Jump to content
You need to play a total of 10 battles to post in this section.
Pseudoscope

% fire chance

17 comments in this topic

Recommended Posts

Beta Tester
45 posts
6,573 battles

hi Devs,

 

Just had a round in my Fletcher firing he. 143 hits and 3 fires. The stats say I have an 8% chance of setting a fire, why is it I only get a 2% chance?

 

And yes I realise you have a stupid damage formula that takes into account a lot of factors other than just that. But what is the point of stating there is an 8% chance of a fire? This number means nothing and is miss leading. Also I realise that they might be running fire surpression but still these numbers are way off.

 

Can we please get some meaningful stats? / reworks the %fire

 

Regards 

Pseudo

Share this post


Link to post
Share on other sites
Senior Moderator
3,837 posts
2,602 battles

That's not how RNG works. It doesn't retain memory of previous results. Lets say the "final" probability of starting a fire is 1%, it doesn't mean one fire will start every 100 hits. It means that each hit has a 1% chance of starting a fire. So it's possible to get 1000 hits or more without ever starting a fire. Likewise, if you have 99% chance of starting a fire, it doesn't mean you can never have 2 hits in a row that doesn't start a fire, but the probability of this happening is extremely low.

 

Furthermore, you can't start a fire on a section that is already on fire. So some of your total hits may have been on sections that were already on fire.

Share this post


Link to post
Share on other sites
Beta Tester
45 posts
6,573 battles

I understand how the fire mechanic works, the places fires can be set, I've even seen the formula that takes into consideration the ship resistance and the 'fire suppression' perk. I also have a solid knowledge on statistics, as such probability and chance (or RNG) is something I understand fully. Statistics is the science of data. 

 

8% means approximately 2 out of every 25 shells should start a fire. Now I understand that some shells may have hit the belt armor and that 'they might have been on fire' (they weren't). But giving the high shell arcs of the fletcher it is unlikely that they landed anywhere but the deck. I have no problem with the fire% being so low, my problem is the misleading stats.

 

It is very unlikely if I had an 8% fire chance I would only set 2% worth of fires, especially over a sample of 143 hits, this sample size should have netted me ~11 fires. The fact I only got 2 isn't because of RNG or 'unlucky rolls'. It's because the quoted stats are wrong or at least, misleading.

Share this post


Link to post
Share on other sites
Member
892 posts
7,853 battles

I would think the formula is based on every shell having a x% to set fire instead of x% of shells fired setting fires. Case in point, I can fire my first salvo and still set fire to the target

Share this post


Link to post
Share on other sites
Senior Moderator
4,798 posts
1,924 battles

I would think the formula is based on every shell having a x% to set fire instead of x% of shells fired setting fires. Case in point, I can fire my first salvo and still set fire to the target

 

pretty much this.  Also each ship/class has a modifier on the chance of being set alight.  (google it)

Share this post


Link to post
Share on other sites
Member
892 posts
7,853 battles

 

pretty much this.  Also each ship/class has a modifier on the chance of being set alight.  (google it)

 

Also, where the shells land will be part of a modifier

Share this post


Link to post
Share on other sites
Beta Tester
45 posts
6,573 battles

I would think the formula is based on every shell having a x% to set fire instead of x% of shells fired setting fires. Case in point, I can fire my first salvo and still set fire to the target

 

You don't say? Have you ever done coding before? have you seen the formula they use? 

 

 

pretty much this.  Also each ship/class has a modifier on the chance of being set alight.  (google it)

 

Yeah I know, if you read the post it says I know about the fancy formula, which means I know about the inputs and variables.

 

Actually, 143 hits is too small a sample size. It's not enough to prove that the stats are wrong.

 

I never said the stats are wrong. I said they are misleading. And seems you've got soo much spare time on your hands, please do go and do a statistical analysis on all the possible encounters and come back to me with some hard evidence. I'd love to see it. 

 

Edited by Pseudoscope

Share this post


Link to post
Share on other sites
2
[BOTES]
Member
94 posts
3,501 battles

 

I never said the stats are wrong. I said they are misleading. And seems you've got soo much spare time on your hands, please do go and do a statistical analysis on all the possible encounters and come back to me with some hard evidence. I'd love to see it. 

