31 replies to this topic

[Xavori]

Xiphias, on 02 November 2017 - 03:07 PM, said:

You literally don't understand how statistics work.

I know you are, but what am I!

Seriously. Random numbers are random.

That's what you keep skipping over. 12 vs 11...uh... Okay, 12 potatoes vs 11 comp players? 12 average players vs 1 god-like player and 10 average players? 12 good players who don't use voip vs 11 who do? Are you getting it yet?

AND THERE IS ZERO I CAN DO TO MAKE THE OTHER 11 PLAYERS INTO GOOD PLAYERS WHO WORK TOGETHER.

Heck, just getting 11 other players who work together, good or not, doesn't matter

p.s. A well driven mech with LRM's can absolutely shape a battlefield. Kinda like how artillery works IRL...

Then again, you've made it perfectly clear you aren't exactly the sharpest tool in the shed.

So one more time: Random numbers are random.

[Xavori]

Wait. I got it. The example anyone could understand.

Let's say you have a random number generator that creates the numbers 1-10 11 times and adds them together. The program then writes in a 12th number that is always 10.

You want the highest number possible. You've even cheated and made one of the numbers the best it can be. Is this going to produce top-notch results the majority of the time?

No.

You'll get everything between 21 and 120. Your mean is going to be ~60. That's barely above average. And that's even though you cheated and made sure one of your numbers was the absolute best it could be.

So even with that absolute best possible number 8.3% of the time, random numbers are still random.

[Nightbird]

Guys, just let Matchmaker take the blame

[Xavori]

Nightbird, on 02 November 2017 - 04:03 PM, said:

Guys, just let Matchmaker take the blame

Ya know. If the matchmaker and PSR were accurate, everyone would eventually have 50/50 W/L records in solo drops.

[Nightbird]

Xavori, on 02 November 2017 - 04:08 PM, said:

Ya know. If the matchmaker and PSR were accurate, everyone would eventually have 50/50 W/L records in solo drops.

Exactly, if you're below 1:1, it's the MM's fault. If you're above 1:1, it's also the MM's fault. I'm above 1:1 because I buy mechpacks, therefore the MM put me in better teams.

[Xavori]

Nightbird, on 02 November 2017 - 04:16 PM, said:

Exactly, if you're below 1:1, it's the MM's fault. If you're above 1:1, it's also the MM's fault. I'm above 1:1 because I buy mechpacks, therefore the MM put me in better teams.

Funny thing tho...

That Cyclops Spotlight build I posted...actually has a 19W 10L record to go with it's meh 18 kill 17 death K/D ratio. It absolutely makes my teammates better even tho it itself is pretty meh in terms of leaderboard stats (my Raven 3L does the same thing with the same leaderboard-free results). The Cyclops is built entirely around that concept.

It can 'see' and lock on to mechs over 1500m away. That means it tells my teammates where those mechs are when I press R (ya know, maybe I should write some poems about how OP the R key is or something...). It provides ECM coverage to mechs near it. It also has the rather unique ability to boost the sensor range of nearby teammates because of the built-in B2K computer. So when I started putting my take on that together, that was my thought. And it directed everything.

When it came to weapons, I kept going with the idea that I'm always going to be hanging out with as many of my teammates as possible which means I cannot count on any short range weapons ever being useful. So it has none. I also went way overboard with the number of LRM's I'm carrying not because I care at all about trying to lurm things to death (it only has a single LRM 20 launcher), but because it's really useful to make enemies hide behind buildings while your team maneuvers against them, and I can do that pretty much nonstop. Oh, and if I do get shots, I go for legs...mostly because when I'm driving an IS mech I like being honorless freebirth scum....but also because that's more useful to my team than trying to burn through 100 points of CT armor using nothing but 2 ERLL's and and the occasional lucky LRM.

In turn, this mech has proven wildly useful in making teams just a lil' bit better even if it means my damage, k/d, etc. don't really benefit. And I'm okay with that.

Of course, at the opposite end of the teamwork spectrum I have a Mist Lynx I tore all the armor off of in order to create a sneaky sniper. It cannot risk getting shot at which means it spends as much time as possible staying as far away from everybody, including teammates, as it can while taking potshots at bad guys. It gets pretty good damage per match doing that even though it has a terribad W/L record and K/D ration (cuz if anything even sneezes at it...poof). However, it's lots of fun for me to play such a challenging creation even tho doing so is murder on my stats.

[Xiphias]

Xavori, on 02 November 2017 - 03:42 PM, said:

Wait. I got it. The example anyone could understand.

Let's say you have a random number generator that creates the numbers 1-10 11 times and adds them together. The program then writes in a 12th number that is always 10.

You want the highest number possible. You've even cheated and made one of the numbers the best it can be. Is this going to produce top-notch results the majority of the time?

No.

