105 replies to this topic

[El Bandito]

Some people claim River City is the most common map. I call that horsedung.

Freaking Alpine (ALPINE!) is more popular than both River City combined in my case.

Edited by El Bandito, 13 March 2013 - 08:38 AM.

[Livewyr]

17
17
17
20
17
12
19
19

Good thing you got going here Agent..nothing like some numbers and facts to prove or disprove an observation.

[Agent of Change]

Livewyr, on 13 March 2013 - 08:37 AM, said:

Good thing you got going here Agent..nothing like some numbers and facts to prove or disprove an observation.

Thanks, sometimes I get a bug up my *** to work the numbers and what i think we see more than anything else in the complaints about random maps is threefold:

1. Confirmation Bias - expecting to see a map you don't like makes it seem more common when it shows up.
2. Streakiness - small pool of maps means the likely hood of seeing them in streaks or very close together is not uncommon.
3. Small data sets - at this point any given individual isn't going to be able to get a decent view of the big picture with only a couple hundred matches tracked.

[Tikkamasala]

42,40,46,48,44,38,35,33

[ItsAPotato]

Skyscream Sapphire, on 13 March 2013 - 07:51 AM, said:

Busted? What are you talking about? It is in fact a statistical confirmation of what I knew all along:

Potato whines a lot for no reason

I admit it, complaining is one of my more favorite activities.

[Sprouticus]

9,2,7,4,4,6,5,9

[hammerreborn]

17
13
25
17
27
24
16
24


Lol river city night is my 2nd lowest played (its also my most successful map).

Edited by hammerreborn, 13 March 2013 - 09:18 AM.

[DDDs]

18
17
18
16
23
22
19
18

[Surtr]

10
5
5
9
10
13
3
10

[Skyscream Sapphire]

hammerreborn, on 13 March 2013 - 09:17 AM, said:

Lol river city night is my 2nd lowest played (its also my most successful map).

Isn't funny how, with low sample sizes, some maps can be so much more successful than others? You may as well disconnect if you see me on the other side of Frozen City Night because I'm 17-0 since the stats patch! lol

I also win RC Night 70% more than I win on regular RC. Disproportionate wins on night maps? I must be Batman.

[Vincent Lynch]

17
12
14
17
14
12
12
13

[Agent of Change]

Skyscream Sapphire, on 13 March 2013 - 09:25 AM, said:

Isn't funny how, with low sample sizes, some maps can be so much more successful than others? You may as well disconnect if you see me on the other side of Frozen City Night because I'm 17-0 since the stats patch! lol

I also win RC Night 70% more than I win on regular RC. Disproportionate wins on night maps? I must be Batman.

OI, give me your numbers there peanut gallery.

[ItsAPotato]

Skyscream Sapphire, on 13 March 2013 - 09:25 AM, said:

Isn't funny how, with low sample sizes, some maps can be so much more successful than others? You may as well disconnect if you see me on the other side of Frozen City Night because I'm 17-0 since the stats patch! lol

I also win RC Night 70% more than I win on regular RC. Disproportionate wins on night maps? I must be Batman.

Like that time you disconnected the ONE TIME we lost on it?

Agent of Change, on 13 March 2013 - 09:28 AM, said:

OI, give me your numbers there peanut gallery.

He can't. Sky and I always drop together, so it would duplicate my stats

[Agent of Change]

ItsAPotato, on 13 March 2013 - 09:31 AM, said:

He can't. Sky and I always drop together, so it would duplicate my stats

hmm. i'm wondering if that would actually skew the data it the size of the sample set i'm getting.

i mean that's an interesting question. Wether or not to teach two players who end up ion teh saem games (teamed or theoretically random) as the same or separate entities. the grand picutre of all games by all players would by neccessity include teams.

Also Updated Data 5 is in the OP now.

We are at 5300 games worth of data.

[Matt Minus]

Agent of Change, on 13 March 2013 - 08:41 AM, said:

Thanks, sometimes I get a bug up my *** to work the numbers and what i think we see more than anything else in the complaints about random maps is threefold:

1. Confirmation Bias - expecting to see a map you don't like makes it seem more common when it shows up.
2. Streakiness - small pool of maps means the likely hood of seeing them in streaks or very close together is not uncommon.
3. Small data sets - at this point any given individual isn't going to be able to get a decent view of the big picture with only a couple hundred matches tracked.

It never ceases to amaze me how many players in games based on numbers cannot grasp basic probability. Nothing is as amusing as people on MMO forums postulating wildly complex formulas to explain what could just as easily be attributed to randomness.

[Agent of Change]

Matt Minus, on 13 March 2013 - 09:37 AM, said:

It never ceases to amaze me how many players in games based on numbers cannot grasp basic probability. Nothing is as amusing as people on MMO forums postulating wildly complex formulas to explain what could just as easily be attributed to randomness.

Can't agree more.

[Wittich]

10
11
4
11
6
11
7
9

[LeMans]

Here's mine:
5, 5, 14, 11, 11, 6, 5, 10

Thanks for the effort, was wondering what kind of deviation others were seeing...

[Skyscream Sapphire]

Agent of Change, on 13 March 2013 - 09:36 AM, said:

hmm. i'm wondering if that would actually skew the data it the size of the sample set i'm getting. i mean that's an interesting question. Wether or not to teach two players who end up ion teh saem games (teamed or theoretically random) as the same or separate entities.

Yes, adding my drops would absolutely skew the data. You are attempting to treat the map choice as an independent random variable, like rolling the dice each time. Since Potato and I were (almost) always grouped together during this sample period, our variables are obviously dependent. It would be like rolling the dice once and recording the value twice. It would amplify the variance.

I've been trying to figure out what the minimum sample size required to get an XX.X% confidence that we're within a Y% margin of the 12.5% hypothesis, but I haven't quite been able to wrap my head around it yet (while also trying to be productive at work, lol).

[Mechteric]

11
14
11
12
11
8
7
21


so my river city night is only 7/95=7.37%
but Alpine woah, 21/95=22.1%

Edited by CapperDeluxe, 13 March 2013 - 10:06 AM.