RTCW stats analysis for years 2020 and 2021 in NA

import matplotlib
import seaborn as sns
import pandas as pd
import requests
import os
import numpy as np

Load stats from public repository

cache files locally as they will never change again

url2021 = 'https://s3.amazonaws.com/donkanator.com/stats/endseason/stats_na_6_2021.gz'
url2020 = 'https://s3.amazonaws.com/donkanator.com/stats/endseason/stats_na_6_2020.gz'

if not os.path.exists('stats_na_6_2021.gz'):
    with open('stats_na_6_2021.gz', 'wb') as out_file:
        content = requests.get(url2021, stream=True).content
        out_file.write(content)
        print("downloaded 2021 file")
else:
    print("2021 file exists")
    file_size = os.path.getsize('stats_na_6_2021.gz')
print("File Size is :", round(file_size/1024/1024,1), "MB")
        
        
if not os.path.exists('stats_na_6_2020.gz'):
    with open('stats_na_6_2020.gz', 'wb') as out_file:
        content = requests.get(url2020, stream=True).content
        out_file.write(content)
        print("downloaded 2020 file")
else:
    print("2020 file exists")
    file_size = os.path.getsize('stats_na_6_2020.gz')
print("File Size is :", round(file_size/1024/1024,1), "MB")

2021 file exists
File Size is : 1.4 MB
2020 file exists
File Size is : 0.8 MB

# Basics
stats2020 = pd.read_parquet("stats_na_6_2020.gz")
print("Stats2020 :", len(stats2020), " lines")
stats2021 = pd.read_parquet("stats_na_6_2021.gz")
print("Stats2021 :", len(stats2021), " lines")

Stats2020 : 39085  lines
Stats2021 : 65498  lines

#Bring the files together and see differences
stats= pd.concat([stats2020,stats2021], axis=0, ignore_index=True)
print("For you information: Different columns across datasets:")
set(stats2021.columns).symmetric_difference(set(stats2020.columns))

For you information: Different columns across datasets:

{'Accuracy', 'Headshots', 'Revives'}

#Total number of kills per year
stats['year'] = pd.DatetimeIndex(stats['match_date']).year
stats['quarter'] = pd.DatetimeIndex(stats['match_date']).quarter #this is for later
stats['month'] = pd.DatetimeIndex(stats['match_date']).month #this is for later
stats[["year","Kills"]].groupby("year").sum()

	Kills
year
2020	250477
2021	451438

#Set pandas format
pd.set_option('display.max_rows', 100)
pd.set_option("display.max_columns", 15)
pd.set_option("display.max_colwidth", 20)
pd.set_option("display.width", 400)
pd.options.display.float_format = '{:.2f}'.format

Inspect the dataset

stats.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 104583 entries, 0 to 104582
Data columns (total 56 columns):
 #   Column               Non-Null Count   Dtype  
---  ------               --------------   -----  
 0   AdjScore             104583 non-null  int32  
 1   Airstrike            104583 non-null  int32  
 2   All_Deaths           104583 non-null  int32  
 3   Artillery            104583 non-null  int32  
 4   Colt                 104583 non-null  int32  
 5   Deaths               104583 non-null  int32  
 6   Dynamite             104583 non-null  int32  
 7   Flame                104583 non-null  int32  
 8   Grenade              104583 non-null  int32  
 9   Killer               104583 non-null  object 
 10  Kills                104583 non-null  int32  
 11  Knife                104583 non-null  int32  
 12  Luger                104583 non-null  int32  
 13  MG42                 104583 non-null  int32  
 14  MP40                 104583 non-null  int32  
 15  OSP_Damage_Given     104583 non-null  int32  
 16  OSP_Damage_Received  104583 non-null  int32  
 17  OSP_Deaths           104583 non-null  int32  
 18  OSP_Eff              104583 non-null  int32  
 19  OSP_Gibs             104583 non-null  int32  
 20  OSP_Kills            104583 non-null  int32  
 21  OSP_Player           104583 non-null  object 
 22  OSP_Score            104583 non-null  int32  
 23  OSP_Suicides         104583 non-null  int32  
 24  OSP_TK               104583 non-null  int32  
 25  OSP_Team             104583 non-null  object 
 26  OSP_Team_Damage      104583 non-null  int32  
 27  Panzerfaust          104583 non-null  int32  
 28  Sniper               104583 non-null  int32  
 29  Sten                 104583 non-null  int32  
 30  Suicides             104583 non-null  int32  
 31  TK                   104583 non-null  int32  
 32  TKd                  104583 non-null  int32  
 33  Thompson             104583 non-null  int32  
 34  Venom                104583 non-null  int32  
 35  class                104583 non-null  object 
 36  game_result          104583 non-null  object 
 37  map                  104583 non-null  object 
 38  match_date           104583 non-null  object 
 39  osp_guid             104583 non-null  object 
 40  pb_guid              104583 non-null  object 
 41  player_strip         104583 non-null  object 
 42  round_diff           104583 non-null  int32  
 43  round_guid           104583 non-null  object 
 44  round_num            104583 non-null  int32  
 45  round_order          104583 non-null  int32  
 46  round_time           104583 non-null  int32  
 47  round_win            104583 non-null  int32  
 48  side                 104583 non-null  object 
 49  team_name            104583 non-null  object 
 50  Accuracy             65498 non-null   float64
 51  Headshots            65498 non-null   float64
 52  Revives              65498 non-null   float64
 53  year                 104583 non-null  int64  
 54  quarter              104583 non-null  int64  
 55  month                104583 non-null  int64  
dtypes: float64(3), int32(37), int64(3), object(13)
memory usage: 29.9+ MB

stats.iloc[:,30:45].describe()

	Suicides	TK	TKd	Thompson	Venom	round_diff	round_num
count	104583.00	104583.00	104583.00	104583.00	104583.00	104583.00	104583.00
mean	1.73	0.39	0.39	1.63	0.01	156.59	1.50
std	2.00	0.71	0.67	2.97	0.20	183.74	0.50
min	0.00	0.00	0.00	0.00	0.00	-180.00	1.00
25%	0.00	0.00	0.00	0.00	0.00	0.00	1.00
50%	1.00	0.00	0.00	0.00	0.00	87.00	1.00
75%	2.00	1.00	1.00	2.00	0.00	300.00	2.00
max	46.00	12.00	8.00	30.00	21.00	811.00	2.00

stats.describe(include=['O'])

	Killer	OSP_Player	OSP_Team	class	game_result	map	match_date	osp_guid	pb_guid	player_strip	round_guid	side	team_name
count	104583	104583	104583	104583	104583	104583	104583	104583	104583	104583	104583	104583	104583
unique	503	1903	3	6	5	28	8474	2651	79	1	8617	3	1
top	john_mullins	cakel	Axis		R1MSB	Ice	2020-04-02 23:28:15			notused	9b16cb2ef5e01515...	Defense	notused
freq	4134	3553	51615	73958	52504	19970	174	70711	99923	104583	26	51614	104583

stats.shape

(104583, 56)

