Jump to content

- - - - -

The Base Statistics Page.


8 replies to this topic

#1 TooDumbToQuit

    Member

  • PipPipPipPipPipPipPipPip
  • The Death Wish
  • 1,539 posts

Posted 30 April 2016 - 11:01 AM

I know I should keep everything in one thread but I thought this was semi-important for new guys. When I found this page https://mwomercs.com/profile/stats it helped me see things a bit better.

It has a number of interesting areas and more of a few that stuck out.

One was Weapons and my accuracy with them(or lack of). I saw that I was only hitting 10% of my Narc targets. I was surprised to see how poorly I'm doing with missiles. (this Mech came with 2 large LRM launchers).

My plan was to go with short streak missiles because they "never miss". Either yes, they do miss a lot or maybe the target went out of range often but the very low numbers were a shock.

#2 Spike Brave

    Member

  • PipPipPipPipPipPipPip
  • The Solitary
  • The Solitary
  • 695 posts
  • LocationIn your base, killing your dudes

Posted 30 April 2016 - 11:13 AM

Streaks also take the most direct path to the target. Lot of times, they'll hit cover or another mech that is on that path.

#3 Rogue Jedi

    Member

  • PipPipPipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 4,908 posts
  • LocationSuffolk, England

Posted 30 April 2016 - 11:47 AM

View PostLikeUntoGod, on 30 April 2016 - 11:01 AM, said:

I know I should keep everything in one thread but I thought this was semi-important for new guys. When I found this page https://mwomercs.com/profile/stats it helped me see things a bit better.

It has a number of interesting areas and more of a few that stuck out.

One was Weapons and my accuracy with them(or lack of). I saw that I was only hitting 10% of my Narc targets. I was surprised to see how poorly I'm doing with missiles. (this Mech came with 2 large LRM launchers).

My plan was to go with short streak missiles because they "never miss". Either yes, they do miss a lot or maybe the target went out of range often but the very low numbers were a shock.

do not misunderstand, with regard to the missiles that is the percentage of the missiles which miss, if you fire an LRM20 even if you hit against smaller Mechs some of the missiles will miss wide due to spread, so even if every salvo you ever fire hits you are unlikely to get much past 80% accuracy, if you are firing paired LRM20s that gets even worse as they spread over twice the area of a single LRM20, to the point that with paired LRM20s close to half your missiles are likely to miss.

with regards to the streaks, they do not always hit, they just do not fire without a lock, there is nothing to stop you getting a lock on a target 800m away and firing the streaks, in which case you are firing far past the max range, also AMS can shoot them down or the enemy can put something between them and the missiles, or you could hit a destroyed component for reduced damage.

the numbers for Lasers also tend to be misleading, if you check your laser hit percentage that should be quite high, but if you then check the damage done and check the total number of shots you have made those numbers will not add up, you could have 90% accuracy with Mediun Lasers but you have fired 10,000 shots at 90% accuracy and you hit for the full duration that would be 45,000 damage however you would likely find yourself having done 20-30 thousand because you will usually miss with part of the lasers duration, but if you hit with even the smallest portion of the beam burn time that counts as a hit.

#4 Khereg

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 919 posts
  • LocationDenver, CO

Posted 02 May 2016 - 07:10 AM

Some of the stat pages have some inaccurate data (e.g. on my Weapons stats, for some weapons I show an unrealistic number of shots fired for the number of matches I've played, which seems to indicate the games played counter is broken).

However, there is a wealth of data there to help a player understand where he/she is doing well and where they need to improve. I highly recommend going a little stats nerd on it now and again and dumping it into a spreadsheet for geek analysis fun time.

Also, the stats page only records data from regular queue, not Faction Play. Just something to keep in mind when you're looking things over.

Edited by Khereg, 02 May 2016 - 07:11 AM.


#5 D V Devnull

    Member

  • PipPipPipPipPipPipPipPipPip
  • Little Helper
  • Little Helper
  • 4,245 posts
  • LocationUnknown... Except for the stars, it's kind of dark here!

