

Flaw In Paul's Psr Thinking
#1
Posted 27 September 2015 - 07:34 AM
First, I don't necessarily agree. It seems most of the players that say they are in T4 also mention that they like to play a variety of mechs or play non-optimized builds. But that isn't my main point.
If this game rewards teamwork and teamwork is the behaviour you want to promote, why would you place new players in T4? Wouldn't you want to place them in the tier where teamwork is the norm not the exception? Wouldn't you want them observing what you think are the "right" behaviours from the start?
#2
Posted 27 September 2015 - 07:39 AM
I personally have been playing this game for 3 years, and still have not learned all the vital things. Especially quirk values on all mechs.
Edited by El Bandito, 27 September 2015 - 07:44 AM.
#3
Posted 27 September 2015 - 07:42 AM
El Bandito, on 27 September 2015 - 07:39 AM, said:
As I pointed out this is a much different, less competitive game than LoL. They are not even close and cannot be compared with the exception of they are online games that people play.
#4
Posted 27 September 2015 - 07:49 AM
Ted Wayz, on 27 September 2015 - 07:42 AM, said:
The logic is all the same. You cannot feed fresh meat into the grinder right away. You will only drive away new players like that. Being competitive has nothing to do with it.
Edited by El Bandito, 27 September 2015 - 07:51 AM.
#5
Posted 27 September 2015 - 08:00 AM
El Bandito, on 27 September 2015 - 07:49 AM, said:
The logic is all the same. You cannot feed fresh meat into the grinder right away. You will only drive away new players like that. Being competitive has nothing to do with it.
So put them in T3 then. T3 doesn't mean they are better players than T4 players, just they use better teamwork according to Paul. And I don't think seeding in 1 in 12 is "feeding meat to the grinder". In LoL 1 in 5 would be, especially in a game that has actual roles and where you need independent play (lanes) to succeed.
#6
Posted 27 September 2015 - 08:15 AM
Ted Wayz, on 27 September 2015 - 08:00 AM, said:
I do not know how your brain worked out that there will be only 1 fresh meat seeded in a T3 match. Especially when the game hits Steam.
#7
Posted 27 September 2015 - 08:48 AM
El Bandito, on 27 September 2015 - 08:15 AM, said:
I do not know how your brain worked out that there will be only 1 fresh meat seeded in a T3 match. Especially when the game hits Steam.
I didn't say in a T3 match, sorry if you didn't pick that up. If you put them in T3 it would be a better fit according to Paul's reasoning.
What I meant to convey is seeding 1 in 12 in the upper tiers would not have much of an effect. Match maker could easily ensure a 1 in 12 and one on each team seeding in all T2 and T1 matches. Still T3 makes the most sense.
#8
Posted 27 September 2015 - 08:50 AM
In tier 5 they can still see some sort of "teamwork" because they would be matched with/versus tier 3/4
#9
Posted 27 September 2015 - 09:04 AM
-k
#11
Posted 27 September 2015 - 09:13 AM
Kdogg788, on 27 September 2015 - 09:04 AM, said:
-k
BS.
I enjoy rolling quickie(draws), lolcust, trollmando, ferrit, etc.etc.
I also wrote a love guide to HGN (http://mwomercs.com/...-love-for-them/)
I play to have fun, not to be a slave.
Edited by Stefka Kerensky, 27 September 2015 - 09:14 AM.
#12
Posted 27 September 2015 - 09:13 AM
And for the really good players, it sucks to hold new players' hands through the match as well.
It's a lose-lose proposition, in my book.
#13
Posted 27 September 2015 - 09:16 AM
New players would get wrecked in tier 3, even if they didn't see tier 1 or 2 opponents, for the simple reason that they'd be playing with and against people who had the money to fully outfit and customize their machines. Even with a completely unjustified assumption that piloting skill and knowledge would be equal, that's a huge disadvantage already.
Tier 4 is probably the right place so new players have the ability to go up or down, as opposed to just up.
#14
Posted 27 September 2015 - 09:17 AM
Ted Wayz, on 27 September 2015 - 07:34 AM, said:
First, I don't necessarily agree. It seems most of the players that say they are in T4 also mention that they like to play a variety of mechs or play non-optimized builds. But that isn't my main point.
If this game rewards teamwork and teamwork is the behaviour you want to promote, why would you place new players in T4? Wouldn't you want to place them in the tier where teamwork is the norm not the exception? Wouldn't you want them observing what you think are the "right" behaviours from the start?
I always thought that the flaw is you have 11 people not working as a team, and unless you carry the team you're not considered a team player either.
The chart needs to look like this. With more areas for no change, and a faster way to accel if you did good. Notice, there's no "small" change.
Regardless of win or lose, if you did poorly, you're going down.
Regardless of win or lose, if you did OK, you'll stay the same.
If you lose, and did good, you'll still stay the same as your contribution may not have been team orientated.
If you win, and did good, you'll go up as your contribution was likely team orientated.
If you win, or lose, and you did awesome, you're obvious in the wrong place, so you'll be going up.
Paul punishes and rewards win/loss too much. Tier is all about skill placement, so that's why keeping you in a certain spot is essential, and getting you into a different tier faster that matches you is also essential.

