409 replies to this topic

[Grus]

Zergling, on 11 April 2017 - 10:08 PM, said:

As Carl Vickers said, the whole game is a team game. It doesn't matter what mode you are playing, if you are playing at all effectively you will have a positive W/L ratio.
If you aren't playing effectively, you will have a negative W/L ratio.

I'll agree to this and Carl's but only to a point, ill defer to my "potatos are real" post. Using LRMS is definitely not easy mode, but they can be a powerful tool when used correctly. If I'm in my maddog moveing around the battlefield, focusing fire on engaged targets and getting3/4 kills and my team just gets wiped out for what ever reason. I don't see how w/l for games should come into play when asked how good a person is in a particular mech.

To put it differently, if someone has a 8/1 kd in a mech but a1/8 win rate, is that person a bad player in that mech? Or is he/she just not getting good teams?

[Zergling]

Grus, on 12 April 2017 - 08:27 AM, said:

I'll agree to this and Carl's but only to a point, ill defer to my "potatos are real" post. Using LRMS is definitely not easy mode, but they can be a powerful tool when used correctly.

They can be, but in most circumstances direct-fire weapons are the better option.

A good player can still be effective when using LRMs, but they will almost always be more effective using direct fire weapons.

Grus, on 12 April 2017 - 08:27 AM, said:

If I'm in my maddog moveing around the battlefield, focusing fire on engaged targets and getting3/4 kills and my team just gets wiped out for what ever reason. I don't see how w/l for games should come into play when asked how good a person is in a particular mech.

Win/loss results of individual battles don't matter from a player performance standpoint, because there are many battles that will be won or loss regardless of personal performance.
What matters is W/L over a large number of battles, because that will reveal player performance in the battles that are close enough that the actions of a single player can influence the battle towards a win or loss.

And the number of battles that can be influenced like that are far from rare; I estimate about 50% of all battles can be influenced towards a win by a player of high skill.


Going further into that, there are three types of random battles:

1) Battles that you will win regardless of what you do
2) Battles that you will lose regardless of what you do
3) Battles that are close enough that personal performance can either result in a win or a loss

The proportions are probably around 25% #1, another 25% #2, and about 50% being #3.

This means that a player that is completely useless and fails to influence any of battle type #3s to produce a win, will end up with just 25 out of 100 wins, and a 0.33 W/L.

An average player will only influence half of the battles of type #3 to a win, and fail in the other half, resulting in 50 wins to 50 losses, and a 1.00 W/L.

A fantastic player that manages to influence ALL of battle type #3 to a win, will end up with 75 out of 100 wins, and a 3.00 W/L.

Grus, on 12 April 2017 - 08:27 AM, said:

To put it differently, if someone has a 8/1 kd in a mech but a1/8 win rate, is that person a bad player in that mech? Or is he/she just not getting good teams?

A player with such a high K/D and low W/L would likely be focusing on stealing kills at the expense of their teammates, which is why they have a low W/L.

A good player will have high W/L, K/D, K/B and match score stats. They might drop in some stats over the short term, but over hundreds of battles they will have fairly high stats in all categories.

[Grus]

I got mixed feelings about that, "stealing" a final blow on a mech can be a **** move no argument there. But a mech with all its guns at Red core is still able to put lead down range. So removing guns from the field isn't a bad thing.

But let me ask this. What about 8kmd per death and 1/8 win rate.

[Zergling]

Grus, on 14 April 2017 - 08:32 AM, said:

I got mixed feelings about that, "stealing" a final blow on a mech can be a **** move no argument there. But a mech with all its guns at Red core is still able to put lead down range. So removing guns from the field isn't a bad thing.

But let me ask this. What about 8kmd per death and 1/8 win rate.

I'd define 'stealing' as someone that intentionally holds fire until the enemy mech is near dead, so they can take the killing blow. That is the sort of behaviour that pads kill stats at the expense of teammates, so it produces higher kill stats but lower W/L.

And it is pretty much impossible to score 8 kill most damage per death with a 0.125 W/L over a statistically significant number of battles, so its a non-issue.
Even if it was possible, it'd still be a non-issue, because kills are less important than winning. It doesn't matter how many kills you score if you don't win the battle, because the only goal that matters is winning.

People judge by W/L first; the other stats are honestly just there to see if a person is being carried.

Eg, if someone has a high W/L but poor K/D, K/B and avg Match Score, they are probably just being carried in group queue, or the number of battles in their sample is low enough that they just got lucky with teams.
In the reverse case, if someone has poor W/L but decent K/D, K/B and avg Match Score, they are padding their kill and damage stats.

It is pretty simple: a genuinely good player will have good W/L, K/D, K/B and avg Match Score, over a large number of battles.
If they are lacking in one of those stats, then they aren't as good as their other stats indicate.

Edited by Zergling, 14 April 2017 - 02:55 PM.