What data looks like

stats.head(10)

	AdjScore	Airstrike	All_Deaths	Artillery	Colt	Deaths	Dynamite	...	team_name	Accuracy	Headshots	Revives	year	quarter	month
0	10	1	10	2	0	7	0	...	notused	NaN	NaN	NaN	2020	2	4
1	9	0	8	0	0	3	0	...	notused	NaN	NaN	NaN	2020	2	4
2	42	0	6	0	0	5	0	...	notused	NaN	NaN	NaN	2020	2	4
3	18	1	12	0	0	9	0	...	notused	NaN	NaN	NaN	2020	2	4
4	20	0	6	0	0	3	0	...	notused	NaN	NaN	NaN	2020	2	4
5	2	0	9	0	0	8	0	...	notused	NaN	NaN	NaN	2020	2	4
6	6	2	7	0	0	5	0	...	notused	NaN	NaN	NaN	2020	2	4
7	24	0	5	0	0	4	0	...	notused	NaN	NaN	NaN	2020	2	4
8	8	0	7	0	0	4	0	...	notused	NaN	NaN	NaN	2020	2	4
9	13	0	10	0	0	6	0	...	notused	NaN	NaN	NaN	2020	2	4

10 rows × 56 columns

More meaningful column selection

stats[["round_guid", "map", "match_date" , "Killer", "OSP_Team","Kills", "OSP_Damage_Given", "game_result", "round_num", "round_win"]].head(20)

	round_guid	map	match_date	Killer	OSP_Team	Kills	OSP_Damage_Given	game_result	round_num	round_win
0	00a51f47c49706a2...	Ice	2020-04-08 18:54:37	murkey	Axis	9	2202	R1MSB	1	1
1	00a51f47c49706a2...	Ice	2020-04-08 18:54:37	cliffdark	Axis	3	406	R1MSB	1	1
2	00a51f47c49706a2...	Ice	2020-04-08 18:54:37	chinamann	Axis	8	1350	R1MSB	1	1
3	00a51f47c49706a2...	Ice	2020-04-08 18:54:37	jimmy	Allies	7	1308	R1MSB	1	0
4	00a51f47c49706a2...	Ice	2020-04-08 18:54:37	mean	Allies	7	1361	R1MSB	1	0
5	00a51f47c49706a2...	Ice	2020-04-08 18:54:37	miles	Allies	1	818	R1MSB	1	0
6	800f707fc5f4f006...	Ice	2020-04-08 19:00:58	murkey	Allies	11	4983	WON	2	1
7	800f707fc5f4f006...	Ice	2020-04-08 19:00:58	cliffdark	Allies	5	1559	WON	2	1
8	800f707fc5f4f006...	Ice	2020-04-08 19:00:58	chinamann	Allies	7	2636	WON	2	1
9	800f707fc5f4f006...	Ice	2020-04-08 19:00:58	jimmy	Axis	2	2388	LOST	2	0
10	800f707fc5f4f006...	Ice	2020-04-08 19:00:58	mean	Axis	8	2930	LOST	2	0
11	800f707fc5f4f006...	Ice	2020-04-08 19:00:58	miles	Axis	3	1433	LOST	2	0
12	ad23b7d74145c9e5...	Ice	2020-04-08 19:06:39	murkey	Allies	7	874	R1MSB	1	0
13	ad23b7d74145c9e5...	Ice	2020-04-08 19:06:39	cliffdark	Allies	2	607	R1MSB	1	0
14	ad23b7d74145c9e5...	Ice	2020-04-08 19:06:39	chinamann	Allies	6	1491	R1MSB	1	0
15	ad23b7d74145c9e5...	Ice	2020-04-08 19:06:39	jimmy	Axis	4	898	R1MSB	1	1
16	ad23b7d74145c9e5...	Ice	2020-04-08 19:06:39	mean	Axis	5	1036	R1MSB	1	1
17	ad23b7d74145c9e5...	Ice	2020-04-08 19:06:39	miles	Axis	2	508	R1MSB	1	1
18	c805865abb22ac97...	Ice	2020-04-08 19:10:08	murkey	Axis	4	1440	WON	2	1
19	c805865abb22ac97...	Ice	2020-04-08 19:10:08	cliffdark	Axis	4	1502	WON	2	1

Clean the dataset

Some cleaning steps I have discovered mostly after the fact

print("Lines before general cleansing: ", len(stats))

stats_original = stats.copy()

#Remove some aliases
stats = stats[~stats["Killer"].isin(["WolfPlayer9999", "blackmagik0inc.", "blackmagik7inc."])]

#Remove unknown maps
stats = stats[~stats["map"].isin(['Not listed', 'No info'])]

#Remove disconnected players that have no OSP stats and team affiliation
stats = stats[~stats["OSP_Team"].isin(['0'])]

print("Lines after general cleansing: ", len(stats))

Lines before general cleansing:  104583
Lines after general cleansing:  102828

#Remove games with less than 5v5 players
player_nums = stats["round_guid"].value_counts()
games_under5_players = player_nums[player_nums < 9].index.values
print("Lines before removing smaller matches: ", len(stats))
stats = stats[~stats['round_guid'].isin(games_under5_players)]
print("Lines after removing smaller matches: ", len(stats))

Lines before removing smaller matches:  102828
Lines after removing smaller matches:  99580

#Remove some duplicate players probably caused by rename process
stats.drop_duplicates(["Killer", "round_guid"],inplace=True)

# player_dups = stats.groupby(["Killer", "round_guid"]).count()["round_num"].reset_index()
# player_dups[player_dups["round_num"]>1].head(100)

Analysis

Set the stage - who are the players and what are the maps?

#We will work with round2 (final) stats
tmp_heat = stats[stats["round_num"] == 2]

#players
tmp_games = tmp_heat[["Killer", "round_guid"]].groupby(["Killer"]).agg({'round_guid' : "count"})
tmp_games.sort_values(by="round_guid", ascending=False).head(30)
#limit players to at least 120 games
tmp_games_filt = tmp_games[tmp_games["round_guid"]>120].sort_values(by="round_guid")

#Maps
tmp_maps = tmp_heat[["map", "round_guid"]].groupby(["map"]).agg({'round_guid' : pd.Series.nunique})
tmp_maps.sort_values(by="round_guid", ascending=False).head(20)

	round_guid
map
Ice	760
Frostbite	647
Village	501
Adlernest	446
Beach	386
Base	322
Cipher	226
Escape	184
Password	181
Assault	147
Sub	113
UFO	41
Tunordic	29
Tundra	17
Chateau	16
Brewdog	11
Delivery	10
Radar	6
Keep	6
Tram	4

