View unanswered posts | View active topics It is currently October 17th, 2017, 7:35 pm



Reply to topic  [ 27 posts ]  Go to page Previous  1, 2
 SDE Spawner Tiers: Ranking Spawn Points in Classic Mode 
Author Message
Bottle Cap
User avatar

Joined: December 11th, 2015, 12:14 am
Posts: 7
iamfanboy wrote:
Each +1B adds 4/6 of a Star to an average roll, with the full +3B giving the average equivalent of a Green Star to each roll.


FYI only the average is the same, a three blue dice roll has more chance of rolling unreliably (and they often do)

Multiple blue dice, however, have the added bonus of greater likelyhood, greater total, and less dud-rolled hearts, illustrated in the second graph. Yes I am pretty bored at the moment.

Image Image


December 13th, 2015, 3:08 pm
Profile
Denizen
User avatar

Joined: March 15th, 2012, 7:26 am
Posts: 310
Teowulff wrote:
Interesting!!

You haven't rated Rock Pile yet .. and sometimes you use "medium" or "medium/high", in other cases numbers. Bramble Knight M/H means .. a 4? And Fungal Growth an "M" .. equals a 3?

Yeah, sorry about that; I made an editorial change halfway through adding that category.

I also moved the Bramble Knight from A to B; the Bramble Knight is a one-trick poodlepony, and the Grobbit's awkwardness to use inhibits its usefulness.


As far as the dice go, HexZyle...

It's hard to evaluate the usefulness of each dice. Yes, the blues are the worst statistically, but they're also essential, because without them you're relying too much on Heal potions. Reds are nice, but too many are overkill as you roll more Potions than you need.

And that's not even counting the odds they give. 50% of the time blues give no Stars at all, making them just so likely to whiff that you can't rely on them - I've rolled a whopping total of THREE Stars on a Gruesome George attack using 9 blues - but you also need them.

I thought about using more complicated formulae to evaluate 'em but couldn't really come up with anything that felt right and was easy to read like just using the averages. Maybe I'm just not that great of a mathematician. Let's do a raw comparison of 3B to 1G.

1G has 17% of whiffing, 67% of 2+, 33% of 3+, and a 16% of 4.

3 blues is hard odds to calculate, with 1/216 possibilities, and here's the raw numbers that compare after hand-counting all the dice throw combinations (yes, I spend half an hour doing it, be thankful!):

12.5% of 0 stars
25% of 1 star
29% of 2 stars
20% of 3 stars
10% of 4 stars
3% of 5 stars
.5% of 6 stars.

Judging from this, 3B has a better chance of not whiffing at all (12.5% to 17%), 13.5% of 4+ stars, 33.5% of 3+ stars, and 62.5% of 2+ stars.

Overall, the G has a slightly higher chance of rolling more consistently, but 4% ain't much of a much.

Frankly, the 1G to 3B comparison came out a lot closer than I thought it would, and the closeness of that means I don't have to calculate further odds to justify the use of averages in rating the dice.

_________________
Lootimeter Mode - A Fast, Fun, Fair Mode for Consul versus Heroes!
SDE Hero Tiers: Ranking Heroes in Classic Mode
SDE Boss Tiers: Ranking Bosses/Minibosses in Classic Mode
SDE Spawner Tiers: Ranking Spawn Points in Classic Mode


December 15th, 2015, 2:28 am
Profile
Bottle Cap
User avatar

Joined: December 11th, 2015, 12:14 am
Posts: 7
iamfanboy wrote:
1G has 17% of whiffing, 67% of 2+, 33% of 3+, and a 16% of 4.

3 blues is hard odds to calculate, with 1/216 possibilities, and here's the raw numbers that compare after hand-counting all the dice throw combinations (yes, I spend half an hour doing it, be thankful!):

12.5% of 0 stars
25% of 1 star
29% of 2 stars
20% of 3 stars
10% of 4 stars
3% of 5 stars
.5% of 6 stars.

Erm well, I already had represented that distribution of odds for the 3B in my first graph...

iamfanboy wrote:
Judging from this, 3B has a better chance of not whiffing at all (12.5% to 17%), 13.5% of 4+ stars, 33.5% of 3+ stars, and 62.5% of 2+ stars.

Overall, the G has a slightly higher chance of rolling more consistently, but 4% ain't much of a much.

Frankly, the 1G to 3B comparison came out a lot closer than I thought it would, and the closeness of that means I don't have to calculate further odds to justify the use of averages in rating the dice.


