12 replies to this topic

[Tim East]

How accurate is the thing? I ask, because I stumbled onto this today.

C-MACHINE GUN 1 290 292 100.69%

The only way I can think of for that to make sense is if machine gun hits can pass through one target into another?

I'll grant you that the sample size for that weapon is really low because I don't use MGs much since before the clans came out, but still. It struck me as really weird was all.

Edited by Tim East, 15 June 2015 - 07:25 PM.

[Bulletsponge0]

weapons stat page is very unreliable

[process]

I'd have to assume the stats page includes tachyons.

[mailin]

Do you have some magic bullets in your MG ammo? Those numbers don't make any sense at all.

[Rushin Roulette]

The only half way logical answer to those stats (apart from Tachyons) would be that The stats are not calculated by bullets hit/bullets fired*100, but the accuracy is somehow determined by damage dealt inrelation to the bullets fired.

This means that 3 factors will throw off the calculations;
1. Partial damage due to damage transfer to a section further in the mech (50% or 25% damage if you hit an arm/leg or a side torso being forwarded to the center torso).
2. Partial damage due to final bullet to a critically damaged and crucial section with less HP than the bullet potentially deals (pretty unlikely with MGs... but you never know)
3. Extra damage dealt by double or tripple criticals (highly likely with a high ROF weapon such as the MG)

Edited by Rushin Roulette, 16 June 2015 - 04:04 AM.

[Wildstreak]

I only look at stats to see how well I am doing in Mechs and that list is still mixed up regarding older chassis.

[Tarogato]

The weapon stats are broken. They number they display for "Matches", "Accuracy" and "Time Equipped" is highly inaccurate. I can only assume that the "Fired", "Hit", and "Damage" are also inaccurate, but I've studied my and other players weapon stats extensively, and I've found those numbers to be more reliable. (plus, the samples are much larger, so errors are more easily discounted)

If you want to gauge how well you are doing with a weapon, try this formula
Damage ÷ [ Shots Fired × Potential Damage Per Shot ]

For instance, look at my IS medium laser.

369,105 (damage I have done) ÷ [ 189,111 (number of times I've fired a Medium Laser) × 5 (amount of damage a medium should deal) ]

= 369,105 (damage I have done) ÷ 945,555 (total damage I could have potentially done)

= 0.390358044

= 39% efficiency

---

Note that the "potential damage per shot" is handled differently for many weapons. For an AC/20, it's "20". For a cERLL, it's "11". For a Clan SRM4, it's "8", for an Inner Sphere SRM4, it's "8.6". For LRMs, it's "1". For stuff like MGs and Flamers, I haven't bothered to figure it out.

I find it useful to compare like weapons. So I compare all my lasers to each other, and I compare all my autocannons to each other, and see where I'm weakest and strongest. It's all about dealing the maximum damage per shot. The closer you are to a final value of 1.0 (100%), the more efficient you are at using that weapon. You'll find that many of your numbers will range between 30% and 60%, and rarely above 60%. Usually people who are very careful with their gauss and AC/20 shots will have 75% with those weapons.

(oh, and by the way, the "Accuracy" shown on the stats page for lasers, is worthless. If you fire a medium laser and only graze an enemy mech with it, dealing 0.00001 damage, it still counts as a solid hit for the stats counter. So again... worthless. Use the formula. )

---

Yes, I already have a spreadsheet for this if you're really interested.

Edited by Tarogato, 16 June 2015 - 05:25 AM.

[JayKay17]

Stats page is - unfortunately - totally unreliable, at least the weapon and Mechs sections.
The stats page tells me, that I did in 1 match, that lasted 00:10:46, with small pulse lasers an exorbitant 9,354 points of damage.
Until now I didn't find a way to bend time or manipulate other laws of physics to have managed that feat, so it has to be the CODE.

[Roger That]

Yes, unreliable. I put a ticket in over a year ago about my stats just to see what would happen. Nothing happened.

[Tim East]

Rushin Roulette, on 16 June 2015 - 04:04 AM, said:

The only half way logical answer to those stats (apart from Tachyons) would be that The stats are not calculated by bullets hit/bullets fired*100, but the accuracy is somehow determined by damage dealt inrelation to the bullets fired.

This means that 3 factors will throw off the calculations;
1. Partial damage due to damage transfer to a section further in the mech (50% or 25% damage if you hit an arm/leg or a side torso being forwarded to the center torso).
2. Partial damage due to final bullet to a critically damaged and crucial section with less HP than the bullet potentially deals (pretty unlikely with MGs... but you never know)
3. Extra damage dealt by double or tripple criticals (highly likely with a high ROF weapon such as the MG)

But it specifically says fired 290, hit with 292.

Tarogato, on 16 June 2015 - 05:20 AM, said:

(oh, and by the way, the "Accuracy" shown on the stats page for lasers, is worthless. If you fire a medium laser and only graze an enemy mech with it, dealing 0.00001 damage, it still counts as a solid hit for the stats counter. So again... worthless. Use the formula. )

I wonder if it is possible to get laser hits to register more than once by striking multiple mechs with a single firing of the beams. That could be the source of this, though given the refresh rate on MGs I have no idea how I swept people that fast.

[JC Daxion]

Tim East, on 16 June 2015 - 09:05 AM, said:

But it specifically says fired 290, hit with 292.

Stats are totally borked.. and not counting right.. So it is not surprising that, it is counting more hits than amount fired.. I hope they track down the issue one day and fix it, but i think that is very low on the list of things to do.

[Wildstreak]

Well, I figure Stats Page fixes is down the middle to bottom of the To Do list.

[Stickjock]

Moving over to Website & Forum Feedback...

And have noticed the same issues as well, Pilot Module stats as well are not accurate.