Jump to content

Statistics

Feedback v1.2.197

  • You cannot reply to this topic
77 replies to this topic

#61 jay35

    Member

  • PipPipPipPipPipPipPipPip
  • The Hammer
  • The Hammer
  • 1,597 posts

Posted 08 March 2013 - 08:37 AM

Eh, I've lost 100 basis points on my KDR since these expanded stats when live with the last patch, but it's not a big deal. When you're busy experimenting with new builds you expect your stats to suffer a bit.

The funny thing about the Stats, though, is they currently present a very inaccurate picture since they don't go back to when I started playing.

They also don't appear to account for Assists, which is a significant part of team-based gameplay.

Is it true there is a stats wipe still forthcoming in the next few weeks? I hope so. I'd like to see the stats start fresh.

Edited by jay35, 08 March 2013 - 08:43 AM.


#62 jay35

    Member

  • PipPipPipPipPipPipPipPip
  • The Hammer
  • The Hammer
  • 1,597 posts

Posted 08 March 2013 - 08:40 AM

View PostXendojo, on 08 March 2013 - 04:57 AM, said:

Great...now people can judge your skill completely out of context.

Yep, and if they do something foolish like make them publicly viewable, without the player's ability to control whether their stats are public or private, that will immediately start to impact gameplay, as players who care about stats will be less likely to engage the enemy. You thought R&R was bad when it came to discouraging combat, wait 'til you see stat-lovers afraid to engage, disconnect, or do other strange things to avoid hurting their KDR.

The simple solution is to do like most products (Steam is a good example), and ensure the player can control the viewability of their personal stats/profile.

#63 Mad Elf

    Member

  • PipPipPipPipPip
  • 181 posts
  • LocationGlasgow

Posted 08 March 2013 - 10:47 AM

OK, bit of a bug here:

Weapon: TAG
Matches: 25
Fired: 4
Hit: 199
Accuracy: 4,975.00%
Time Equip: 01:57:31
Damage: 0

I have TAG in a "locked permanently on" group (for the moment, until the devs implement a better way), so I think I can see what's going on: I push the button once, holding it down, and the weapon cycles many times.

This means that the "Hit" and "Accuracy" stats for TAG are completely meaningless. Someone suggested a "time on target" stat for lasers? Might be worth while here.


However, it does raise a question: if I hold down the button that fires my MLs, my ACs, or my LRMs, they will fire continuously until I let go. Does this count as a "single firing" for them as well?

If so, the "Hit" and "Accuracy" stats are completely meaningless for all weapons. I will test this later...


EDIT - TEST RESULTS


No, holding down the fire button on a MPLAS counts as firing it several times.

However, I had it in the same group as my TAG for this experiment, and it didn't log being fired at all -- although it hit several times more.


So, most weapons' stats are OK. TAG's are broken, and inappropriate for the weapon anyway.

Edited by Mad Elf, 08 March 2013 - 04:32 PM.


#64 Segoris

    Member

  • PipPip
  • 34 posts

Posted 08 March 2013 - 04:53 PM

Re: public stats - So what? This is not changing much of anything as people are going to play selfishly and try to hide, use others as shields, kill steal, whatever else regardless. It has been done long before any public stats were available, and will be done long after we're all done playing games. If someone judges you based on stats and your stats are not up to what are being sought after, then the person judging is probably not worth playing with for one's play style preference.
Additionally, for those waiting for the last second to kill steal, this will easily be reflected in their own stats when you they have 50 kills but only 500 damage (hopefully assists are added since that would also be telling to see 50 kills, 500 damage, but only 5 assists).


View PostSeraphims Blood, on 08 March 2013 - 01:13 AM, said:

Yes, you could, divide total damage by number of matches ... but this seems like a lot of work every time you enter the stats page. I'm not saying that total damage it TOTALLY useless, it's just that the more you play this number will be getting bigger and bigger not really serving any purpose. Right now its OK, because we can check how stats are calculated (privilege of beta testers :P), but after all stats will be polished I don't see any valuable information in this statistic (just like mech stat > xp earn)

Why? Average damage = total damage / number of matches.


And I support having average, just don't remove total is my concern. Large numbers or not, I would rather have that number than anything else since I prefer being able to figure out things like average per kill (hopefully per assist soon enough), damage per hit, etc., but I also want a stat for anything which gives xp/cbills to better evaluate where I can improve. That is a main reason why I want raw data instead of processed. That, and we can see if weapons are acting as they should if crits, ammo explosions, and whatever else is split out (doubtful, but would be nice).

I enjoy raw stats more, though having averages calculated for commonly looked at stats (damage being the main one) would be welcomed.

View PostSeraphims Blood, on 08 March 2013 - 01:43 AM, said:

also, components destroyed on mech (not on enemy mechs) would be nice to have for each mech


Yes please.

