll eventually be back at step 1.
They just print whatever they think is interesting. There’s really no correlation between the rng & mana cost.
While I can’t give an answer to that, there’s one thing we know for sure :
If we compare 2 similar effects, the cheapest is (most of the time) the most random.
I don’t think there’s an amount of rng allowed for each mana cost. I think the effect is decided, then they choose the price depending (among other things) on the rng of the effect.
They target an expected power level, that on average wouldn’t swing games too much.
Yogg 1.0 was above their target, which is what led to its dominance in tournament play and subsequent nerf. They wanted it to be weaker than it ended up being, a nonserious showstopper for the Timmys, not the Spikes.
Actually opening the question to all.
So, how to gauge a expected power level?
The current mana cap is 10mana. As you can see, since higher mana cost can mean more stats/effects, then where is the line?
I do not expect an answer since these are very dependent on Dev, but interested to see other poster’s having any deeper views of the topic. (e.g. effects on future designs)
Average of outcomes, weighted by chance of ocurring.
It requires testing to confirm.
Same process as quantum mechanics, actually, which despite being inherently random is actually quite predictable for a large number of events.
I do understand the concept. However, as we ‘know’ that the limited testing capability that they have, how do they determine?
or does the ‘strategy’ to implement, monitor, and nerf after expansion applies, that removes the burden to focus too much on that expected power level?
At that point you’re out of the math of how you generate expected power level and into the game design. They have an uncertainty on everything they do: they can’t predict what we’ll come up with with perfect accuracy, so they do their best and will readjust later.
They run simulations. Gather the results and make adjustments after launch.
For the most part they can gauge the power level of a card by how much it takes to remove it, how much potential damage it can cause, additional effects like draw, discover, etc…
I question this back to my question.
If we take a few examples, we can see where my question comes from.
DK Guldan: Zul’jin : both 10mana and have ability to generate an incredible amount of value. We can estimate the amount of power it can generate, yet as we move forward to future expansion, what will be the bar?
Noting, these cards permanently impact WILD and a potential baseline to future designs.
While it make no sense to limit designs to peg stats/effects to mana cost, on the other hand it makes no sense to game balance if there is no some form of design control?
and if nerf thereafter is the solution, then it ease the responsibility to balance the design on the onset?
There are so many ways to get the same result though. Really it comes down to assigning a range of values to represent each tier or mana cost if you will.
So let’s keep it simple for now and use attack/health as our only values.
0 Cost = 1/1 or 0/2
1 Cost = 2/2, 1/3, 3/1 or 0/4
continued down the chain.
Once we have that base value we can then assign a value to a new mechanic and increase the cost according to the exchange value.
Taunt costs 1HP or 1ATK or 1LP or 1 Mana to attach to a selected card, let’s say.
2/2 with Taunt could then cost 2 Mana
These are arbitrary numbers mind you but at it’s essence that’s the idea. Adjustments to the values are made based on the extra benefit. The simplest card that showed the idea was 10 Mana Pyro-Blast, IMO. From there you just move it up and down as it effects the meta. That’s where the sims come in.
Edit: Wild is not something that can really be included in the equation much. They can HoF cards and apply nerfs easy enough though.
What you refers to does does represents an “arbitrary” baseline. However I am questioning the random elements.
We can clearly see it applies on Mad Bomber, where the 2mana 3/2 + 3dmg random effect. As we move on to Yogg and Yogg Box, both at 10mana where the former provides an extra body.
At lower mana cost, it easier to relate. But at higher mana cost, it is harder to relate as to draw the line where is that design control.