Humans Overpowered? - With Stats

[Kiwi]

I happened to glance over here today, and I decided I would see if all these clams (coming from me also) were true about humans being overpowered.  I took an average of what team won what percent of the time for every map that has been played ten or more times.  These are the results I got:
Humans - 55% wins
Alien - 39% wins
Tie - 2%

I understand that the GPP is still undergoing construction, and 1.2 isn't out, but I just wanted to point out that humans currently are winning a lot more than aliens.  Nearly in a 3:2 ratio.

[cron]

I don't believe those clams.

[Kiwi]

Click the link and look, add them up if you want
- Kiwi

[CreatureofHell]

Fishy things are so much more believable than crustaceans 

[cron]

Listen here shrimp  , you can't muscle me into believing some obscene clams.  I'm too smart for that. 

[cron]

Oh yes, and:

⎛11:31⎞ < FrameShifter> i thinks humies are a bit stronger nao
⎛11:31⎞ < FrameShifter> or maybe alien players can't get used to all the changes

[Kiwi]

Hmm, no.  Those numbers are very real.

[FreaK]

I've been playing on there a lot and this is what I've observed. When the game is less than 8v8 but greater than 4v4 it's pretty balanced but above 8v8 it feels impossible to even touch humans in huge groups. I don't think trem is balanced for those big games in tremulous. I think playing slots should be reduced or the server slots should be reduced an another server should be put up.

Basically I think aliens have the advantage in smaller games, humans have the advantage in large games but it's pretty balanced when it's 6v6 7v7.

[temple]

Yeah Freak, but hasn't that always been true for Tremulous?

[David]

55+39+2 != 100

What's the rest of it?

[Kiwi]

It's 96, because tremstats rounds and I round. (slightly)

Edit: If you take my rounding out, at least to 2 decimal places:
Humans: 55.25%
Aliens: 38.96%
Tie: 2.58%

- Kiwi

[Plague Bringer]

I happened to glance over here today, and I decided I would see if all these clams (coming from me also) were true about humans being overpowered.  I took an average of what team won what percent of the time for every map that has been played ten or more times.  These are the results I got:
Humans - 55% wins
Alien - 39% wins
Tie - 2%

I understand that the GPP is still undergoing construction, and 1.2 isn't out, but I just wanted to point out that humans currently are winning a lot more than aliens.  Nearly in a 3:2 ratio.

[Kiwi]

Correct, I wanted to see if these claims were true.
A claim is:

an assertion of something as a fact

So, I wanted to see if humans win more often than aliens is true.

[CreatureofHell]

Correct, I wanted to see if these claims were true.
A claim is:

an assertion of something as a fact

So, I wanted to see if humans win more often than aliens is true.

Oh you mean claims! I honestly did not realise  I thought you had just used some unheard-of meaning of clams.

[FreaK]

they are just saying you misspelled claims as clam in your post

[David]

Wins by time and by player count, from both tremstats.  (Only counting games with players, and ignoring specs)

Code Select Expand


+----------+--------+--------+-----+------+-------+
| time     | aliens | humans | tie | draw | total |
+----------+--------+--------+-----+------+-------+
|   0 to 5 |    297 |    231 |   0 |    8 |   536 |
|  5 to 10 |    682 |    708 |   0 |    7 |  1397 |
| 10 to 15 |    373 |    565 |   0 |    4 |   942 |
| 15 to 20 |    267 |    269 |   0 |    3 |   539 |
| 20 to 25 |    140 |    134 |   0 |    0 |   274 |
| 25 to 30 |     94 |     87 |   0 |    2 |   183 |
| 30 to 35 |    107 |     90 |   0 |    5 |   202 |
| 35 to 40 |     67 |     61 |   0 |    7 |   135 |
| 40 to 45 |     31 |     37 |   0 |    4 |    72 |
| 45 to 50 |      2 |      2 |  74 |    0 |    78 |
| 50 to 55 |      1 |      1 |   0 |    0 |     2 |
| 55 to 60 |      0 |      3 |   0 |    0 |     3 |
| 60 to 65 |      1 |      1 |   0 |    0 |     2 |
| 65 to 70 |      1 |      0 |   0 |    0 |     1 |
| 70 to 75 |      2 |      1 |   0 |    0 |     3 |
| 75 to 80 |      0 |      0 |   2 |    0 |     2 |
| 85 to 90 |      1 |      0 |   0 |    0 |     1 |
|    total |   2066 |   2190 |  76 |   40 |  4372 |
+----------+--------+--------+-----+------+-------+