You'll get everything between 21 and 120. Your mean is going to be ~60. That's barely above average. And that's even though you cheated and made sure one of your numbers was the absolute best it could be.

So even with that absolute best possible number 8.3% of the time, random numbers are still random.

Your example is the correct one to use, your conclusion from it is erroneous though. You've stated right here that your mean is going to be above average. That means you are having a statistically significant effect on the result through your actions. The system is no longer completely random and is biased towards the one that generates the 10.

Your mistake I think is that you are interested in getting the highest number possible, you aren't. What you are trying to get is a number that is higher than another randomly generated team that doesn't have your perfect 10. You don't need top notch results every time, you just need results that are slightly better than your opponents. They will also have their share of problems so as long as you can edge them out you will trend positively.

Let me actually do this to show you how it works.

Here is an example: https://docs.google....#gid=1406785854

It generates 400 matches with 23 of the players being completely random between 1-100. The 24th player is always on team 1. The first sheet has random but static values for the other players. If you change the "skill" of the player you get the following KDRs

01 = 0.56
10 = 0.60
20 = 0.65
30 = 0.76
40 = 0.88
50 = 0.97
60 = 1.16
70 = 1.29
80 = 1.42
90 = 1.60
100 = 1.86

Unsurprisingly the above average players have WLRs that are higher than 1 and the worse players have ones that are less than 1.

There's also a second sheet that regenerates all the numbers every time a cell value is edited (I'd recommend downloading it since it's faster to update that way). If you select and empty cell use use the delete key you can see various WLRs for the same skill player over different sets of 400 matches. Unsurprisingly you'll see some variation in the number. What you will notice is that the higher skill players usually have higher WLRs and the worse players have lower (sub 1) ones.

The more matches you add the tighter these groupings get. While it's possible for a skill 1 player to end up with a higher WLR than a skill 100 it is extremely unlikely.

In practice the distribution will be different which means a really good player like Proton can have an even greater effect on a match (say skill of 150) because there would be extremely few other players at that level while the vast majority would be grouped around 50.

Quote

Ya know. If the matchmaker and PSR were accurate, everyone would eventually have 50/50 W/L records in solo drops.

Close to it. The problem is that T1 is bloated and has a massive skill gap between the top and the bottom. This post gives a good graph of it. If skill was more evenly distributed and matched things would trend to 50/50.

[ForceUser]

This exact same argument comes up in every game like this that I've ever played. I remember a particularly animated discussion in word of warships around this exact topic.


What I've noticed is that there is usually a correlation between the skill of the player and whether they think they can have an affect on their W/L(edit: and all their other stats too) or not. This makes sense since the worse the player the less of a positive effect they have so the more they believe that player skill does not over the long run affect W/L and that the leaderboard or stats in general is meaningless. It's easier to blame team mates, match maker and random randomness than their own lack of ability OR drive to do better.

Same reason why some people run special snowflake builds. Easier to run a bad build and blame it for their poor performance than running a good build and still being bad.

THAT SAID, something that is a little bit different in MWO is that compared to some other games it is harder (but not impossible) to carry hard in MWO but that is down to specific game mechanics regarding dealing damage. There is no bouncing/reflecting/true tanking of damage like you can see in games like world of tanks or world of warships or to a lesser degree war thunder ground forces. Additionally how spotting mechanics works in those games. Watching replays you'll notice that those 1v5 or 1v6 games often relies on either completely negating incoming damage or never getting spotted in the first place.

That's not to say you can't absorb larger than normal amounts of damage through making use of the damage transfer mechanic/torso twisting, using speed to make other players miss, using ECM to avoid detection or simply using range to avoid return fire but a good enemy player can mitigate those actions again through targeting, positioning and situational awareness.

And that's what it comes down to, better players will have better stats because they are better. And that's a hard fact for a lot of people to swallow.

Edited by ForceUser, 02 November 2017 - 11:07 PM.

[Xavori]

ForceUser, on 02 November 2017 - 11:06 PM, said:

And that's what it comes down to, better players will have better stats because they are better. And that's a hard fact for a lot of people to swallow.

No. I'd be totally okay with better players having better stats EXCEPT....

Your stats are in so very many ways tied to your teammates, and in solo drops, you have zero control over that.

Your stats on the leaderboard are also tied to events that are often random. Kills are not kills, but are instead last hit counts. You can do 1 damage to a mech but get the kill if that damage is the one that finally brings down the enemy mech.

Your stats are easy to manipulate in ways that have nothing to do with game objectives. If you want high damage, shoot off components. This will also boost your KMDD stats. If you want high win %, run with a 4-5 man team. More people than that and the tonnage restrictions start becoming prohibitive. Less than that, and we're back to random numbers are random territory.

Your stats don't reflect other things you can do to make your teammates more successful. The leaderboards don't track time spent capturing objectives, scouting other mechs, soaking damage, shooting down missiles, and so much else that goes into good team based play.