#65 Segoris

    Member

  • PipPip
  • 34 posts

Posted 08 March 2013 - 05:04 PM

View PostSeraphims Blood, on 08 March 2013 - 01:43 AM, said:

also, components destroyed on mech (not on enemy mechs) would be nice to have for each mech


Since we really do want a lot of the same things, I drew up a quick diagram (which will probably format horrible when posted, but hopefully it comes through okay). Think I got it all

What you say you want:
MECHS: matches || wins || losses || ratio || max dmg || avg dmg || avg last 20 || avg cbills
-- weapons split by mech as a sub category here
WEAPONS: matches || accuracy || max dmg || avg dmg || #matches equip || overall equip %

What I am saying I want (bit more excessive, admittedly and I actually found I wanted more while drawing this diagram sadly):
MECHS: matches || Win || loss || ratio || total XP || avg XP || max XP || cbills || capture || assists/spotting/components/tag/narc/whatever xp (tag/narc maybe moved to weapons sub-category, AMS usage as well would be nice).
-- weapons split by mechs as a sub category here, and I would move "overall equip %" to this category
-- [new request] engines by death per mech. I would love to see every time I am dealt an XL engine explosion to see how my survivability with real stats
-- [new request] filters by wins/losses, so you can see damage/xp/cbills by win/loss, though if this was ever put in I would also want time played to see time per win/loss. While time is not very telling, it would just be welcomed

WEAPONS: matches || shot || hit || accuracy || total dmg || avg dmg || max dmg || kill shots

I would be happy if time equipped stayed, but that is the first thing which should be removed imo. Average last 20 would be cool, but I would not care either way for it as it will be obvious if I am trending up, down, or stagnating as my stats move or not.

#66 Enigmos

    Member

  • PipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 1,290 posts
  • LocationPhiladelphia

Posted 09 March 2013 - 06:58 AM

New Stats system: In my founder's Cat I had eight matches last night, but only seven appear today after playing another match. Is the stat system losing data somehow?

Otherwise the new stat system looks much better, more complete. It just also needs to be accurate.

Edited by OriginalTibs, 09 March 2013 - 06:59 AM.


#67 Snib

    Member

  • PipPipPipPipPipPipPip
  • 689 posts

Posted 09 March 2013 - 03:19 PM

View PostXendojo, on 08 March 2013 - 06:20 AM, said:


Soon enough they will be public. Did you read the rest of my post?


Oh are they? Fair enough. Take it as a compliment then that somebody is interested enough in you to look up your stats.

I don't disagree with the rest of your post, but for me it is even more simple. Given all the instability, bugs, imbalance and netcode/latency issues the stats cannot be compared anyway, which is also why i'm watching the Elo attempts with a raised eyebrow.

#68 Butane9000

    Member

  • PipPipPipPipPipPipPipPipPip
  • Elite Founder
  • 2,788 posts
  • LocationGeorgia

Posted 10 March 2013 - 03:00 PM

Can you allow us to sort the columns out from highest to lowest?

#69 AegisKay

    Member

  • PipPip
  • Bad Company
  • Bad Company
  • 26 posts

Posted 11 March 2013 - 04:39 AM

What's the big deal about public stats? I know some people see it as an opportunity to boost their k/d or whatever but what about those people who genuinely want to improve? Say I was in the top 1000 and wanted to get into the top 500, there'd be a trend in that top 500 that would suggest where I'd need to improve to get there. Maybe my accuracy is really high so I think I should be there but my deaths are much higher than the average. I then learn I need to die less to get better.

#70 BloodyDziq

    Member

  • PipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 159 posts
  • LocationPoland

Posted 11 March 2013 - 06:49 AM

@Segoris

Yes, I basically agree :)

Would be nice to incorporate average damage taken to a component (body part) with component being destroyed and engine on the mech during match.

I would see it something like this in the mech stats:

Centurion CN9-A ..........................(overal stats)
- weapon stats:
---- medium laser ..........................(stats)
---- srm 6 ......................................(stats)
- engine and components
---- XL 250 (engine dedicated stats: matches || engine destroyed || avg speed || % of max speed in match)
------ (no need to list all components here you all get the idea)
------ left torso(component dedicated stats: destruction ||avg damage || max damage || avg friendly fire damage || max friendly fire damage)


What do you all think about this? Would this information be of any use to you?

#71 Luxan

    Member

  • PipPip
  • Bad Company
  • Bad Company
  • 41 posts

Posted 11 March 2013 - 07:33 AM

Looking at my stats, I see that LRMs are doing less damage than advertised (about 1.45 / missile instead of 1.8) and SRMs are doing more damage than advertised (about 3 / missile instead of 2.5).

All of my energy and ballistic weapons are showing less damage than advertised, but this is expected. I sometimes fire them outside of optimal range and don't always keep the beam weapons on target for their full duration.

But I would expect missiles to either do full damage or none. They hit or they don't, right? And how are SRMs doing more damage than they should?

Has anyone else noticed this in their stats? Is this an indication of a missile bug or bugs?

