92 replies to this topic

[dario03]

Jiffy, on 14 May 2019 - 02:03 PM, said:

I think this was true a while ago back when there were more players, but I find that it's not hard to maintain about a 3.0 W/L, 6.0+ K/D, and ~60% survival rate in solo these days. Mostly because people will literally just sit and let you shoot them. I play more solo queue than anything else. Not even playing up my own abilities, it's just the state of the game anymore.

K/D and survival rate I could see, especially if you avoid battle if hurt. But W/L can be a bit difficult since you are just one of twelve. It can of course be done at times, but maintaining it wouldn't be so easy. One of my months that was all solo ended with a bit over 2w/l at nearly 200 games, though at around 120 games it had been over 3w/l. No major change in what I was doing, just the way the win and loss streaks went.
I guess maybe if you run nothing but really strong well rounded mechs, like a good heavy or assault mech you could do it. But I don't think it would be easy and your k/d and survival might take a hit.

[Insufficient Skill]

Wil McCullough, on 12 May 2019 - 11:19 PM, said:

Let's put it this way. I was consistently in the top 5% of players back when i played and guys like bearclaw wrecks my face when we're on opposite teams. The gap between top 5% and top 1-2% is HUGE.

I would say top 1-2 % is way too broad, especially going by skewed stats like in Jarls. Heck, even I had a single season, where I made above 99.5% on Jarl's. In my experience, it's the guys consistently above 99.5%, so top 0.5%, are really top tier.

Those guys can gut you even piloting a stock mech on any map in any mode.

[Jman5]

Jiffy, on 14 May 2019 - 02:03 PM, said:

I think this was true a while ago back when there were more players, but I find that it's not hard to maintain about a 3.0 W/L, 6.0+ K/D, and ~60% survival rate in solo these days. Mostly because people will literally just sit and let you shoot them. I play more solo queue than anything else. Not even playing up my own abilities, it's just the state of the game anymore.

I would just be careful. It's not hard for even a small number of games in the group queue to have an outsized impact on your overall WLR. Imagine for a second you play 100 games in the Solo Queue winning 72, losing 28 for a 2.57 WLR. It would only take 12 extra wins in the Group Queue to move that up to a 3.0 WLR. So even if only about 10% of your games are group queue, a high winrate in GQ they can really move the needle.

This is why it kills me that group and solo queue have a merged leaderboard. It's just so hard for anyone to know what stats are realistic. I would love to know for certain that all the Division A comp guys are managing 3.0+ WLR in the solo queue and in Light mechs. It would clue me in that you guys are definitely doing some thing(s) right that I'm doing wrong.

[Jiffy]

Jman5, on 14 May 2019 - 04:39 PM, said:

I would just be careful. It's not hard for even a small number of games in the group queue to have an outsized impact on your overall WLR. Imagine for a second you play 100 games in the Solo Queue winning 72, losing 28 for a 2.57 WLR. It would only take 12 extra wins in the Group Queue to move that up to a 3.0 WLR. So even if only about 10% of your games are group queue, a high winrate in GQ they can really move the needle.

This is why it kills me that group and solo queue have a merged leaderboard. It's just so hard for anyone to know what stats are realistic. I would love to know for certain that all the Division A comp guys are managing 3.0+ WLR in the solo queue and in Light mechs. It would clue me in that you guys are definitely doing some thing(s) right that I'm doing wrong.

It's arguably easier in light mechs like the Wolfhound to pull off crazy carries, since they have a good mix of agility, tankiness, and firepower. Yes, losing a few matches can easily throw off W/L, but for the past couple seasons I've done mostly solo, and was around 3.0 W/L. Admittedly, this is nothing to aspire to. Solo queue is a far cry from anything resembling good gameplay.

[Bowelhacker]

Jarl's only measures Quick Play performance?

[justcallme A S H]

Bowelhacker, on 14 May 2019 - 05:32 PM, said:

Jarl's only measures Quick Play performance?

Yes. Solo and GroupQ are combined as PGI does not split them out.

[justcallme A S H]

Jman5, on 14 May 2019 - 04:39 PM, said:

This is why it kills me that group and solo queue have a merged leaderboard. It's just so hard for anyone to know what stats are realistic. I would love to know for certain that all the Division A comp guys are managing 3.0+ WLR in the solo queue and in Light mechs. It would clue me in that you guys are definitely doing some thing(s) right that I'm doing wrong.

Yep, kills me as well. Having them split would be so much better just from information sake and interest.

I rarely do GroupQ due to my timezone - And by rarely I mean max of 10 games a month outta me 250-300. I average a 2.5 WLR in SoloQ. Some outlier seasons I've hit 3.0 however I find that is more of a rare occurrence than a norm. Plus I **** around a fair bit cause I stream etc - run objectively rubbish builds and stuff which probably isn't helping my teams win as much as I could in full / strict META.