If you want meaningful leaderboards, the comp play leaderboards are where you should be looking. There you should see the best teams moving to the top because the only thing being tracked, and the only objective they are working towards is winning. Over time, the best teams will be at the top.

But in solo queues, random numbers are random.

[Kroete]

ForceUser, on 02 November 2017 - 11:06 PM, said:

And that's what it comes down to, better players will have better stats because they are better. And that's a hard fact for a lot of people to swallow.

Thats right, but only if you have all the numbers needed for it and lots of the numbers (at least 100 matches a season to see a trend, more then 1000 matches if you want some meaningfull stats).

The mwo statistics base on wl, kd and matchscore (which is based mostly on damage), but as example
assists (showing for good teamplay), kmdd (doing the work but not the kill), spotting (teamplay again),
are not used for the mwo statistics.

They are only a small part and they show only if you are are good at doing damage (streak-boats),
killing (standing back and vulturing open enemys), and wins (hope my team has better players then their team).
They show not the real big picture, the stats we get are only showing a tendency without any proof and the possibility to game the numbers as showen in the above examples.

Edited by Kroete, 03 November 2017 - 04:28 AM.

[Dr Cara Carcass]

Xavori, on 03 November 2017 - 04:20 AM, said:

No. I'd be totally okay with better players having better stats EXCEPT....

Your stats are in so very many ways tied to your teammates, and in solo drops, you have zero control over that.

Your stats on the leaderboard are also tied to events that are often random. Kills are not kills, but are instead last hit counts. You can do 1 damage to a mech but get the kill if that damage is the one that finally brings down the enemy mech.

Your stats are easy to manipulate in ways that have nothing to do with game objectives. If you want high damage, shoot off components. This will also boost your KMDD stats. If you want high win %, run with a 4-5 man team. More people than that and the tonnage restrictions start becoming prohibitive. Less than that, and we're back to random numbers are random territory.

Your stats don't reflect other things you can do to make your teammates more successful. The leaderboards don't track time spent capturing objectives, scouting other mechs, soaking damage, shooting down missiles, and so much else that goes into good team based play.

If you want meaningful leaderboards, the comp play leaderboards are where you should be looking. There you should see the best teams moving to the top because the only thing being tracked, and the only objective they are working towards is winning. Over time, the best teams will be at the top.

But in solo queues, random numbers are random.

Ohh thats ********. You always have controll over your stats. I often get 2:12 stomps with me gettign 600dmg 2 kills and everybody else was below 100dmg exept for one or two wo get 400dmg and the assists. Shure those games drag my average down. But compared to my team and the enemy team i am often the guy with the highest damage and the highest matchscore. I am not saying majority of stomps i recive but often enough. If you play good you get better results - period! When i start a game and i see 2-3 good players on the other side or in ym team i know where to go what to do and i know that i can initiate things coz those players will watch the minimap and so on.
If i play agaisnt them i am the best chance of my team to kill them If i ge that i often get a win, if i fail at that i often lose. But yeah, sometimes you are simply not on the side where you could shoot the good players. You take down 2 noobs while the rest of your team died to the 3 good enemy players. Its a 1v3 at that moment. But this shows directly the influence you have as a better player. It doesnt negate the fact that luck sometimes helps, that sometimes crazy **** happens and so on. but over time it will all cancel out and your skill will show. Perhaps you could get better coz one season you really had bad luck.
Last season i had 50% of my games with full lrm teams that fail to move one inch, or games where people tried to snipe over 700m with srms and spl. And poeple trying to get a lock for mrms and never shootthose mrms because that reticule never changes. Overall i had about the same matchscore i always had. better than 98% of the other players in this game according to isens list. Sure i often think i could have done better if those teams werent so stupid, but then i also remember that its my fault for beliving a radom pug game in tier 1 will not have lrm only assaults coz the xp bar knows more ways up than down. Its my fault for trying to move the team at all. the didnt reacto to voice they didnt react to chat, why would they react to the big red cloud gettign split in half because tehy try to hunt a squirrel. Why would they take the 11 vs 6 advantage and win teh game. Why not lose 2:12 with em and my assasin doing 900+dmg and gettign thsoe two kills out of the 6 who followed me. Or why use teh fact that i killed 3 enemy mediums and its 3:0 and tr to push. No they wait for the emeys to chip away on teh team team with their direct fire coz their lrm boats cant do ****.


Also on the topic of this thread i intentionally planned....
Over night nothing changed. Still 2 more games recorded in the forum leaderboard compared to the in game leaderboard. defenetly a disparity

Edited by Cara Carcass, 03 November 2017 - 05:00 AM.

[Shifty McSwift]

All of my games this season were recorded into the leaderboards, I am not sure about mech/weapon stats, but the number of games on the leaderboard and scores are right for me at least. I took more notice this season and last because my PC started acting up and I was playing far less games in general.