Jump to content
Korean Random
Praetor77

[WN6, WN7 рейтинг]

Recommended Posts

 

 

каких отклонений?
Вы б  тему перечитали прежде чем постить

 

Почему нельзя взять танк, посчитать для каждого танка - средний уровень
кажется, тут проблема с терминами: Вы, наверно, предлагаете считать эффективность на танке. Читали бы тему - знали бы, что API не предоставляет необходимых данных, чтоб реализовать это именно таким образом

 

ротные бои в которых стат падает
снова проблемы с терминологией. Падает эффективность, и то не всегда.

И Вы не привели доводов о том, что

 

Это совершенно идиотское латание дыр.

Share this post


Link to post

Short link
Share on other sites

 

 

Вы, наверно, предлагаете считать эффективность на танке. Читали бы тему - знали бы, что API не предоставляет необходимых данных, чтоб реализовать это именно таким образом

В игре олень эту информацию отображает, а выцепить её нельзя?

 

 

Падает эффективность, и то не всегда.

А wn6, который в основном учитывает дамаг и фраги не падает? дамаг то в ротах всяко меньше, чем во взводе набить можно.

 

 

И Вы не привели доводов о том, что

Чем хеллкат-то не угодил? Тем, что хороший танк?

Share this post


Link to post

Short link
Share on other sites

 

 

а выцепить её нельзя?
В том и дело, что нет

 

дамаг то в ротах всяко меньше
Смотря как играть и в каких ротах ;) Но чаще всего, если в хороших - да. Только ПП стремится при такой игре к 80-90%, так что выбирайте и балансируйте сами

 

Чем хеллкат-то не угодил?
а где в формулах WN трогают Ведьму? Я чего-то не заметил? Ткните, если можно.

Share this post


Link to post

Short link
Share on other sites

 

 

а где в формулах WN трогают Ведьму?

Чуть выше пост с предложением по wn8, я как-то не правильно его понял?

 

 

 

Только ПП стремится при такой игре к 80-90%, так что выбирайте и балансируйте сами

Ну, если взвод хеллов собрать процент побед - в районе 70-90%, только дамаг средний раза в два выше.

Share this post


Link to post

Short link
Share on other sites

 

 

я как-то не правильно его понял?
видимо, да. Человек только сообщил, что работает над новой формулой. её ещё нет, а когда появится - в дискуссии и с аргументами можно будет вносить изменения.

 

Ну, если взвод хеллов собрать процент побед - в районе 70-90%, только дамаг средний раза в два выше.
ну так кто мешает? Я таких пациентов в рандоме часто встречаю. Если соло - мало чего стоят обычно. Хотя мой ПП тоже взводный.

Удивляют люди, которым рейтинг надо побольше, а не объективнее...

Share this post


Link to post

Short link
Share on other sites

 

 

Если соло - мало чего стоят обычно.

Иди по-вобй соло, когда в команде 14 раков и тебя по всем флангам давят. Не раз набивал 5-7 фрагов и слив, поэтому стараюсь взводы искать. Кроме того - это интереснее.

 

 

видимо, да.

И что он там про хелла написал, раз я не правильно понял?

 

 

Удивляют люди, которым рейтинг надо побольше, а не объективнее...

Мне он нафиг не нужен, просто когда в роту не берут потому что ты желтый - это немного удивляет, при том что у меня много боев на полтосе и арте. Поэтому если за рейтинг то объективный, а он то как не объективный. О том и речь я тут веду.

Share this post


Link to post

Short link
Share on other sites

 

 

Иди по-вобй соло, когда в команде
не мне расскажете,  я в курсе.

 

И что он там про хелла написал, раз я не правильно понял?
привел Ведьму в пример дамажной машины, на которой удобно накручивать ВН6-7, ибо в них упор именно на урон. Надо же делать поправки на то, что один на пт уронит, второй на ст, третий на лт. Вот и хочет сделать поправки на более дамажную технику. Мне кажется, это поможет не отставать светлякам.

 

в роту не берут потому что ты желтый
ищите роты своего уровня, улучшайте уровень, меняйте окрас и будут брать. Меня в фиолетовые тоже редко берут

Share this post


Link to post

Short link
Share on other sites

каких отклонений?