Player win percentages

tmp_wins = stats[stats["round_num"] == 2]
tmp_wins = tmp_wins[tmp_wins["Killer"].isin(tmp_games_filt.index.values)]
tmp_wins = tmp_wins.groupby(["Killer","game_result"]).count()["Kills"].reset_index()
tmp_wins.columns = ["Killer","game_result","Count"]
tmp_wins =  tmp_wins.pivot(index="Killer",columns="game_result", values="Count")
tmp_wins["Total"] = tmp_wins.sum(axis=1)
tmp_wins["Win%"] = tmp_wins["WON"]/tmp_wins["Total"]
tmp_wins["Draw%"] = tmp_wins["FULLHOLD"]/tmp_wins["Total"]
tmp_wins["Loss%"] = tmp_wins["LOST"]/tmp_wins["Total"]

pd.options.display.float_format = '{:.1%}'.format
tmp_wins[["Win%","Draw%","Loss%"]].sort_values(by="Win%", ascending=False).head(50)

game_result	Win%	Draw%	Loss%
Killer
diego	54.9%	4.5%	40.6%
kazz	53.5%	8.0%	38.5%
brandon	52.7%	5.6%	41.8%
pingrage	52.4%	9.7%	37.8%
ding	51.8%	5.9%	42.2%
murkey	51.7%	7.3%	41.0%
druwin	50.9%	6.9%	42.2%
vein	50.9%	4.7%	44.4%
xill	50.9%	6.1%	43.1%
ipod	50.4%	4.7%	44.9%
pimp	50.0%	8.0%	42.0%
yeniceri	50.0%	4.4%	45.6%
kittens	50.0%	7.1%	42.9%
john_mullins	49.9%	7.4%	42.7%
fro	49.9%	7.1%	43.1%
raiser	49.0%	7.3%	43.6%
miles	48.8%	6.3%	44.9%
mmbird	48.8%	7.2%	44.0%
jaytee	48.7%	8.3%	43.0%
bru	48.6%	7.4%	44.0%
donka	48.4%	6.8%	44.8%
illkilla	48.3%	5.1%	46.6%
tragic	48.2%	8.9%	42.9%
reker	48.0%	6.4%	45.6%
yyz	47.8%	6.8%	45.3%
souldriver	47.7%	6.1%	46.2%
source	47.6%	6.5%	45.9%
flogzero	47.5%	7.9%	44.6%
donkey	47.4%	6.7%	45.9%
detdet	47.2%	3.6%	49.2%
conscious	47.1%	7.0%	45.9%
warrior	47.1%	7.3%	45.6%
rob	46.8%	7.5%	45.7%
shaz	46.7%	5.2%	48.1%
doza	46.6%	5.9%	47.5%
nigel	46.4%	7.1%	46.5%
cliffdark	46.3%	6.6%	47.1%
parcher	46.2%	7.0%	46.7%
dillweed	45.9%	7.3%	46.8%
fonze	45.7%	7.6%	46.7%
joep	45.7%	5.1%	49.1%
blackmagik	45.6%	7.0%	47.4%
cakel	45.6%	7.2%	47.2%
firehot	45.5%	6.8%	47.7%
sem	45.3%	3.4%	51.4%
jam	45.0%	10.0%	45.0%
luna	45.0%	8.5%	46.5%
mooshu	45.0%	7.0%	48.0%
eternal	44.9%	5.8%	49.3%
jimmy	44.7%	4.8%	50.5%

Here we probably need to state that the games are handpicked/adjusted for fairness and those with higher percentages are probably performing better than expected, people down the middle are just right, and the bottom is not shown.

Player wins by map

#Get heatmap of player vs map by win percentage
#limit maps to at least 40 games played
tmp_maps_filt = tmp_maps[tmp_maps["round_guid"]>40].sort_values(by="round_guid")

#This is a lazy count of round2 holds. Winning* includes fullholds here
tmp_heat = tmp_heat[["Killer","map", "round_win", "round_guid"]].groupby(["Killer","map"]).agg({'round_win' : "sum", 'round_guid' : "count"}).reset_index()
tmp_heat = tmp_heat[tmp_heat["Killer"].isin(tmp_games_filt.index.values)]
tmp_heat = tmp_heat[tmp_heat["map"].isin(tmp_maps_filt.index.values)]
tmp_heat = tmp_heat[tmp_heat["round_guid"]>4]
tmp_heat["win%"] = tmp_heat["round_win"]/tmp_heat["round_guid"]

Least favorite maps per player

tmp_heat.sort_values(by="win%").head(50)

	Killer	map	round_win	round_guid	win%
1156	jam	Cipher	0	14	0.0%
2031	rob	UFO	0	5	0.0%
488	conscious	Escape	0	5	0.0%
1207	jimmy	Password	0	8	0.0%
539	cypher	Assault	0	5	0.0%
608	detdet	Assault	1	7	14.3%
2385	tragic	UFO	1	7	14.3%
332	bru	Sub	2	14	14.3%
1330	kep	Sub	1	6	16.7%
2520	vulki	Escape	1	6	16.7%
2229	souldriver	Assault	1	6	16.7%
497	conscious	UFO	1	6	16.7%
793	duster	Village	2	12	16.7%
1234	joep	Sub	1	6	16.7%
226	blackmagik	Escape	2	10	20.0%
941	festus	UFO	1	5	20.0%
2515	vulki	Adlernest	2	10	20.0%
968	flogzero	Escape	2	9	22.2%
299	brandon	Sub	2	9	22.2%
2238	souldriver	UFO	2	8	25.0%
2093	sem	Cipher	4	16	25.0%
1170	jaytee	Base	3	12	25.0%
351	caffeine	Password	3	12	25.0%
1478	mean	UFO	2	8	25.0%
2557	warrior	Password	3	12	25.0%
2553	warrior	Escape	6	24	25.0%
377	cakel	Sub	6	23	26.1%
2175	silentstorm	Password	6	23	26.1%
2525	vulki	Village	5	19	26.3%
937	festus	Password	7	26	26.9%
616	detdet	Sub	5	18	27.8%
1875	prowler	Password	12	43	27.9%
545	cypher	Frostbite	7	25	28.0%
2469	virkes	Ice	7	25	28.0%
1328	kep	Password	2	7	28.6%
1204	jimmy	Cipher	4	14	28.6%
2609	xill	Assault	2	7	28.6%
2519	vulki	Cipher	2	7	28.6%
2002	risk	Sub	2	7	28.6%
1300	kazz	Adlernest	8	28	28.6%
2470	virkes	Password	2	7	28.6%
511	corpse	Password	5	17	29.4%
1606	nigel	Base	23	77	29.9%
784	duster	Assault	3	10	30.0%
1464	mean	Adlernest	3	10	30.0%
1175	jaytee	Cipher	3	10	30.0%
814	elsa	Base	17	56	30.4%
713	donkey	Assault	11	36	30.6%
540	cypher	Base	4	13	30.8%
1531	mmbird	Cipher	5	16	31.2%

Most effective maps per player

tmp_heat.sort_values(by="win%", ascending=False).head(50)