+---------+--------+--------+-----+------+-------+
| players | aliens | humans | tie | draw | total |
+---------+--------+--------+-----+------+-------+
|       1 |     50 |     57 |   5 |    1 |   113 |
|       2 |     48 |     53 |   7 |    0 |   108 |
|       3 |     53 |     48 |   6 |    0 |   107 |
|       4 |     51 |     38 |   3 |    0 |    92 |
|       5 |     76 |     75 |   4 |    0 |   155 |
|       6 |     72 |     49 |   2 |    1 |   124 |
|       7 |     62 |     80 |   3 |    3 |   148 |
|       8 |     62 |     67 |   1 |    1 |   131 |
|       9 |     87 |     91 |   3 |    3 |   184 |
|      10 |     65 |     78 |   4 |    1 |   148 |
|      11 |     82 |     74 |   3 |    3 |   162 |
|      12 |     62 |     64 |   2 |    1 |   129 |
|      13 |     66 |     70 |   2 |    1 |   139 |
|      14 |     41 |     69 |   1 |    0 |   111 |
|      15 |     51 |     55 |   1 |    4 |   111 |
|      16 |     55 |     58 |   7 |    3 |   123 |
|      17 |     57 |     73 |   0 |    1 |   131 |
|      18 |     47 |     65 |   3 |    0 |   115 |
|      19 |     44 |     53 |   0 |    1 |    98 |
|      20 |     39 |     51 |   1 |    1 |    92 |
|      21 |     34 |     41 |   0 |    0 |    75 |
|      22 |     33 |     32 |   0 |    0 |    65 |
|      23 |     34 |     43 |   0 |    1 |    78 |
|      24 |     25 |     37 |   0 |    0 |    62 |
|      25 |     19 |     31 |   1 |    0 |    51 |
|      26 |     19 |     24 |   0 |    2 |    45 |
|      27 |     18 |     22 |   1 |    0 |    41 |
|      28 |     11 |     14 |   0 |    0 |    25 |
|      29 |      7 |      6 |   0 |    0 |    13 |
|      30 |      5 |      8 |   0 |    0 |    13 |
|      31 |      5 |     11 |   0 |    0 |    16 |
|      32 |      2 |      5 |   1 |    1 |     9 |
|      33 |      1 |      0 |   0 |    0 |     1 |
|      34 |      7 |      4 |   1 |    0 |    12 |
|      36 |      0 |      3 |   0 |    0 |     3 |
|      37 |      1 |      1 |   1 |    0 |     3 |
|      48 |      1 |      1 |   0 |    0 |     2 |
|   total |   1392 |   1551 |  63 |   29 |  3035 |
+---------+--------+--------+-----+------+-------+



With just game with 6-24 players, its  1018 to aliens, 1150 to humans, 33 ties and 25 draws.

I think (might be wrong) that a tie is time limit and draw is votes etc.

[Kiwi]

they are just saying you misspelled claims as clam in your post

haha, I didn't even realise that  

@David:
That data does make sense, and you can graph it like this:

Where the red line represents the aliens and the green line represents the humans.
The x-axis is the number of players, and the y-axis is the amount of wins.
You can see that as long as the number of players is less than 6.5, the aliens have an advantage.  As long as you have over 6.5 players, the humans have a greater chance.

Regards,
Kiwi

[David]

Here's an interesting table, it's who won and the average number of lines said. (chats per minute, no adjustment for player counts)
I was expecting the winner to on average talk more, but instead it seems aliens are more chatty. :s

(Only counting games that lasted more than 5 min)

Code Select Expand


+-------------+-------+-----------+-----------+-----------+-----------+------------+------------+
| game_winner | games | alien_avg | alien_dev | human_avg | human_dev | public_avg | public_dev |
+-------------+-------+-----------+-----------+-----------+-----------+------------+------------+
| aliens      |   715 |       1.8 |       2.0 |       1.5 |       1.8 |        4.2 |        4.0 |
| draw        |     9 |       2.3 |       1.4 |       1.9 |       1.4 |        5.8 |        4.1 |
| humans      |   888 |       1.9 |       2.2 |       1.4 |       1.6 |        4.3 |        4.1 |
| none        |    87 |       0.6 |       1.6 |       0.3 |       0.8 |        1.7 |        3.3 |
| tie         |    33 |       1.0 |       1.3 |       0.7 |       0.8 |        2.1 |        2.4 |
+-------------+-------+-----------+-----------+-----------+-----------+------------+------------+


(If there's any random bits of data people want say and I'll see what I can do)

[Asvarox]

I wanted to post a nice wall of text, but...

Reduce the flamer damage! The Flamer's damage increased in Phase 2, splash damage was reduced in Phase 3. What are the consequences? Flamer can now one hit a dretch, and almost insta burn any other class (except for rant). Considering it's ammo and type, I believe it's now the strongest weapon in the game, way over luci. The splash damage reduce in phase 3 made things even worse, it made flamer even less "selftoastable". It wasn't much of an issue until few days ago - now many people rediscovered this weapon and "overuse" it. Now once once humans got s2, they are almost unstoppable.

I didn't enjoyed almost at all last few days of played the GPP as aliens and humans alike, just because of the flamer. It was already stated, that it will be changed in phase 4. The point is that nobody specified when it will happen (as nobody knows). I know I am not on the position to suggest anything only on my judgment, but looking at phase 3 start date (3 weeks ago) I'd say that it's right time to start next one. I see no point in keeping the flamer as overpowered as it is now.

[Kiwi]

That's true, but aliens also get many things at s2 such as poison, which will waste a medikit or cause a human to run back to his base, and amazingly fast regen speeds at a booster.  Aliens also get adv. goons at s2 which can help reduce camping.  Even after all of this, I still believe that humans are overpowered, and I believe that the stats show that.