Posted 02 May 2016 - 08:06 AM

View PostKhereg, on 02 May 2016 - 07:10 AM, said:

Also, the stats page only records data from regular queue, not Faction Play. Just something to keep in mind when you're looking things over.

Unless PGI changed something in the last two hotfixes since the "FW, Phase 3" release, that is no longer correct. I discovered my Quick Play stats were being polluted with FW activity. PGI should have provided a second page for the FW stuff. :(

~D. V. "I got caught the wrong way by this." Devnull

#6 Morggo

    Member

  • PipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 670 posts
  • LocationCharlotte, NC, USA

Posted 02 May 2016 - 08:44 AM

I also do the spreadsheet analysis.
Usually grab a copy of my stats pages for mech and weapons end of each weekend and monthend.
I'll not only compare changes over last period to look for trends (both good and bad) but just to help know which mech's I'm piloting better in general (this will often surprise you).

I agree some metrics are a bit wonky, or need a bit more refinement to be better used... especially lasers as mentioned above. I like to calc my ratio of true damage per shot hitting and compare that to the "perfect" damage for the weapon. That way you also know the average damage per burn you are actually delivering. Once I started tracking this particular stat for lasers it caused be to pay more attention to holding and burning longer.. so far I've greatly improved my ERLarge and Medium lasers. :)



#7 Koniving

    Welcoming Committee

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • The Guide
  • The Guide
  • 23,384 posts

Posted 02 May 2016 - 09:04 AM

View PostLikeUntoGod, on 30 April 2016 - 11:01 AM, said:

My plan was to go with short streak missiles because they "never miss". Either yes, they do miss a lot or maybe the target went out of range often but the very low numbers were a shock.


"How Streaks don't miss."


They defy all laws of physics.

How they used to work in closed beta.

When they had the laws of physics.

Thing is, the lore of the Streak is that it will only fire when its dedicated computer system, sensors, etc. take all the data of the surrounded area and plot out a path with a 100% chance of hitting the target, and if there was even a chance the enemy could evade it or if the percentage chance was 99.9%, it'd lock the launchers and refuse to fire at all.
PGI isn't bright enough to program such a thing (but to be fair I don't even think a big company like DICE could either, though the Javelin comes close in BF3/BF4 but even that can miss), and even then the limitations of it would suck. Especially since the standard lore streak is no smarter than a regular SRM in terms of the missile itself, the difference there is the rapid path-programing port for streaks when they enter the launcher.

So for MWO, they won't fire without a lock, and they will defy the laws of physics to any extent (short of hitting objects) to nail their target. 40 donuts around it? Sure! Complete U-turn to hit a target behind you -- if you still have that lock, YES! (requires target decay and 360 degree target retention and a fast mech like a Raven 3L). Corner-missiling! Definitely! (In fact it's a big way to troll people). Shoulder-perch missiles? YES! (Stand on enemy, fire streaks, enjoy!)

What's that?! Enemy on the ridge on top of you? Stand just below by edge. Unlock arms, put "O" crosshair on sensor target square, get lock and fire! Insert troll face!

Streak rear-shots are made fun of here, but it is a real thing.


The series playlist.

Edited by Koniving, 02 May 2016 - 12:49 PM.


#8 Koniving

    Welcoming Committee

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • The Guide
  • The Guide
  • 23,384 posts

Posted 02 May 2016 - 12:50 PM

Fixed link to the last video.

#9 Not A Real RAbbi

    Member

  • PipPipPipPipPipPipPipPip
  • Survivor
  • Survivor
  • 1,688 posts
  • LocationDeath to Aladeen Cafe

Posted 03 May 2016 - 12:11 AM

View PostLikeUntoGod, on 30 April 2016 - 11:01 AM, said:

I know I should keep everything in one thread but I thought this was semi-important for new guys. When I found this page https://mwomercs.com/profile/stats it helped me see things a bit better.

It has a number of interesting areas and more of a few that stuck out.

One was Weapons and my accuracy with them(or lack of). I saw that I was only hitting 10% of my Narc targets. I was surprised to see how poorly I'm doing with missiles. (this Mech came with 2 large LRM launchers).