	Killer	map	round_win	round_guid	win%
2044	robmaster	UFO	5	5	100.0%
1784	pimp	Password	5	5	100.0%
1182	jaytee	Sub	5	5	100.0%
1518	miles	UFO	10	11	90.9%
1310	kazz	Password	15	17	88.2%
627	diego	Cipher	18	21	85.7%
1562	mooshu	UFO	12	14	85.7%
1169	jaytee	Assault	11	13	84.6%
1168	jaytee	Adlernest	5	6	83.3%
787	duster	Cipher	5	6	83.3%
631	diego	Password	5	6	83.3%
1993	risk	Assault	8	10	80.0%
513	corpse	Sub	4	5	80.0%
554	cypher	Village	12	15	80.0%
748	doza	UFO	4	5	80.0%
665	ding	Assault	8	10	80.0%
1527	mmbird	Adlernest	9	12	75.0%
336	bru	UFO	6	8	75.0%
2263	source	UFO	9	12	75.0%
1319	kep	Assault	6	8	75.0%
1797	pingrage	Cipher	14	19	73.7%
1115	illkilla	Password	14	19	73.7%
2027	rob	Password	13	18	72.2%
2616	xill	Password	5	7	71.4%
2653	yyz	Beach	10	14	71.4%
235	blackmagik	UFO	5	7	71.4%
291	brandon	Beach	29	41	70.7%
2654	yyz	Cipher	7	10	70.0%
1320	kep	Base	7	10	70.0%
1313	kazz	Village	18	26	69.2%
2548	warrior	Assault	18	26	69.2%
1910	raiser	Sub	17	25	68.0%
768	druwin	Escape	27	40	67.5%
729	donkey	UFO	4	6	66.7%
783	duster	Adlernest	12	18	66.7%
2632	yeniceri	Assault	8	12	66.7%
727	donkey	Sub	22	33	66.7%
2448	vein	Sub	4	6	66.7%
1321	kep	Beach	4	6	66.7%
1792	pingrage	Adlernest	10	15	66.7%
2427	vacs	Adlernest	4	6	66.7%
1778	pimp	Base	10	15	66.7%
1876	prowler	Sub	4	6	66.7%
1181	jaytee	Password	12	18	66.7%
1305	kazz	Cipher	8	12	66.7%
1029	fro	UFO	6	9	66.7%
763	druwin	Adlernest	39	59	66.1%
626	diego	Beach	27	41	65.9%
1620	nigel	Password	44	67	65.7%
666	ding	Base	21	32	65.6%

Heatmap graph

#Prepare the graph
tmp_heat_pivot = tmp_heat.pivot(index="Killer",columns="map", values="win%")
sns.set(rc={'figure.figsize':(20,25)})
s = sns.heatmap(tmp_heat_pivot, cbar=True, cmap='viridis', annot=True, fmt='.1%')
for t in s.texts: t.set_text(t.get_text()+ "%")
s.tick_params(top=True, labeltop=True)
s.tick_params(axis='x', labelrotation=90)
s.set_xlabel('X-Axis', fontsize=0)
s.set_ylabel('Y-Axis', fontsize=0)

Text(156.453125, 0.5, 'Y-Axis')

People we haven't seen in a while

tmp_last_seen = stats[["Killer","round_guid","match_date"]].groupby(["Killer"]).agg({'round_guid' : "count",'match_date' : np.max}).copy()
tmp_last_seen.columns = ["Rounds_played", "Last_seen"]
tmp_last_seen = tmp_last_seen[tmp_last_seen["Rounds_played"]>40].sort_values("Last_seen").reset_index()

tmp_last_seen['year'] = pd.DatetimeIndex(tmp_last_seen['Last_seen']).year
tmp_last_seen['month'] = pd.DatetimeIndex(tmp_last_seen['Last_seen']).month
tmp_last_seen[((tmp_last_seen["year"]==2021) & (tmp_last_seen["month"]<12)) | (tmp_last_seen["year"]<2021)]

	Killer	Rounds_played	Last_seen	year	month
0	sonar	75	2020-04-08 21:49:03	2020	4
1	boydarilla	67	2020-04-24 00:09:12	2020	4
2	kotip	78	2020-05-10 21:02:57	2020	5
3	paper	146	2020-05-11 23:27:44	2020	5
4	oliokath	169	2020-07-30 22:43:17	2020	7
5	gut	142	2020-08-11 17:25:09	2020	8
6	dash	42	2020-08-12 23:29:02	2020	8
7	spuddy	43	2020-08-13 21:43:46	2020	8
8	sloth	74	2020-08-20 23:16:48	2020	8
9	aristotle	127	2020-08-30 23:34:07	2020	8
10	mean	529	2020-09-24 23:11:19	2020	9
11	dresserwood	147	2020-10-17 22:56:55	2020	10
12	duster	282	2020-10-25 22:22:36	2020	10
13	pasek	139	2020-11-01 22:02:50	2020	11
14	wang	42	2020-11-05 22:16:06	2020	11
15	knifey	43	2020-11-05 23:06:02	2020	11
16	scrilla	98	2020-11-12 22:33:46	2020	11
17	booty	75	2020-11-30 22:21:44	2020	11
18	caffeine	295	2020-12-04 23:47:49	2020	12
19	fox	105	2020-12-12 22:43:41	2020	12
20	nizou	56	2021-01-09 22:00:10	2021	1
21	scrill4	89	2021-01-15 05:38:37	2021	1
22	ceres	63	2021-01-16 22:50:20	2021	1
23	vodka	201	2021-01-23 04:35:00	2021	1
24	flogzero	1238	2021-03-18 21:56:18	2021	3
25	fatihot	62	2021-04-18 22:10:12	2021	4
26	lasher	86	2021-05-15 00:47:23	2021	5
27	cash	158	2021-05-17 03:04:28	2021	5
28	jimmy	583	2021-05-17 04:14:16	2021	5
29	cliffdark	491	2021-05-22 21:31:50	2021	5
30	pimp	352	2021-05-24 04:17:06	2021	5
31	kep	263	2021-06-06 23:37:00	2021	6
32	kazz	430	2021-06-29 21:33:31	2021	6
33	blaz	41	2021-07-07 00:39:32	2021	7
34	kaiser	66	2021-07-14 23:22:19	2021	7
35	shat	187	2021-08-17 02:16:45	2021	8
36	naper	44	2021-08-19 04:04:59	2021	8
37	vis	119	2021-08-20 19:44:55	2021	8
38	luna	257	2021-08-20 20:39:50	2021	8
39	tragic	384	2021-08-20 20:39:50	2021	8
40	virus	62	2021-08-20 20:39:50	2021	8
41	cypher	265	2021-08-20 20:39:50	2021	8
42	siluro	67	2021-08-20 20:39:50	2021	8
43	blox	48	2021-08-21 02:19:21	2021	8
44	blackmagik	1192	2021-08-22 23:15:39	2021	8
45	sixers	45	2021-08-23 23:03:04	2021	8
46	prowler	1314	2021-08-31 04:43:45	2021	8
47	cky	227	2021-09-02 03:22:35	2021	9
48	elver	57	2021-09-03 00:01:35	2021	9
49	biggi	143	2021-09-05 07:10:32	2021	9
50	gorthrax	82	2021-09-15 03:37:47	2021	9
51	gatorhead	74	2021-09-16 23:01:12	2021	9
52	rev9	108	2021-09-30 21:18:11	2021	9
53	raiser	1893	2021-10-04 03:09:34	2021	10
54	shaz	267	2021-10-08 03:35:27	2021	10
55	jaytee	456	2021-10-11 01:44:57	2021	10
56	shrek	53	2021-10-21 00:06:30	2021	10
57	corpse	665	2021-11-06 02:07:55	2021	11
58	playa	90	2021-11-06 03:11:33	2021	11
59	utrolig	153	2021-11-06 03:11:33	2021	11
60	vacs	282	2021-11-10 21:49:38	2021	11
61	jam	562	2021-11-11 22:31:40	2021	11
62	vorg	51	2021-11-13 04:20:35	2021	11
63	sem	295	2021-11-14 19:42:42	2021	11
64	zed	44	2021-11-21 23:43:57	2021	11
65	kamizama	69	2021-11-22 23:51:06	2021	11
66	conscious	1188	2021-11-29 23:48:21	2021	11