 

 

Для чего вообще делается этот рейтинг? Чтобы понимать, кто играет хорошо - а кто плохо. Понятное дело, что папок, которые бьют детей на младших левелах - надо занижать, это сделано в WN6. В wn7 опять планку опустили и дали повод фармить стату на Т49, когда 5лвл в массе - нубы.

Теперь предложение урезать статут по отдельным танкам. Смысл? Появятся другие, которые интереснее остальных на своем уровне, будем и их зарезать?

 

Почему нельзя взять танк, посчитать для каждого танка - средний уровень, а потом считать рейтинг игрока относительно среднего уровня на танке пропорционально кол-ву боев на нем?

 

А так получается что взводного хелла предлагают порезать, а ротного полтоса никто поднять не предлагает. Да и вообще, ротные бои в которых стат падает - почему-то никто не предлагает учитывать, о какой тогда субъективности рейтинга речь?

 

Using google translate which is pretty terrible, but the idea is very simple. Use player-tank data to calculate top100 damage and kills like what appears in the service record when using XVM. This will allow honest comparison of each tank´s "power". For WN7, if you have the same player play an A-20 and a Su122-54, his WN7 would surely be lower than if he played Tetrarch and Object 704.

 

The idea is to use per-tank top100 stats to multiply times the number of games a player has on each tank, then divide that by number of battles. Then you divide his overall account average by this number. We call this rSTATS, for ratio. So rFRAGS would be the players average frags divided by what the top100 would have gotten for the amount of games he has played in each tank. Same for rDAMAGE, rSPOT, rCAP, rDEFENSE, etc.

 

So, for example, for M4 Sherman top100 numbers are 1000 damage and 2.5 frags per game. For T1 3 frags and 400 damage. For Je100 1.5 frags and 3000 damage. Lets suppose a player opens 4 accounts and plays different tanks on each of them.

 

Lets say player A plays 10000 games on T1 averaging 3 frags and 500 damage. Average tier 1. WN7 (including low tier penalty) is about 1700.

Player B plays 5000 games on T1 and 5000 on JE100, averaging 2.25 frags and .1700 damage. Average tier 5.5. WN7 is 2750.

Player C plays 5000 games on M4 and 5000 on JE100,average 2 frags and 2000 damage. Average tier 7.5. WN7 is 2430.

Played D plays 10000 games on Je100. WN7 is 1840.

 

For WN8, his score would be around 2150 for all four cases.

Share this post


Link to post

Short link
Share on other sites

@Praetor77,

All is well. People just do not understand what you want to do.

But where will you get the top100 data on the tanks? API WG for a long time does not give values to individual tanks.

Share this post


Link to post

Short link
Share on other sites
ищите роты своего уровня, улучшайте уровень, меняйте окрас и будут брать. Меня в фиолетовые тоже редко берут

поэтому уже месяц на хелке только во взвод, уже +200 к WN6

 

 

For WN8, his score would be around 2150 for all four cases.

How to deal with light tanks that work on detection of the enemy, not the damage? Playing on their player has a low rating.

 

 

But where will you get the top100 data on the tanks? API WG for a long time does not give values to individual tanks.

You can collect a representative sample of those who use statistics. This information is available for statistics. I think that the developers can do it. But perhaps it is time-consuming.

It is not clear why the proposed take TOP100. You can take the average of the level and he already believed.

Edited by Little_boo

Share this post


Link to post

Short link
Share on other sites
How to deal with light tanks that work on detection of the enemy, not the damage? Playing on their player has a low rating.

your dmg will be comperade to he dmg of top-100 on the same wehicle, so if top players have 500 dmg per battle, it'll be OK to have 450 and keep good rating. That is quite obvious.

Edited by sech_92

Share this post


Link to post

Short link
Share on other sites

Hello, I should introduce myself here. I was the one who essentially came up with the idea for using top stats in WN8. In response to the question asking how the top100 stats will be calculated, I have thought of a different way of doing it which doesn't use actual top 100 stats, but predicted top stats. Using the current formula we have (and data), if Praetor77 agrees (I think he will), we will extrapolate or interpolate and predict the damage (and other stats) expected for a player with 2000 WN8. Then, we will change the values of the top stats to the newly-calculated stats and repeat the process until the numbers settle around a certain area. The final numbers will be the top stats we use.