My plan was to go with short streak missiles because they "never miss". Either yes, they do miss a lot or maybe the target went out of range often but the very low numbers were a shock.


The stats are notoriously bugged. Weapons stats, DOUBLY so. For instance, I have (according to the stats page) used IS Gauss in ONE match since the last stat reset. I fired it 299 times. Considering that you get 10 rounds of Gauss ammunition per ton, that means I'd have needed at least THIRTY TONS of Gauss ammo on a single mech to do that. This is, as I'm sure you recognize, DIFFICULT at least. (A fellow reddit user managed to design a KGC build with a Gauss rifle and 30 tons of ammo, but it's terribly slow and short of armor. It CAN be done, however it is incredibly inefficient and kinda pointless, as it would take about 30 minutes to fire all those rounds and a typical match only lasts 15.)

Stats page tells me I've fired an IS AC/20 640 times, and that I did so in a total of 11 matches. That's somewhere in the neighborhood of 60 shots/match, which means about 8.5 tons of ammo expended per match on average. The majority of DUAL-AC/20 builds don't carry that much ammo.

So yeah, take the stats with a BIG grain of salt. They're awful. They're also not separated by game mode. That may be fine and dandy prior to 19-April-2016, but now we have SCOUTING in there. It's 4v4, it's part of Faction Warfare, and it REFLECTS IN YOUR STATS LIKE QUICK PLAY MATCHES. Don't believe me? Go do a Scouting drop real quick, then have a look at the stats page 15 minutes later for that same mech. Your damage, kills, deaths, etc., will all reflect that match. And since SCOUTING is different from Domination, Conquest, Assault, and Skirmish, and in some pretty big ways, that skews your data into meaninglessness.

I was starting to look HARD at my stats before Phase 3 dropped. Interesting stuff. And then, it all went to ruin. Bummer, that.

Something to look at, though, is this:

For each weapon, take the total damage done with it and divide by the total number of times fired. That's your damage/shot (for missile weapons, this is PER MISSILE regardless of how many the launcher fires, but it still may be worth looking in on). Now, compare that damage/shot to the max/optimal damage for that weapon. For instance, an IS ERPPC does ten points of damage MAX (not counting critical). If you're averaging 7.3 damage per trigger pull, then you're doing pretty well. If you're averaging more like 5 per pull, you have some 'splaining to do Lucy.

Take that information (which, unlike some other weapons stats, is actually fairly reliable), and determine which of your weapons are most damage-efficient for you. That is, with which weapons systems are you getting the highest percentage of the weapon's potential damage each time you fire? When you want to farm a challenge event, or you just feel like putting on your A-game, use/boat THOSE weapons. When you're toying around, or trying to improve your play with underachievers, then use the weapons that you have a lower % with.

Now, you can go SUPER nuts and put that weapon data alongside your mechs' KDRs and damage/match, and figure out which weapons to boat on which mechs in order to maximize your potential efficiency.

For instance, I see by doing so that I have a high damage/hit (>100% of listed damage/shot) for the SRM-4, and highest (as of 13-April) damage/match in the JR7-O "Oxide" hero mech, so I can conclude that to be a pretty good match for me.

I also see that I have (more recently, though it's VERY skewed by Faction Warfare Invasion matches being included in the mech stats) a higher-than-average (for me) KDR, among my best, in the BL-6-KNT. But none of the weapons I can mount on it are among my top performers for damage/shot and damage/hit.

It's right about then, that you realize the limits to the usefulness of those statistics. One of my best mechs uses one of my least-efficient weapons systems (TWO of them, actually).

But trying to discipline yourself to fire lasers ONLY at targets within the optimal range, in order to shore-up that stat, will cost you a LOT of damage/match and kills and assists (ESPECIALLY) and so on.

Missiles will be skewed a bit, because they either do FULL damage or NO damage. Your hit % with lasers will look great, but then you realize that you regularly get <50% of their damage on the target, and that only one tick of the laser shot needs to hit a target for the whole shot to register as a hit.

And it just gets weirder from there.

We definitely need some love on the stats pages...





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users