Shortest rounds ever played

tmp_map_times= stats[stats["round_num"] == 2]
shortest_rounds = tmp_map_times[["round_guid","round_time"]].groupby(["round_guid"]).agg({'round_time' : np.min}).sort_values(by="round_time").head(3).index.values
stats[stats["round_guid"].isin(shortest_rounds)][["map", "match_date", "Killer", "side", "game_result", "round_time"]].sort_values(by=["round_time","match_date", "side"])

	map	match_date	Killer	side	game_result	round_time
47344	Adlernest	2021-08-19 22:07:20	knicks9999	Defense	LOST	56
47345	Adlernest	2021-08-19 22:07:20	blox	Defense	LOST	56
47352	Adlernest	2021-08-19 22:07:20	biggi	Defense	LOST	56
47353	Adlernest	2021-08-19 22:07:20	festus	Defense	LOST	56
47354	Adlernest	2021-08-19 22:07:20	mika9999	Defense	LOST	56
47355	Adlernest	2021-08-19 22:07:20	pug9999	Defense	LOST	56
47346	Adlernest	2021-08-19 22:07:20	ipod	Offense	WON	56
47347	Adlernest	2021-08-19 22:07:20	john_mullins	Offense	WON	56
47348	Adlernest	2021-08-19 22:07:20	raiser	Offense	WON	56
47349	Adlernest	2021-08-19 22:07:20	sem	Offense	WON	56
47350	Adlernest	2021-08-19 22:07:20	cky	Offense	WON	56
47351	Adlernest	2021-08-19 22:07:20	eternal	Offense	WON	56
66039	Frostbite	2021-07-15 00:12:24	ipod	Defense	WON	62
66041	Frostbite	2021-07-15 00:12:24	parcher	Defense	WON	62
66042	Frostbite	2021-07-15 00:12:24	donkey	Defense	WON	62
66043	Frostbite	2021-07-15 00:12:24	gatorhead	Defense	WON	62
66044	Frostbite	2021-07-15 00:12:24	murkey	Defense	WON	62
66045	Frostbite	2021-07-15 00:12:24	fonze	Defense	WON	62
66038	Frostbite	2021-07-15 00:12:24	brandon	Offense	LOST	62
66040	Frostbite	2021-07-15 00:12:24	john_mullins	Offense	LOST	62
66046	Frostbite	2021-07-15 00:12:24	kittens	Offense	LOST	62
66047	Frostbite	2021-07-15 00:12:24	vacs	Offense	LOST	62
66048	Frostbite	2021-07-15 00:12:24	donka	Offense	LOST	62
66049	Frostbite	2021-07-15 00:12:24	nigel	Offense	LOST	62
7660	Beach	2020-08-26 23:18:16	raiser	Defense	LOST	64
7661	Beach	2020-08-26 23:18:16	conscious	Defense	LOST	64
7662	Beach	2020-08-26 23:18:16	donkey	Defense	LOST	64
7663	Beach	2020-08-26 23:18:16	spaztik	Defense	LOST	64
7667	Beach	2020-08-26 23:18:16	donka	Defense	LOST	64
7668	Beach	2020-08-26 23:18:16	fenix	Defense	LOST	64
7664	Beach	2020-08-26 23:18:16	fro	Offense	WON	64
7665	Beach	2020-08-26 23:18:16	john_mullins	Offense	WON	64
7666	Beach	2020-08-26 23:18:16	murkey	Offense	WON	64
7669	Beach	2020-08-26 23:18:16	fonze	Offense	WON	64
7670	Beach	2020-08-26 23:18:16	pingrage	Offense	WON	64
7671	Beach	2020-08-26 23:18:16	rob	Offense	WON	64

Shortest rounds per map and Average map durations

tmp_map_times_res = tmp_map_times[["round_time","map","OSP_Kills"]].groupby(["map"]).agg({"round_time": [np.min, np.mean], "OSP_Kills": "mean"})
tmp_map_times_res.columns = ["Fastest(sec)", "Average(min)", "Kills Per Person/Min"]
tmp_map_times_res["Average(min)"] = tmp_map_times_res["Average(min)"]/60
tmp_map_times_res["Kills Per Person/Min"] = tmp_map_times_res["Kills Per Person/Min"]/60
pd.options.display.float_format = '{:.2f}'.format
tmp_map_times_res

	Fastest(sec)	Average(min)	Kills Per Person/Min
map
Adlernest	56	4.69	0.26
Assault	113	5.78	0.18
Base	85	3.94	0.15
Beach	64	5.31	0.20
Braundorf	199	4.12	0.17
Breakout	208	3.64	0.14
Brewdog	126	4.36	0.22
Castle	228	3.80	0.16
Chateau	98	2.75	0.12
Church	260	4.33	0.18
Cipher	82	4.96	0.25
Delivery	212	4.45	0.23
Escape	98	4.25	0.22
Frostbite	62	4.43	0.21
Ice	87	5.64	0.24
Keep	132	5.51	0.25
Kungfugrip	92	2.02	0.04
Market garden	1009	16.82	0.28
Password	135	6.81	0.28
Radar	233	5.28	0.26
Sub	82	3.13	0.15
Tram	479	15.05	0.50
Tundra	99	7.00	0.25
Tunordic	79	6.42	0.25
UFO	99	6.83	0.22
Village	144	6.49	0.23

