44 replies to this topic

[Valleye]

7
3
3
2
4
4
2
5

Edited by Valleye, 08 March 2013 - 08:37 AM.

[Dyallo]

Total Matches: 40
Total RC (Night and Norm): 17

Edited by Dyallo, 08 March 2013 - 11:24 AM.

[Kunae]

Dyallo, on 08 March 2013 - 08:28 AM, said:

Total Matches: 24
Total RC (Night and Norm): 12

How many were RCN, as that's the one we're tracking?

[Kekkone]

The wording of the topic might screw with the numbers, as it will probably attract people who have an issue or a streak of RC(N). Anyway i updated my own numbers on the first page of this topic.

[Tikkamasala]

ItsAPotato, on 08 March 2013 - 07:53 AM, said:

I agree, the sample size is too small at this point to make anything statistically significant.

But I was hoping to merge my data with everyone else's to get a larger set, rather than wait another 2 weeks (and I'm bored at work)

You need to be careful to not include the same match twice, i.e. if you often drop with someone else don't include his data. The same goes for a small Elo-bracket and/or a small (time dependent) player base where it's quite likely to drop with the same players.

Edit:

Forest Colony 11
Frozen City 14
Caustic Valley 19
River City 21
Forest Colony Snow 18
Frozen City Night 19
River City Night 10
Alpine Peaks 10

Edited by Tikkamasala, 08 March 2013 - 08:46 AM.

[Agent of Change]

Updated with Thontors and Valleye data


Individual maps:

ForC 21 + 7 = 28 (+3.5)
FroC 18 + 3 = 21 (-2.5)
CauV 23 + 3 = 26 (+1.5)
RivC 21 + 2 = 23 (+1.5)
FoCS 20 + 4 = 24 (-.5)
FrCN 25 + 4 = 29 (+4.5)
RvCN 24 + 2 = 26 (+1.5)
Alpline 15 + 5 = 20 (-4.5)

Total games 166 + 30 = 196

Map probability: 12.5% = 24.5
Set Probability: 25% = 49

Sets:

ForC (Combined) 52 (+3)
FroC (combined) 50 (+1)
RivC (combined) 49 (+0)
Caustic/alpine 46 (-3)


Some error for rounding but all individual maps are with +/- 4.5 of individual probability and the "sets" are all within +/-3 so given the limited sample set it's looking pretty spot on to random.

[ItsAPotato]

Kekkone, on 08 March 2013 - 08:33 AM, said:

The wording of the topic might screw with the numbers, as it will probably attract people who have an issue or a streak of RC(N). Anyway i updated my own numbers on the first page of this topic.

I was thinking this. When I created the thread in my boredom, I wasn't planning on keeping track of all maps or all stats.

Now that we've started, though, I'd rather like to keep track. Perhaps a new thread, in which the initial post specifies a specific format to use, and which includes all maps.

[Bilbo]

5,5,3,4,7,5,6,5.

[Anjian]

I totally agree. I ranted and posted it before that the frequency of River City Maps are too high. And I am not the only one. In my chat channel, I hear frequent complaints about it.

[ItsAPotato]

Agent of Change, on 08 March 2013 - 08:40 AM, said:

Updated with Thontors and Valleye data


Individual maps:

ForC 21 + 7 = 28 (+3.5)
FroC 18 + 3 = 21 (-2.5)
CauV 23 + 3 = 26 (+1.5)
RivC 21 + 2 = 23 (+1.5)
FoCS 20 + 4 = 24 (-.5)
FrCN 25 + 4 = 29 (+4.5)
RvCN 24 + 2 = 26 (+1.5)
Alpline 15 + 5 = 20 (-4.5)

Total games 166 + 30 = 196

Map probability: 12.5% = 24.5
Set Probability: 25% = 49

Sets:

ForC (Combined) 52 (+3)
FroC (combined) 50 (+1)
RivC (combined) 49 (+0)
Caustic/alpine 46 (-3)


Some error for rounding but all individual maps are with +/- 4.5 of individual probability and the "sets" are all within +/-3 so given the limited sample set it's looking pretty spot on to random.