 

Google Translate (very bad translation; I won't do this again):

 

Здравствуйте, я должен представить себе здесь. Я был тем, кто по существу придумал идею для использования Статистика лучших в WN8. В ответ на вопрос, спрашивая, как Top100 статистика будет рассчитываться, я думал о другой способ сделать это, который не использует фактические Топ 100 статистика, но предсказал Статистика лучших. Следуя уже существующей формуле имеем (и данные), если Praetor77 согласен (я думаю, что он будет), мы будем интерполировать или экстраполировать и прогнозирования ущерба ожидается для игрока с 2000 WN8. Тогда, мы изменим значения верхней статистика вновь рассчитанным статистика и повторить процесс, пока числа не останавливаются на определенной точке. Окончательные цифры будут Статистика лучших мы используем.

Ничего себе, Google Translate это плохо для этого.

 

O_o

 

! The stats we are using for analysis are from the entirity of the dossier files on the website vbaddict.net. If I remember correctly, it is from 175000 players from NA, EU and SEA who uploaded their dossier files.

 

!! As a result of this, maybe we can change (average / top) in XVM to (average / top / WN8=average / WN8=2000).

 

!!! Also, TEFF can be changed to be more accurate (by simply using WN8). We can then compare all of the tanks and come up with a theoretical linear "power rating" for each tank. I'm not certain whether TEFF can be used in-battle in XVM, but if it can, then certainly WN8 per-tank can be also. With this, we can show a player's skill level in a specific tank multiplied by the "power rating" of the tank in order to have a "danger rating". This will assist (greatly) in calculation of "Chance of Winning" and also help players decide which enemies they need to eliminate earlier, in a much more accurate way than just looking at what tank they are using and their Ratings (WN6, Efficiency, Victory %, TEFF).

Edited by bjshnog

Share this post


Link to post

Short link
Share on other sites

shnog anwered some things... but here I go...

 

@Praetor77,

All is well. People just do not understand what you want to do.

But where will you get the top100 data on the tanks? API WG for a long time does not give values to individual tanks.

http://www.vbaddict.net/wot.php

 

поэтому уже месяц на хелке только во взвод, уже +200 к WN6


 

 

How to deal with light tanks that work on detection of the enemy, not the damage? Playing on their player has a low rating.


 

 

You can collect a representative sample of those who use statistics. This information is available for statistics. I think that the developers can do it. But perhaps it is time-consuming.

It is not clear why the proposed take TOP100. You can take the average of the level and he already believed.

Average was tested first and does not work well. WE tried the average between average and top and that did not work well either.

 

your dmg will be comperade to he dmg of top-100 on the same wehicle, so if top players have 500 dmg per battle, it'll be OK to have 450 and keep good rating. That is quite obvious.

Exactly.

Share this post


Link to post

Short link
Share on other sites

shnog anwered some things... but here I go...

 

I was going to ask that you refer to me as "bjshnog" on this forum, since "shnog" has more of a chance to cause confision.

Share this post


Link to post

Short link
Share on other sites

 

 

who uploaded their dossier files

 

to (average / top

distribution of this players not equal distribution of all players. This is an average of the obviously more powerful players. In this way we can judge the top, but does not average.

@bjshnog,

I like the idea of ​​the analysis of specific machines on which the player played. I myself had proposed in February. I just want everything to be done soundly.

Share this post


Link to post

Short link
Share on other sites

distribution of this players not equal distribution of all players. This is an average of the obviously more powerful players. In this way we can judge the top, but does not average.

 

We use the current formula we have to predict what damage (or other stat) is expected for WN8 = 2000. After this, we replace all of the stats that we used in the previous WN8 formula. Then, Praetor77 can reanalyse the stats with the evolutionary algorithm and have a new formula. We repeat the process many times until the top stats don't vary as much.