Best individual performance in any match

Blank class is either engineer or (most likely) a medic

tmp_best_perf = stats[stats["round_num"] == 2]
tmp_best_perf = tmp_best_perf["Kills"].sort_values().tail(10)
stats[stats["Kills"].isin(tmp_best_perf)][["map", "match_date", "Killer", "Kills", "class", "round_time"]].sort_values(by=["Kills"], ascending=False)

	map	match_date	Killer	Kills	class	round_time
315	Tram	2020-04-10 00:54:19	eternal	43	L	1200
28373	Password	2020-05-10 23:11:16	bru	39	P	900
100642	Adlernest	2021-09-12 07:29:49	parcher	39		600
78023	Frostbite	2021-05-12 00:17:51	ipod	39	L	600
61695	Password	2021-01-16 22:24:12	donkey	39	P	900
33203	Password	2020-10-15 22:28:10	john_mullins	39	L	878
54340	Adlernest	2021-12-25 00:17:28	LitoriousJackson...	37	P	592
309	Tram	2020-04-10 00:54:19	source	37	P	1200
293	Tram	2020-04-10 00:33:31	murkey	37	S	1200
33680	Adlernest	2020-10-18 00:34:27	joep	37	P	600
50066	Ice	2021-08-30 23:24:26	fro	36	P	600
59769	Base	2021-01-04 23:42:02	john_mullins	36	L	695
10517	Password	2020-12-17 22:49:29	fatihot	36	P	900
13893	Ice	2020-02-27 22:23:21	source	36	L	575
56052	Adlernest	2021-02-06 22:58:13	john_mullins	36		596
66328	Adlernest	2021-07-16 22:08:11	brandon	36	P	600
3502	Tram	2020-04-03 23:59:16	kittens	36		1200
2549	Ice	2020-04-01 21:08:19	brandon	36	P	600
94035	Adlernest	2021-10-20 23:44:52	warrior	36		600
60293	Password	2021-01-08 00:02:39	john_mullins	35	L	900
57033	Adlernest	2021-02-16 21:44:22	fro	35	P	600
50276	Adlernest	2021-12-01 22:15:55	ipod	35	P	600
103142	Ice	2021-09-24 22:08:16	john_mullins	35	P	600
56041	Adlernest	2021-02-06 22:46:29	kittens	35	P	596
50707	Adlernest	2021-12-02 23:16:29	ipod	35	P	600
4051	Password	2020-04-23 22:57:21	kittens	35		900
3515	Tram	2020-04-03 23:59:16	murkey	35	S	1200
3111	Market garden	2020-04-02 23:13:33	source	35		1009
318	Tram	2020-04-10 00:54:19	murkey	35		1200
46799	Frostbite	2021-08-17 01:32:56	druwin	34	P	463
32817	Tunordic	2020-10-14 22:03:42	john_mullins	34	L	600
80450	Password	2021-05-30 01:11:05	fro	34	P	815
86276	Adlernest	2021-11-22 23:23:32	ipod	34	P	600
86335	Ice	2021-11-24 04:18:33	ipod	34	P	600
305	Tram	2020-04-10 00:54:19	parcher	34		1200
101899	Escape	2021-09-18 05:34:41	murkey	34		600
102180	Adlernest	2021-09-19 01:50:15	john_mullins	34	P	590
288	Tram	2020-04-10 00:33:31	donka	34		1200

Best individual performance in a 6v6

games_over6_players = player_nums[player_nums > 12].index.values
tmp_best_perf_2 = stats[~stats['round_guid'].isin(games_over6_players)]
tmp_best_perf_2 = tmp_best_perf_2["Kills"].sort_values().tail(10)
stats[stats["Kills"].isin(tmp_best_perf_2)][["map", "match_date", "Killer", "Kills", "class", "round_time"]].sort_values(by=["Kills"], ascending=False)

	map	match_date	Killer	Kills	class	round_time
283	Tram	2020-04-10 00:33:31	source	41	S	1200
76956	Adlernest	2021-05-07 01:05:20	biggi	41	S	415
28373	Password	2020-05-10 23:11:16	bru	39	P	900
78023	Frostbite	2021-05-12 00:17:51	ipod	39	L	600
61695	Password	2021-01-16 22:24:12	donkey	39	P	900
33203	Password	2020-10-15 22:28:10	john_mullins	39	L	878
100642	Adlernest	2021-09-12 07:29:49	parcher	39		600
33680	Adlernest	2020-10-18 00:34:27	joep	37	P	600
293	Tram	2020-04-10 00:33:31	murkey	37	S	1200
54340	Adlernest	2021-12-25 00:17:28	LitoriousJackson...	37	P	592
309	Tram	2020-04-10 00:54:19	source	37	P	1200
13893	Ice	2020-02-27 22:23:21	source	36	L	575
10517	Password	2020-12-17 22:49:29	fatihot	36	P	900
56052	Adlernest	2021-02-06 22:58:13	john_mullins	36		596
59769	Base	2021-01-04 23:42:02	john_mullins	36	L	695
3502	Tram	2020-04-03 23:59:16	kittens	36		1200
66328	Adlernest	2021-07-16 22:08:11	brandon	36	P	600
2549	Ice	2020-04-01 21:08:19	brandon	36	P	600
94035	Adlernest	2021-10-20 23:44:52	warrior	36		600
50066	Ice	2021-08-30 23:24:26	fro	36	P	600

Biggest match ever

player_nums = stats_original["round_guid"].value_counts()
most_players_round_guid = player_nums.head(1)
print("Most players in a single match: ", most_players_round_guid.values[0])
tmp_most_players = stats_original[stats_original['round_guid'].isin(most_players_round_guid.index.values)]

tmp_most_players[["map", "match_date", "Killer", "Kills", "round_time", "OSP_Team"]].sort_values(by=["OSP_Team", "Kills"], ascending=False)

Most players in a single match:  26

	map	match_date	Killer	Kills	round_time	OSP_Team
253	Assault	2020-04-10 00:11:47	parcher	9	406	Axis
257	Assault	2020-04-10 00:11:47	kittens	9	406	Axis
263	Assault	2020-04-10 00:11:47	donka	6	406	Axis
269	Assault	2020-04-10 00:11:47	mooshu	6	406	Axis
254	Assault	2020-04-10 00:11:47	cypher	5	406	Axis
255	Assault	2020-04-10 00:11:47	dillweed	5	406	Axis
258	Assault	2020-04-10 00:11:47	raiser	5	406	Axis
266	Assault	2020-04-10 00:11:47	eternal	5	406	Axis
270	Assault	2020-04-10 00:11:47	murkey	5	406	Axis
272	Assault	2020-04-10 00:11:47	paper	4	406	Axis
275	Assault	2020-04-10 00:11:47	spaztik	4	406	Axis
260	Assault	2020-04-10 00:11:47	wang	1	406	Axis
252	Assault	2020-04-10 00:11:47	john_mullins	0	406	Axis
259	Assault	2020-04-10 00:11:47	source	11	406	Allies
267	Assault	2020-04-10 00:11:47	gut	10	406	Allies
261	Assault	2020-04-10 00:11:47	blackmagik7inc.	9	406	Allies
262	Assault	2020-04-10 00:11:47	caffeine	9	406	Allies
273	Assault	2020-04-10 00:11:47	playa	9	406	Allies
271	Assault	2020-04-10 00:11:47	nigel	8	406	Allies
276	Assault	2020-04-10 00:11:47	spuddy	7	406	Allies
264	Assault	2020-04-10 00:11:47	flogzero	6	406	Allies
265	Assault	2020-04-10 00:11:47	mean	5	406	Allies
268	Assault	2020-04-10 00:11:47	miles	4	406	Allies
274	Assault	2020-04-10 00:11:47	prowler	4	406	Allies
256	Assault	2020-04-10 00:11:47	deadeye	1	406	Allies
277	Assault	2020-04-10 00:11:47	boydarilla	0	406	Allies