I doubt you'd worry about Green dice rolls on 1 star defense enemies, (and if you did, they both have an approximate hitrate of 66%, favoring the 3B with its 19.5% chance of at least one heart) unless you were really desperate for potions. Is it safe to say that most the rolls that matter are against enemies with at least defense 2? Since the average defense roll is usually 2-3B (1.33-2 stars), or a minimum of 2W anyway for Elites. So I guess the total chances for the rolls of a minimum value are very similar, with the Green having only a 4% improved roll rate, and the 3B only getting a 7% chance to roll a heart.

I love probability, I could discuss it all day. I certainly should consider constructing a graph that covers most dice rolls in most situations and analyze which are better than what.


December 16th, 2015, 12:34 pm
Profile
Denizen
User avatar

Joined: March 15th, 2012, 7:26 am
Posts: 310
Dude, when someone questions my math, I check it. Even when it's me.

And I started questioning my math when I read your post. I had to go back and check it to justify my use of the averages instead of more fancified formulae; I'm glad I did because it does show that the use of averages is 'good enough.' 3B to 1G is close enough on a more detailed inspection that I can't justify more time-intensive evaluations that could be used actually playing the game.

The main thing with creating a comprehensive evaluation of dice odds is how they add and interact with each other. By the end game, you're never rolling a 3B or a 1G in isolation, after all, it's usually 3B1R1G or 1W4B1R or similar combinations. Never was this more driven home to me when I was trying to evaluate Starfire's Hoard ability and figuring out the chances of whiff to success in drawing from the Treasure deck. Plus there's interactions with Hex and Bane to try and figure out in a more complicated model, and I said to heck with that.

Averages create a quick 'at a glance' judgement of the dice roll's value, especially late game when you're trying to figure out if you're wasting time attacking a Boss with regular attacks or should go to Special Attacks instead, or when you can safely Lockpick a Chest (important in Lootimeter, after all you don't get Keys!), and so on. As long as it comes out 'good enough' I'm happy.

_________________
Lootimeter Mode - A Fast, Fun, Fair Mode for Consul versus Heroes!
SDE Hero Tiers: Ranking Heroes in Classic Mode
SDE Boss Tiers: Ranking Bosses/Minibosses in Classic Mode
SDE Spawner Tiers: Ranking Spawn Points in Classic Mode


December 16th, 2015, 1:34 pm
Profile
Bottle Cap
User avatar

Joined: December 11th, 2015, 12:14 am
Posts: 7
iamfanboy wrote:
The main thing with creating a comprehensive evaluation of dice odds is how they add and interact with each other. By the end game, you're never rolling a 3B or a 1G in isolation, after all, it's usually 3B1R1G or 1W4B1R or similar combinations. Never was this more driven home to me when I was trying to evaluate Starfire's Hoard ability and figuring out the chances of whiff to success in drawing from the Treasure deck. Plus there's interactions with Hex and Bane to try and figure out in a more complicated model, and I said to heck with that.


You're completely right. I'll be sure to take these into account if I do end up making one. Especially with the Hex and Bane, because they're a complete overhaul of a dice's probabilities since it can affect any dice in the roll depending on the roll itself, not just one isolated dice.


December 17th, 2015, 12:02 pm
Profile
Bottle Cap

Joined: May 13th, 2017, 10:21 am
Posts: 2
Totally disagree with you about the shallow graves. I was playing Arena with the graves, and my opppnent was playing with Trent. The shallow grave ended up massacring Trent. I personally disagree because of the necromancer and bone pile ability, your rarely have to respawn. As long as you do it right, you could kill the heroes at the first spawning point.


May 13th, 2017, 10:39 am
Profile
Bottle Cap
User avatar

Joined: April 9th, 2016, 12:28 am
Posts: 10
Location: Australia
You may need to re-visit these spawn points since card updates in 2.0

I would put your rocktop spawner close to SS tier.
Shells that block line of sight and movement. Arm 2 and 2 HP with no reward for killing them... Means you use your throw effect to move your models for extra movement not for damage.

That and the crusher has no hiding..... Turtle-mancer? Those shells just became monsters that deal damage.


September 13th, 2017, 2:05 pm
Profile
Display posts from previous:  Sort by  
Reply to topic   [ 27 posts ]  Go to page Previous  1, 2

Who is online

Users browsing this forum: akai, Bing [Bot], Danteire and 4 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group
Designed by STSoftware.