#15
Posted 27 September 2015 - 09:25 AM
It would be interesting to see the lowest number of games any of the tier 1's has. I'm thinking there won't be many or any under 5k of games as a guess.
#16
Posted 27 September 2015 - 09:25 AM
MoonUnitBeta, on 27 September 2015 - 09:17 AM, said:
The chart needs to look like this. With more areas for no change, and a faster way to accel if you did good. Notice, there's no "small" change.
Regardless of win or lose, if you did poorly, you're going down.
Regardless of win or lose, if you did OK, you'll stay the same.
If you lose, and did good, you'll still stay the same as your contribution may not have been team orientated.
If you win, and did good, you'll go up as your contribution was likely team orientated.
If you win, or lose, and you did awesome, you're obvious in the wrong place, so you'll be going up.
Paul punishes and rewards win/loss too much. Tier is all about skill placement, so that's why keeping you in a certain spot is essential, and getting you into a different tier faster that matches you is also essential.

I'm fine with such placements, but it might be biased against lighter mechs under current reward system.
#17
Posted 27 September 2015 - 09:27 AM
#18
Posted 27 September 2015 - 09:41 AM
#19
Posted 27 September 2015 - 10:04 AM
#20
Posted 27 September 2015 - 10:06 AM
Ted Wayz, on 27 September 2015 - 08:48 AM, said:
What I meant to convey is seeding 1 in 12 in the upper tiers would not have much of an effect. Match maker could easily ensure a 1 in 12 and one on each team seeding in all T2 and T1 matches. Still T3 makes the most sense.
And if you put them in T3 they can also end up being the one T3 guy in a match full of T1 players, if their luck sucks for their entire cadet life, which is exactly what initially seeding them in T4 is supposed to avoid.
They´re still getting thrown to the wolves, because they can still be matched with T3 and T2 players... but at least they`re not getting tossed in front of a whole pack of Comp overlords with min/Maxed mechs and full module sets while themselves piloting their first trial mech.
There`s a big difference between a trial mech and a build that makes full use of quirks, in the hands of a player who`s played at least a few hundred matches
"A FEW HUNDRED?? BS!!" you say? No, think about it:
He had to level at least 2 mechs thru elite and the one he´s currently in to master, and unless he`s got too much RL money to spend he had to farm an additional 10-15 million + for modules. if we assume earnings that (for a beginner) are above average at 200k /match, that`s 50 matches for just the modules (if we assume 10mil), another 10 mil for the mech he`s currently in, and another 10 mil for the second mech he bought (assuming the first was "free" after cadet bonus and tutorial), that´s already a bare minimum of 150 matches..
All assuming he knew from the second he logged on exactly what he was doing boith in game and in teh mechlab, and what his favorite mech was going to be, and never wasted a single cBill..... In reality, the number will probably be about twice if not triple that for most people.
If we assume 10 matches /hr, you have someone who`s been playing for half a week nonstop without eating,. sleepinging, or using the loo, and he`s getting matched with a player that started literally 2 minutes ago.
And that´s the BEST CASE scenario for the new recruit if matched against T1 players.
The worst case, which is equally likely when seeded in T3, is getting tossed in against veterans with thousands of matches each under their belt that know the maps and small grooup tactics like the back of their hand. He may as well just dc right after the drop, as his match score will be insignificantly different.
Edited by Zerberus, 27 September 2015 - 10:30 AM.
6 user(s) are reading this topic
0 members, 6 guests, 0 anonymous users