 

If you meant that skill distribution in a specific tank does not reflect skill distribution over the entire server, then I agree, but that does not matter. The players who have lower skill (decided by WN8) will have lower stats in that specific tank, and the process we use to calculate top stats will tell us what to expect from a player with WN8 = 2000.

 

If you meant average specifically, then what I meant in that post was that XVM could show the average stats of the players (which can be calculated on a larger scale by the XVM people) AND the top players of that tank AND the stats to be expected from a player with 2000 WN8. I did not mean "average" as in the stat expected from an average player. That was only so that people could compare themselves to other players of that tank. So, for the Object 907, for example, our stats show an average damage of something like 1200. We know that is incorrect, but we also know that the skill of those players was low, so our process tells us what to expect from a player with a higher skill level.

 

Also, if the XVM developers have a larger database of player stats they would like to share with us, that would be brilliant. :)

 

I like the idea of ​​the analysis of specific machines on which the player played. I myself had proposed in February. I just want everything to be done soundly.

 

There is actually a better way to calculate top stats than what we are doing at the moment, but it is much more complex than the current method (which is already complex enough as it is). At the present moment, we are comparing the player's aggregate stats to the expected top stats, but with the stats we have, we could find individual tank WN8 for each player and then average the results, weighted by number of battles in each tank. I think the most this would do, however, is speed up the process of calculating the top stats for each tank.

 

A very important note (in case you didn't see it already): The current system we are using is independent of player skill distrubution in each tank. This means that if a tank has really bad stats, it is very likely that the players who played the tank are really bad at the game.

Edited by bjshnog

Share this post


Link to post

Short link
Share on other sites

@bjshnog,

The problem of the distribution of non-compliance is also the fact that the distribution of the average level too different. And we can get too high or too low for the requirements of any level (and any tanks).

Edited by seriych

Share this post


Link to post

Short link
Share on other sites

@bjshnog,

The problem of the distribution of non-compliance is also the fact that the distribution of the average level too different. And we can get too high or too low for the requirements of any level (and any tanks).

 

Could you explain what you mean a bit more? I don't quite understand.

Share this post


Link to post

Short link
Share on other sites

 

 

Could you explain what you mean a bit more? I don't quite understand.
Maybe I'm just is not enough understand your method.

But I mean this ... Most load data on appspot - are experienced players and their performance on low-level tanks will be quite possible indicators of newcomers (since they played in these tanks for a long time, but when they came into the game). In a real game a lot of players that play specially at low levels, and it is these players are on top these tanks. Thereby we get underestimated requirements for low-level, and low-level twinks get overly high rating. And this situation is not necessarily a low-level machines.

 

By the way, when we were discussing the figure TOP100 in TEFF, found out that it is is not enough correct. Better to take certain better percentage, rather than a fixed number. For example, the top 0.01%

Share this post


Link to post

Short link
Share on other sites

Maybe I'm just is not enough understand your method.

But I mean this ... Most load data on appspot - are experienced players and their performance on low-level tanks will be quite possible indicators of newcomers (since they played in these tanks for a long time, but when they came into the game). In a real game a lot of players that play specially at low levels, and it is these players are on top these tanks. Thereby we get underestimated requirements for low-level, and low-level twinks get overly high rating. And this situation is not necessarily a low-level machines.

 

I was actually thinking about that a lot today and I have a solution for it. When we calculate the line to predict the stats for WN8 = 2000, the first step is to find the mean point (the average point of the dataset for both parameters). Next, draw a line which goes straight from the origin (because 0 of all stats results in 0 WN8) through that mean point. At tier 1, that line should be multiplied by a certain percentage (for example: 96%). Every point below that line is ignored in the next step. Take the mean point of the remaining points (above the line) and draw a line through that. Ignore the first line. The second line drawn is what we can use to predict stats at WN8 = 2000. However, if the scatterplots still look weird, then we can also ignore all of the points where WN8 < 1000. As the tier value increases, that percentage slightly decreases (because there are less battles in which the player was a complete newbie).

Edited by bjshnog

Share this post


Link to post

Short link
Share on other sites
Guest
This topic is now closed to further replies.

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...