Looks to be that way. Still, not a HUGE sample size, but starting to become significant.

Satisfies my curiosity for a while.

[Deathlike]

TBH, I don't think the randomness is borked by any sense, but when it comes to the MM, there is a tendency to have a streak of running the same map within a session... and it feels like it has the rate of a 50-50 coinflip. The chances of having the same map back-to-back-to-back (factoring in the lame variant maps) is higher than one would want. That would be part of what is causing it, but I swear the MM's RNG for maps bothers the heck outta me.

I don't think it's just a random occurrence.

Edited by Deathlike, 08 March 2013 - 09:01 AM.

[Agent of Change]

With Tikamasala and Bilbo's data:


Individual maps:
Old Total New data New total Difference
Forest Colony 28 + 16 = 44 0
Frozen City 21 + 19 = 40 -4
Caustic valley 26 + 22 = 48 4
River City 23 + 25 = 48 4
Forest Csnow 24 + 25 = 49 5
Frozen Night 29 + 24 = 53 9
River Night 26 + 16 = 42 -2
Alpine 20 + 15 = 35 -9

Total games 197 + 162 = 359

Map probability 12.50% = 44
Set Probability 25% = 89

Sets:
Old Total New data New total Difference
Forest colony 52 + 41 = 93 4
Frozen City 50 + 43 = 93 4
River city 49 + 41 = 90 1
Caustic 46 + 37 = 83 -6


individual maps are at +/- 9
sets are at +/- 6

Edited by Agent of Change, 08 March 2013 - 09:24 AM.

[TexAce]

Agent of Change, on 08 March 2013 - 09:06 AM, said:

With Tikamasala and Bilbo's data:

still working

individual maps are at +/- 9
sets are at +/- 6

http://jsbin.com/olezil/1/

so we have 359 games until now.

NEED MOAR SAMPLES!

Edited by TexAss, 08 March 2013 - 09:24 AM.

[Agent of Change]

TexAss, on 08 March 2013 - 09:23 AM, said:

http://jsbin.com/olezil/1/

so we have 359 games until now.

NEED MOAR SAMPLES!

Thanks, my issue was getting it from excel to here in a reasonably readable format.

anyway back to work, i'll update as soon as we get a few more data sets to add.

[Valleye]

Agent, Did you start a new thread?

The variance is flattening.

V.

[Agent of Change]

Valleye, on 08 March 2013 - 10:14 AM, said:

Agent, Did you start a new thread?

The variance is flattening.

V.

I didn't start a new thread but i'm also confused by your statement.

[Agent of Change]

Bump for data.

[Piranha]

Forest Colony______6
Frozen City_______15
Caustic Valley_____5
River City_________9
Forest Colony_Snow_7
Frozen City Night__7
River City Night__12
Alpine Peaks_______5


I always had the feeling, that I get the maps which basically "require" thermal vision (RCN and FC) more often than the others. Of course, the sample is way to small to prove anything, but so far the stats prove my suspicion!

Edited by Piranha, 13 March 2013 - 04:46 AM.

[Esplodin]

I think this is a combination of things:

1) annoying maps are noticed more
2) people exploiting a bug to just not ever see annoying maps, causing other people to be channeled into RCN and FCD more than the should.

I'd be interested in the number of 8 vs less on those two maps. I've not kept track, but from my off the cuff it happens more frequently there too.

[Agent of Change]

Esplodin, on 13 March 2013 - 04:58 AM, said:

I think this is a combination of things:

1) annoying maps are noticed more
2) people exploiting a bug to just not ever see annoying maps, causing other people to be channeled into RCN and FCD more than the should.

I'd be interested in the number of 8 vs less on those two maps. I've not kept track, but from my off the cuff it happens more frequently there too.

moved my general stat tracking/ fresh data collection to a new thread.

Pirhana I have grabbed your stats for the new sample.