392 replies to this topic

[DisordeR_]

face4stas, on 05 February 2015 - 01:07 AM, said:

 

Guys, only what we have to focus is making system fair. System that is twisted one way or another will gain no respect from players. It now seems to me that 0.3 penalty is too stiff. What about 0.15

 

	battles	wins	platoon wins	Platon wins adjustment	Platoon wins penalty	Adjusted wins	Win Rate	Adjusted win rate	Diff
jmstichler	13560	8502	2889	0.15	433	8069	62.7	59.51	3.19
Serapth	3384	1848	23	0.15	3	1845	54.61	54.52	0.09
fullestnuke	9402	7659	7362	0.15	1104	6555	81.46	69.72	11.74

 

I can only say that it looks better to me. Fullest is definitely at least 10% better than me. 

[thunderthies]

Another, simpler solution, could be to include raw WR and adjusted WR. 

[face4star]

dddqqq, on 05 February 2015 - 01:12 AM, said:

 

Instead of focusing on winrate you could make a system based off of contribution instead. Average damage, spots, k.d, damage ratio, etc. 

 

 

As I mentioned in OP the baseline would be determined for

a) Win Rate

b) Damage

c) Kills

d) Spots (that should be a reflection of the unreported by API spoting damage)

 

Each of these factors would have a weight that would depend on the tank type. It is expected that TDs would make more damage but less spotting when mediums would score exactly opposite.

[originalmadkilla]

jmstichler, on 05 February 2015 - 01:12 AM, said:

 

I can only say that it looks better to me. Fullest is definitely at least 10% better than me. 

 

I just want what's fair also, but do you believe that platooning has only added 3.19% to your normal win rate? Doesn't that seem a bit low?

[dddqqq]

Ah ok. I didn't even look at the OP

[acrisis]

The thing everybody needs to keep in mind, is that the formula is not set in stone. It hasn't even really been used yet, and can be tweaked further down the road.

 

I agree that some adjustment for platooning probably should be made. At the same time, individual experienced players may make a great platoon, inexperienced players that just platoon will probably have no benefit ... and guys with a high win rate to begin with who now constantly play together, potentially live audio chat .. could rule any give game with two sets of eyes and ears.  And what I'm trying to say is if two 40% xp platoon or a 40%& 50% platoon effect may be minimal. The better the combo of xp players in the platoon, the higher the net effect will be.

 

Great work face4stas. 

[Nutellanism]

All I have to say is. Thank you face4stas for even giving up one minute of your time to help. Loving the generosity. Any system you come up with is appreciated man.

[_Virtus]

Looks great! The second calculated statistic is much better than the first in my opinion. Maybe you can offer 2 ratings, one that penalizes platooning and one that doesn't. When you are faced against another player, what if they already are in a platoon? If they are, I would like to be able to check WOTBstars and assume Fullestnuke(example) is a 82% player instead to a 69% player..

Both are still TERRIFYING though when he's on the red team...

[gl123]

face4stas, on 05 February 2015 - 01:01 AM, said:

 

Fine. What would be the fair rate ? 0.1 0.2 0.3 ?

 

It is hard to rate platoon - why? Cause you can platoon with or w/o using  mumblespeak/teamkill. If there is a way to identify that mumble/teamspeak has been used during the game, then my personal opinion is that your 0.7 multiplier is OK. Platooning w/o mumble/teamspeak is almost as playing as individual ... therefore, penalty could be 0.1 e.g. 0.9. But is there a way to know who is playing with or w/o audio connections.

Bottom line is that mumble/teamspeak greatly enhance your chances of win ... however, we also have to notice that some guys (concep, fullestnuke, etc.) are class on its own regardless how they play.

Edited by gl123, 05 February 2015 - 01:21 AM.

[concep]

Why even have an adjusted WR or ? I honestly don't care about WR as much as I care about damage caused to damage received. That is what matters. It means you contributed to your team more. Why penalize platooners in the first place? WN8 doesn't judge a person by platooning and is considered an excellent system.

[thunderthies]

originalmadkilla, on 05 February 2015 - 01:16 AM, said:

 

I just want what's fair also, but do you believe that platooning has only added 3.19% to your normal win rate? Doesn't that seem a bit low?