Maps with their fullhold percentages

tmp_mapfh = stats[stats["round_num"] == 2].copy()
tmp_mapfh = tmp_mapfh[~tmp_mapfh['round_guid'].isin(games_over6_players)]

tmp_mapfh = tmp_mapfh.groupby(["round_guid","map","game_result"]).count()["Kills"].reset_index()
tmp_mapfh = tmp_mapfh.groupby(["map","game_result"]).count()["Kills"].reset_index()
tmp_mapfh.columns = ["map","game_result","Count"]
tmp_mapfh =  tmp_mapfh.pivot(index="map",columns="game_result", values="Count")

pd.options.display.float_format = '{:.2%}'.format
tmp_mapfh["Draw%"] = tmp_mapfh["FULLHOLD"]/tmp_mapfh["WON"] #or lost, it goes both ways
tmp_mapfh["Draw%"].fillna(0).sort_values()

map
Tundra       0.00%
Tram         0.00%
Brewdog      0.00%
Castle       0.00%
Chateau      0.00%
Church       0.00%
Delivery     0.00%
Sub          0.96%
Base         1.46%
Escape       2.25%
Adlernest    3.05%
Frostbite    3.15%
Cipher       3.30%
Password     4.24%
Beach        9.27%
Ice         10.26%
UFO         13.04%
Assault     13.54%
Village     16.44%
Keep        25.00%
Tunordic    40.00%
Name: Draw%, dtype: float64

Damage throughput and KDR per player over time

This is an attempt to see improvement in players over time

pd.options.display.float_format = '{:.1f}'.format

tmp_improv = stats[stats["round_num"] == 2].copy()
tmp_improv = tmp_improv[~tmp_improv["OSP_Team"].isin(['0',''])]
tmp_improv = tmp_improv[tmp_improv["Killer"].isin(tmp_games_filt.index.values)]
tmp_improv = tmp_improv.groupby(["Killer", "year", "quarter"]) \
                       .agg({"OSP_Damage_Given":"sum", "Kills": "sum", "Deaths": "sum", "round_time" :"sum", "round_win" : "sum", "round_guid" : "count"}) \
                       .reset_index()
tmp_improv["year.q"] = tmp_improv["year"].astype(str) + "." + tmp_improv["quarter"].divide(.04).astype(int).astype(str)
tmp_improv["KDR"] = tmp_improv["Kills"]/tmp_improv["Deaths"]
tmp_improv["DPM"] = tmp_improv["OSP_Damage_Given"]/(tmp_improv["round_time"]/60)
tmp_improv["win%"] = tmp_improv["round_win"]/tmp_improv["round_guid"]
tmp_improv_pivot  = tmp_improv.pivot(index="Killer", columns="year.q", values = ["KDR","DPM"])
tmp_improv_pivot= tmp_improv_pivot[tmp_improv_pivot["DPM"]["2021.75"].notna()] #seed out people who quit

kdr_slopes = {}
dpm_slopes = {}
for player, values in tmp_improv_pivot.iterrows():
    mean_kdr = values["KDR"].mean()
    values["KDR"].fillna(mean_kdr, inplace=True)
    kdr_x = values["KDR"].index.values.astype(float)
    kdr_y = values["KDR"].values
    kdr_slopes[player] = np.polyfit(kdr_x, kdr_y,1)[0]
    
    mean_dpm = values["DPM"].mean()
    values["DPM"].fillna(mean_dpm, inplace=True)
    dpm_x = values["DPM"].index.values.astype(float)
    dpm_y = values["DPM"].values
    dpm_slopes[player] = np.polyfit(dpm_x, dpm_y,1)[0]

kdr_df = pd.DataFrame.from_dict(kdr_slopes, orient='index', columns = ["kdr_slope"])
dpm_df = pd.DataFrame.from_dict(dpm_slopes, orient='index', columns = ["dpm_slope"])
pd.options.display.float_format = '{:.3f}'.format

Most improved players by KDR

tmp_improv_pivot_kdr = tmp_improv_pivot["KDR"].join(kdr_df)
tmp_improv_pivot_kdr[tmp_improv_pivot_kdr["kdr_slope"]>0.02].sort_values(by=["kdr_slope"], ascending=False)

	2020.100	2020.25	2020.50	2020.75	2021.100	2021.25	2021.50	2021.75	kdr_slope
Killer
miles	0.957	0.650	0.896	0.933	1.074	1.164	1.521	1.204	0.350
rob	0.417	0.977	0.977	0.937	1.032	0.902	1.500	1.073	0.335
john_mullins	1.234	0.490	0.621	1.207	1.431	1.250	1.186	1.324	0.331
silentstorm	0.393	0.802	0.802	0.802	0.940	0.792	0.874	1.007	0.229
mooshu	0.749	0.360	0.644	0.669	0.844	0.900	0.871	0.865	0.221
doza	1.304	1.113	0.763	0.593	1.257	1.359	1.275	1.239	0.155
jam	1.355	1.538	1.538	1.538	1.501	1.401	1.820	1.612	0.130
virkes	0.781	0.781	0.781	0.409	0.824	0.781	1.043	0.848	0.112
donkey	1.226	0.872	1.157	1.152	1.108	1.308	1.300	1.132	0.096
fonze	0.881	0.896	0.656	0.986	0.925	0.965	0.905	0.951	0.071
cakel	0.620	0.510	0.588	0.567	0.600	0.626	0.587	0.690	0.052
conscious	0.767	0.819	0.765	0.818	0.819	0.851	0.879	0.831	0.049
robmaster	0.966	0.966	0.966	0.966	0.785	0.968	1.009	1.103	0.043
ipod	1.034	1.034	1.034	1.034	1.185	0.667	1.112	1.172	0.027
corpse	0.932	1.019	0.912	0.750	0.943	0.949	0.953	1.000	0.023

Most improved players by Damage Given per Minute

tmp_improv_pivot_dpm = tmp_improv_pivot["DPM"].join(dpm_df)
tmp_improv_pivot_dpm[tmp_improv_pivot_dpm["dpm_slope"]>0.025].sort_values(by=["dpm_slope"], ascending=False)