It actually depends heavily on what events and the type that are in each given season. I know that may sound crazy but I firmly believe some events lower the quality of games because of the objectives etc. People going off to Yolo a UAV to pop and then die 3s later in the first 60s of the match in a light.

Where some events that don't have silly objectives means that players don't divulge into totally idiotic gameplay.

Edited by justcallme A S H, 14 May 2019 - 06:53 PM.

[GoodTry]

Jman5, on 14 May 2019 - 04:39 PM, said:

I would just be careful. It's not hard for even a small number of games in the group queue to have an outsized impact on your overall WLR. Imagine for a second you play 100 games in the Solo Queue winning 72, losing 28 for a 2.57 WLR. It would only take 12 extra wins in the Group Queue to move that up to a 3.0 WLR. So even if only about 10% of your games are group queue, a high winrate in GQ they can really move the needle.

This is why it kills me that group and solo queue have a merged leaderboard. It's just so hard for anyone to know what stats are realistic. I would love to know for certain that all the Division A comp guys are managing 3.0+ WLR in the solo queue and in Light mechs. It would clue me in that you guys are definitely doing some thing(s) right that I'm doing wrong.

Being a giant nerd, I keep a spreadsheet of all my matches. I'm not a Div A comp guy, but my WLR so far this month in solo queue only is 3.19, and my KDR is 5.96 (151 games total).

If you stretch it back to include last month, though, it's 2.72 WLR and 5.16 KDR (324 games) for solo queue only. I bet a top-tier player could do better :-).

Edit: I missed the comment about "in light mechs." My light mech WLR this month is 2.75, and KDR is 5.48, over 71 solo queue games (no group queue games in light mechs this month).

Edited by GoodTry, 14 May 2019 - 06:54 PM.

[Jman5]

Jiffy, on 14 May 2019 - 04:49 PM, said:

It's arguably easier in light mechs like the Wolfhound to pull off crazy carries, since they have a good mix of agility, tankiness, and firepower. Yes, losing a few matches can easily throw off W/L, but for the past couple seasons I've done mostly solo, and was around 3.0 W/L. Admittedly, this is nothing to aspire to. Solo queue is a far cry from anything resembling good gameplay.

Yeah the "mostly solo" is my issue. You can play mostly solo and your group wins still have an outsized impact on your final WLR. Without seeing an actual breakdown it's just really hard to know.

justcallme A S H, on 14 May 2019 - 06:05 PM, said:

Yep, kills me as well. Having them split would be so much better just from information sake and interest.

I rarely do GroupQ due to my timezone - And by rarely I mean max of 10 games a month outta me 250-300. I average a 2.5 WLR in SoloQ. Some outlier seasons I've hit 3.0 however I find that is more of a rare occurrence than a norm.

Those stats are closer to my experience lately.

Quote

It actually depends heavily on what events and the type that are in each given season. I know that may sound crazy but I firmly believe some events lower the quality of games because of the objectives etc. People going off to Yolo a UAV to pop and then die 3s later in the first 60s of the match in a light.

Where some events that don't have silly objectives means that players don't divulge into totally idiotic gameplay.

I definitely believe that matchscore can be heavily influenced by events. For example on that day 1 event where it asked people to play lights my scores were pretty low. When they get to the Assault part of the event, I bet you're going to see matchscores go up.

You could be right that some events make the outcome a little more 50/50. Would be interesting to track.

GoodTry, on 14 May 2019 - 06:45 PM, said:

Being a giant nerd, I keep a spreadsheet of all my matches. I'm not a Div A comp guy, but my WLR so far this month in solo queue only is 3.19, and my KDR is 5.96 (151 games total).

If you stretch it back to include last month, though, it's 2.72 WLR and 5.16 KDR (324 games) for solo queue only. I bet a top-tier player could do better :-).

The World needs more nerds and their spreadsheets!

[B L O O D W I T C H]

justcallme A S H, on 13 May 2019 - 03:13 PM, said:

Now upgrade that to ATM VGL-3... I mean I've done similar in other mechs/builds so it's nothing out of the ordinary stats wise.

Pretty sure I've done similar in a XL Atlas of all things

Okay, done that now. That thing is such a cheat.

[Alienized]

Overall 937 97.62% percentile
is dat gud?

Edited by Alienized, 20 May 2019 - 01:04 PM.

[Brizna]

If Solo and Group queue stats were separated win/lose ratio would be a really nice statistic to know who is better, since it's not we have to go by MS ranking, and we all know how biased the MS formula is.

[Kubernetes]

If we're talking top tier as in the elite of the elite, then it's got to be the top comp guys. Other measures don't mean much. I'm pretty high up in Jarls, but the skill difference between myself and a true top tier player is pretty apparent to my eyes.

Edited by Kubernetes, 21 May 2019 - 12:41 PM.