Regards,
Kiwi

[Conzul]

It's that God Awful Pulse Rifle.

[temple]

That's true, but aliens also get many things at s2 such as poison, which will waste a medikit or cause a human to run back to his base, and amazingly fast regen speeds at a booster.  Aliens also get adv. goons at s2 which can help reduce camping.  Even after all of this, I still believe that humans are overpowered, and I believe that the stats show that.

Regards,
Kiwi

All of that was possible in 1.1, with the exception of advanced goons.

1.2 Advanced goons are slower to kill bases than the 1.1 advanced goon.  It takes 2 goons coordinating sniping to actually kill anything.  1.1 wasn't like this.  So I don't know if advanced goons, while nerfed, really make a big deal in 1.2.

[Norfenstein]

Adding those up I get 56% humans, 43% aliens. I didn't bother with the other server though, and for some reason EU favors aliens more than humans, so you shouldn't just be looking at one of them.

I don't use tremstats for analysis. I'm not going to post the current graphs I do use yet (because it's a pain converting them and uploading, and I'm probably going to make a post with them this week anyway), but currently they show:

US: 407 Aliens, 475 Humans
EU: 405 Aliens, 324 Humans
Combined: 812 Aliens, 799 Humans

I don't know when tremstats was last reset, and the graphs don't count all maps and games (timelimits, not enough players for an actual game, things like that). At this point I'm not as concerned about the ratio, and only a little more concerned about how straight the lines are (how well the balance scales with game size) -- mostly the statistics are there to confirm I didn't horribly break anything while adjusting things based on my own judgement and your feedback.

[SlackerLinux]

only thing i think is really broken is flamer the phase before i thought was kinda balanced well

[bob0]

they are just saying you misspelled claims as clam in your post

You misspelled clams as clam in your post.

[Kiwi]

All of that was possible in 1.1, with the exception of advanced goons.

That's not true.  The booster regen speed is amazing now, to the point where it can heal a near dead goon or rant in a matter of seconds.  Poison now can also take nearly halve of a humans health.  If you are even slightly hurt and get poisoned, you are forced to use your medikit or run back to your base.

1.2 Advanced goons are slower to kill bases than the 1.1 advanced goon.  It takes 2 goons coordinating sniping to actually kill anything.  1.1 wasn't like this.  So I don't know if advanced goons, while nerfed, really make a big deal in 1.2.

I don't find that very true.  A turret gets killed with 2 barbs, and the arm with 3 hits (direct hits at least).  The exploding barbs + poison also is very fun inside the human base.

only thing i think is really broken is flamer the phase before i thought was kinda balanced well

Hopefully, this will be fixed when ff is turned on.

Adding those up I get 56% humans, 43% aliens. I didn't bother with the other server though, and for some reason EU favors aliens more than humans, so you shouldn't just be looking at one of them.

I don't use tremstats for analysis. I'm not going to post the current graphs I do use yet (because it's a pain converting them and uploading, and I'm probably going to make a post with them this week anyway), but currently they show:

US: 407 Aliens, 475 Humans
EU: 405 Aliens, 324 Humans
Combined: 812 Aliens, 799 Humans

I don't know when tremstats was last reset, and the graphs don't count all maps and games (timelimits, not enough players for an actual game, things like that). At this point I'm not as concerned about the ratio, and only a little more concerned about how straight the lines are (how well the balance scales with game size) -- mostly the statistics are there to confirm I didn't horribly break anything while adjusting things based on my own judgement and your feedback.

I agree that tremstats isn't the most reliable, I was just surprised that it reported such a large favor to humans.  These new stats sound very good, and hopefully they depict the gpp very well.  I'd also like to say that you devs have done a great job so far, and I hope you continue to.  I just want to give some constructive feed back, and what things feel like to me.

Regards,
Kiwi

[temple]

I don't find that very true.  A turret gets killed with 2 barbs, and the arm with 3 hits (direct hits at least).  The exploding barbs + poison also is very fun inside the human base.

It takes forever to shoot 2 barbs now.  So you need to 2 goons to actually kill a ret.  Otherwise you can repair as fast as the barb recharges.

[bob0]

I don't find that very true.  A turret gets killed with 2 barbs, and the arm with 3 hits (direct hits at least).  The exploding barbs + poison also is very fun inside the human base.

It takes forever to shoot 2 barbs now.  So you need to 2 goons to actually kill a ret.  Otherwise you can repair as fast as the barb recharges.

Allow me to paraphrase:
The 1/5th of a second that was added between barb launches is too HUGE of a difference that requires another whole goon to kill a turret, even when barbs have splash damage when they miss.

[AppleJuice]

@temple: Play an organized clan match/pick-up game and you'll see how good coordinated goon barbing can be (better than in 1.1), depending on the base

[temple]

@temple: Play an organized clan match/pick-up game and you'll see how good coordinated goon barbing can be (better than in 1.1), depending on the base

I believe it.  I'm just saying it requires 2 goons and teamwork.