 

You're assuming he plays exclusively with players of equal or greater WR than his own. There has to be some leeway to account for players who platoon to help others. There's a handful who signed up for the Training Grounds thing. In those instances, it's ver plausible that platooning could actually hinder the chance to win.

[Serapth]

Its hard to say,  I think my winrate platooned is about 75%-80%, but then I've had so few games to go from it could have been a run of good matchmaker luck.  Obviously that's not sustainable, since I don't think many players have winrates in that range and I'm not a rockstar by any means.

 

If you wanted, since I almost never platoon, if someone wanted to partner opposite, I could do a run of games and see what the effect on my winrrate would be.  

[_law]

gl123, on 04 February 2015 - 06:19 PM, said:

 

It is hard to rate platoon - why? Cause you can platoon with or w/o using  mumblespeak/teamkill. If there is a way to identify that mumble/teamspeak has been used during the game, then my personal opinion is that your 0.7 multiplier is OK. Platooning w/o mumble/teamspeak is almost as playing as individual ... therefore, penalty could be 0.1 e.g. 0.9. But is there a way to know who is playing with or w/o audio connections.

 

anyone who has ever played with grant will tell you that you don't need ts to win 99.99% of the time.

[Serapth]

face4stas, on 05 February 2015 - 01:14 AM, said:

 

As I mentioned in OP the baseline would be determined for

a) Win Rate

b) Damage

c) Kills

d) Spots (that should be a reflection of the unreported by API spoting damage)

 

Each of these factors would have a weight that would depend on the tank type. It is expected that TDs would make more damage but less spotting when mediums would score exactly opposite.

 

Whats kinda funny is, all my scout awards... I think they were in either my B2, Matilda or At15

[Nutellanism]

fullestnuke, on 04 February 2015 - 05:21 PM, said:

 

anyone who has ever played with grant will tell you that you don't need ts to win 99.99% of the time.

 

Whats teamspeak?

[concep]

face4stas, on 04 February 2015 - 08:14 PM, said:

 

As I mentioned in OP the baseline would be determined for

a) Win Rate

b) Damage

c) Kills

d) Spots (that should be a reflection of the unreported by API spoting damage)

 

Each of these factors would have a weight that would depend on the tank type. It is expected that TDs would make more damage but less spotting when mediums would score exactly opposite.

 

I guess I better start suicide scouting every game and kill stealing if I'm still alive.

[thunderthies]

concep, on 05 February 2015 - 01:19 AM, said:

Why even have an adjusted WR or ? I honestly don't care about WR as much as I care about damage caused to damage received. That is what matters. It means you contributed to your team more. Why penalize platooners in the first place? WN8 doesn't judge a person by platooning and is considered an excellent system.

 

>>Why even have an adjusted WR?

 

Because platooning influences WR. 

 

>>I honestly don't care about WR as much as I care about damage caused to damage received. That is what matters.

 

Agreed.

 

>>WN8 doesn't judge a person by platooning and is considered an excellent system.

 

True. But this isn't WN8. It's also not WoT.

 

 

[thunderthies]

fullestnuke, on 05 February 2015 - 01:21 AM, said:

 

anyone who has ever played with grant will tell you that you don't need ts to win 99.99% of the time.

 

Yup.

[gl123]

concep, on 05 February 2015 - 01:19 AM, said:

Why even have an adjusted WR or ? I honestly don't care about WR as much as I care about damage caused to damage received. That is what matters. It means you contributed to your team more. Why penalize platooners in the first place? WN8 doesn't judge a person by platooning and is considered an excellent system.

 

agree 1000%

[Serapth]

concep, on 05 February 2015 - 01:19 AM, said:

Why even have an adjusted WR or ? I honestly don't care about WR as much as I care about damage caused to damage received. That is what matters. It means you contributed to your team more. Why penalize platooners in the first place? WN8 doesn't judge a person by platooning and is considered an excellent system.

Platooning is going to affect so many tangibles beyond winrrate.  People that platoon are generally going to also have higher survival rates and more consistent damage, although they won't have as many outlier games.

 

Ideal would be to just discard all the platooned games completely but it doesn't sound like a possibility.