#72 Buckminster

    Member

  • PipPipPipPipPipPipPipPip
  • Elite Founder
  • 1,577 posts
  • LocationBaltimore, MD

Posted 11 March 2013 - 07:47 AM

I think missiles have a splash damage component to them. The average damage per LRM is lower because you probably have missiles that are only partially hitting the mech, maybe doing 1 damage to a leg rather than the 1.8 of a solid hit.

As for SRMs, I think crits and such are included in the damage, so if your SRM takes out a component it is considered as extra damage. I've noticed the same thing with my PPCs - my ERPPCs have an average damage of 10.3, and have average as high as 11.

#73 Kageru Ikazuchi

    Member

  • PipPipPipPipPipPipPipPip
  • The Determined
  • The Determined
  • 1,190 posts

Posted 11 March 2013 - 08:53 PM

I think there needs to be, in addition to the accuracy %, an effectiveness rating ... for example:

- My Medium Laser Stats: Shots: 4,842; Hits: 4,473; Accuracy: 92.38%; Damage: 12,248 ... WOW! 92.38% accuracy, I must be an expert with these! :D

- Effectiveness would use this formula: Total Damage/(Shots*Max Weapon Damage) ... in my case for MLs: 50.59% ... Oh ... I really need to work on keeping my shots on target ... :rolleyes:

This would give us a better way to judge which weapons we use most effectively without using another system and keeping our own records of our stats over time (which I will be doing anyway). :ph34r:

#74 Alik Kerensky

    Member

  • PipPipPipPipPipPip
  • Little Helper
  • Little Helper
  • 406 posts
  • LocationSeattle

Posted 12 March 2013 - 12:42 AM

Kill Assist Stats would be nice.... especially for non-cheese lights. :rolleyes:

#75 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 12 March 2013 - 11:46 PM

View PostMal, on 05 March 2013 - 09:41 PM, said:



That's what I thought... thanks for the confirmation. Not sure that actually makes any sense though, since the number of missiles that hit out of a volley (especially for LRMs)... is pretty much out of our control, even if we have a good solid lock... guess I'll just expect that one to be fairly low most of the time.


edit: Just noticed that on the 'Mech stats tab... the ' is on the wrong side of Mech... it says Mech' instead of 'Mech.. oops.

I find such information more useful, since it tells you how much of a weapon's damage potential is actually reached.

If all players hit generally only with 20 % of their missiles (I figure it'S better, but just saying if they did), then we know that the theoretical DPS of the SRMs is far off from the real damage output. So if you know where to find out that you actually hit with 80 % of all LB10-X AC pellets, you knew that maybe the extra weight investment compared to the SRM6 is worth it.
(But of course, these numbers are just hypothetical.)

In fact, Lasers counting as a hit if just a single tick hits the enemy, even if most of the damage is wasted, might not be so helpful to know. (Luckily there is the damage done statistic as well, together with shots fired and the known damage per shot value, you can figure your actual hit rate or "damage efficiency" yourself.)

Edited by MustrumRidcully, 12 March 2013 - 11:47 PM.


#76 BloodyDziq

    Member

  • PipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 159 posts
  • LocationPoland

Posted 13 March 2013 - 12:46 AM

some kind of sorting would be nice, not saying for every stat and all, but some basic: mechs by tonnage and weapons by groups (energy, ballistic, missile) and types (don't mix regular lasers with pulse lasers)

#77 Mal

    Member

  • PipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 995 posts

Posted 13 March 2013 - 06:00 AM

View PostMustrumRidcully, on 12 March 2013 - 11:46 PM, said:

I find such information more useful, since it tells you how much of a weapon's damage potential is actually reached.

If all players hit generally only with 20 % of their missiles (I figure it'S better, but just saying if they did), then we know that the theoretical DPS of the SRMs is far off from the real damage output. So if you know where to find out that you actually hit with 80 % of all LB10-X AC pellets, you knew that maybe the extra weight investment compared to the SRM6 is worth it.
(But of course, these numbers are just hypothetical.)

In fact, Lasers counting as a hit if just a single tick hits the enemy, even if most of the damage is wasted, might not be so helpful to know. (Luckily there is the damage done statistic as well, together with shots fired and the known damage per shot value, you can figure your actual hit rate or "damage efficiency" yourself.)



The addition of another column of stats could easily address that though. They could just have an "average damage per shot" tacked on to the end. That would easily give the information for anything that doesn't do all it's damage at once (missiles, lasers, hits from beyond 'maximum range', etc).

#78 Warge

    Member

  • PipPipPipPipPipPipPipPip
  • 1,027 posts
  • LocationKiyiv

Posted 18 March 2013 - 07:00 PM

I'd like to see in "Weapon stats" average and max kills by weapon type per match.
As for me stat "Hit" is useless, cause we have "Accuracy".
Stat "Fired" don't give any picture, imho. 'cause on most of mechs there are more than 1 type of each weapon loaded.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users