 

 

Burden of proof is on you matey. Additionally by what everyone is saying in here the chart of fire % would be some multiple hundreds of entries when accounting for (firing ship) vs (target ship) + [damage location] x {multipliers via captains skills, witherer flag, arsonist flag}

Share this post


Link to post
Share on other sites
Alpha Tester
210 posts
4,306 battles

The way i understand how fire chance works is:

 

Ship's base anti-fire coefficient x (1 - Damage Control Mod 1 - Fire Prevention) x (Shell's base chance to set fire + Demolition Expert),

Ship anti-fire coefficient being 1.0 at tiers 1 & 2, and 0.5 at tier 10, also changing Damage Control Mod and Fire Prevention %'s into usable figures, 0.05 and 0.07 respectively. Not taking signal flags into consideration


So, to use the Fletcher as an example, and if the opposing ship has the Damage Control mod and FIre Prevention skill  (targets class doesnt seem to affect it, just the tier):
On tier 7's = 0.7 
x (1 - .05 - .07) x (5 + 3) =4.9% chance

On tier 8's = 0.63 x (1 - .05 - .07) x (5 + 3) = 4.4% chance

On tier 9's = 0.57 x (1 - .05 - .07) x (5 + 3) = 4% chance

On tier X's = 0.5 x (1 - .05 - .07) x (5 + 3) = 3.5% chance

 

Thats just the way i understand it working, because lets face it, if its flat values, 8% chance of fire - 12% (Fp + DCM) giving a -4% chance, that would be just plain stupid lol

 

Either way, we cant get accurate stats on the stat cards, because the ship youre shooting at changes the values individually by tier and whether or not it has the modification and/or the Captain skill.

Edited by Cyanide7662

Share this post


Link to post
Share on other sites
2
[BOTES]
Member
94 posts
3,501 battles

Really. I had been told BB HE had the highest chance of setting a fire (due to AP being more appropriate almost always)

Share this post


Link to post
Share on other sites
Alpha Tester
210 posts
4,306 battles

When i said class doesnt affect it, i meant the class you are shooting at, not what class you are playing, but i rephrased it if it makes you feel any better.

Edited by Cyanide7662

Share this post


Link to post
Share on other sites
Video Contributor
2,261 posts
10,782 battles
8 hours ago, hirokie said:

just had a battle with my takao, 21% fire chances and i hit tirpitz 28 times in 4 salvos somehow still no fire. *sigh 

Read this

Quote

[Fire Resistance Coefficient] x ( 1 - [Damage Control Modification 1] - [Fire Prevention] ) x ( [Projectile Base Fire Chance] + [Demolition Expert] + [Signals] - [Inertial Fuse for HE Shells]) = Fire Chance

Example

Quote

As an example, we can compare the chance of a single Benson high explosive shell lighting a fire on an enemy Bismarck. The Bismarck is using Hull (B) and Damage Control Modification 1, but her commander does not have Fire Prevention. The Benson commander has Demolition Expert (but not Inertia Fuse for HE Shells) and is flying both the Victor Lima and India X-Ray signals. The fire chance would be:

(0.6337) x (1 - 0.05 - 0.00) x (0.05 + 0.02 + 0.005 + 0.005 - 0.00) = 4.82% fire chance


If the Bismarck commander acquires the Fire Prevention skill, the chance would then be:

(0.6337) x (1 - 0.05 - 0.10) x (0.05 + 0.02 + 0.005 + 0.005 - 0.00) = 4.31% fire chance


If the Benson commander didn't have Demolition Expert or the signals equipped, the chance would be:

(0.6337) x (1 - 0.05 - 0.10) x (0.05 + 0.00 + 0.00 - 0.00) = 2.69% fire chance

Or you have to ask RNG if you shot 28 shells to Derpitz and no fire

Share this post


Link to post
Share on other sites
Member
199 posts
18,023 battles

Don't try to understand it , don't try to clarify it . Just play the game and accept the result . WOWS don't give a [content removed] .

 

Profanity. Post edited, user already sanctioned.

~amade

Edited by amade

Share this post


Link to post
Share on other sites

×