	2020.100	2020.25	2020.50	2020.75	2021.100	2021.25	2021.50	2021.75	dpm_slope
Killer
miles	438.269	310.292	386.837	457.548	550.269	520.125	545.217	585.355	136.991
john_mullins	738.901	392.876	485.899	741.356	859.891	720.545	636.479	763.156	121.561
silentstorm	252.664	369.296	369.296	369.296	412.597	339.257	397.625	444.335	69.170
mooshu	365.120	232.480	345.010	468.434	451.595	395.688	427.396	395.331	68.194
conscious	304.481	375.517	339.043	384.003	375.517	431.215	355.988	438.369	52.750
yyz	624.040	624.040	624.040	596.387	519.599	624.040	624.040	756.134	38.369
nigel	461.334	603.504	556.013	532.664	586.037	534.678	527.207	620.218	33.817
yeniceri	228.491	282.257	282.257	282.257	324.583	293.664	265.319	299.226	23.989
donkey	633.793	418.333	567.969	628.135	554.213	576.747	584.162	580.403	22.564
rob	347.712	448.001	448.001	532.803	509.211	421.271	445.232	431.780	22.067
doza	513.143	477.919	467.009	319.481	540.565	510.456	509.639	485.142	19.694
fonze	387.180	381.850	308.635	377.033	416.289	424.204	379.767	379.839	17.871
robmaster	444.375	444.375	444.375	444.375	353.390	480.827	438.035	505.247	17.068
druwin	690.514	704.830	704.830	704.830	735.950	701.606	646.935	749.145	7.823
festus	336.309	363.163	370.572	375.822	375.447	380.802	317.077	372.566	2.246
spaztik	413.281	360.314	429.649	479.164	452.064	427.836	396.019	394.237	1.729
cakel	257.764	270.133	284.831	291.964	293.877	268.555	266.199	267.265	0.056

Various metrics averages

tmp_averages = stats[stats["round_num"] == 2].copy()

tmp_averages = tmp_averages[tmp_averages["OSP_Damage_Given"]>0]
tmp_averages = tmp_averages[tmp_averages["OSP_Kills"]<50]
tmp_averages = tmp_averages[tmp_averages["Killer"].isin(tmp_games_filt.index.values)]
tmp_averages = tmp_averages.groupby(["Killer"]).agg({"OSP_Damage_Given": "sum", 
                                                     "OSP_Damage_Received": "sum", 
                                                    "round_guid" : "count",
                                                    "round_time": "sum",
                                                    "OSP_Kills": "sum",
                                                    "OSP_Deaths": "sum",
                                                    "OSP_Suicides": "sum",
                                                    "AdjScore": "mean"})
tmp_averages["KDR"] = tmp_averages["OSP_Kills"]/tmp_averages["OSP_Deaths"]
tmp_averages["DPM"] = tmp_averages["OSP_Damage_Given"]/(tmp_averages["round_time"]/60)
tmp_averages["DPK"] = tmp_averages["OSP_Damage_Given"]/tmp_averages["OSP_Kills"]
tmp_averages["DPD"] = tmp_averages["OSP_Damage_Received"]/tmp_averages["OSP_Deaths"]
tmp_averages["SPD%"] = tmp_averages["OSP_Suicides"]/tmp_averages["OSP_Deaths"]*100


pd.options.display.float_format = '{:.1f}'.format
tmp_averages

	OSP_Damage_Given	OSP_Damage_Received	round_guid	round_time	OSP_Kills	OSP_Deaths	OSP_Suicides	AdjScore	KDR	DPM	DPK	DPD	SPD%
Killer
blackmagik	1602745	1371946	583	183880	7368	8062	2062	10.5	0.9	523.0	217.5	170.2	25.6
brandon	1914414	1203885	459	140943	8559	6518	2411	8.0	1.3	815.0	223.7	184.7	37.0
bru	3779929	3259296	1215	381348	17991	19793	5179	11.2	0.9	594.7	210.1	164.7	26.2
caffeine	551797	424220	148	50782	2252	2840	939	9.2	0.8	652.0	245.0	149.4	33.1
cakel	2547027	4396417	1758	553025	13239	23506	3825	10.5	0.6	276.3	192.4	187.0	16.3
cliffdark	617497	613806	241	70903	2842	3825	741	10.5	0.7	522.5	217.3	160.5	19.4
conscious	1132064	1581260	583	182985	5683	7663	1157	11.5	0.7	371.2	199.2	206.3	15.1
corpse	757234	853390	328	98025	3784	4313	710	12.6	0.9	463.5	200.1	197.9	16.5
cypher	301086	415511	135	43801	1409	2228	348	12.9	0.6	412.4	213.7	186.5	15.6
deadeye	2104054	3874917	1503	481247	10418	21117	1552	9.6	0.5	262.3	202.0	183.5	7.3
detdet	628038	722741	251	76302	3304	3252	675	11.1	1.0	493.9	190.1	222.2	20.8
diego	481307	630117	265	77549	2118	3272	474	8.5	0.6	372.4	227.2	192.6	14.5
dillweed	988431	1994625	755	242510	4880	10731	940	12.0	0.5	244.6	202.5	185.9	8.8
ding	721303	893675	353	108367	3804	4543	751	13.1	0.8	399.4	189.6	196.7	16.5
donka	4369870	4668699	1722	539669	21843	24113	4272	12.2	0.9	485.8	200.1	193.6	17.7
donkey	3085498	2324775	999	311205	14791	13823	4421	10.1	1.1	594.9	208.6	168.2	32.0
doza	2450289	2463225	931	287217	13199	10991	3466	11.2	1.2	511.9	185.6	224.1	31.5
druwin	1654629	1106871	439	137152	7482	5849	2295	8.2	1.3	723.9	221.1	189.2	39.2
duster	195200	405149	139	44489	1012	1992	123	11.8	0.5	263.3	192.9	203.4	6.2
elsa	2240594	1828086	744	235681	9818	10541	3714	9.1	0.9	570.4	228.2	173.4	35.2
eternal	2830706	2801590	1079	335958	14422	15986	3073	11.8	0.9	505.5	196.3	175.3	19.2
festus	1193781	1505599	667	199267	5896	8814	2310	10.2	0.7	359.5	202.5	170.8	26.2
firehot	484254	1582748	587	184288	2615	7724	528	9.4	0.3	157.7	185.2	204.9	6.8
flogzero	1773105	1747287	617	196190	9088	8398	1840	14.9	1.1	542.3	195.1	208.1	21.9
fonze	3944732	4840515	1930	611738	20205	23955	3474	10.5	0.8	386.9	195.2	202.1	14.5
fro	5606057	3528873	1556	487256	25882	19018	8866	7.0	1.4	690.3	216.6	185.6	46.6
illkilla	1197349	1208940	569	173992	6066	7466	1450	11.7	0.8	412.9	197.4	161.9	19.4
ipod	3047761	2674142	961	290783	14651	12980	3419	10.4	1.1	628.9	208.0	206.0	26.3
jam	922673	718231	280	86297	4867	3415	1192	9.7	1.4	641.5	189.6	210.3	34.9
jaytee	837338	556692	227	77125	3419	3560	1026	7.2	1.0	651.4	244.9	156.4	28.8
jimmy	883437	710812	291	88957	3938	4294	1015	8.6	0.9	595.9	224.3	165.5	23.6
joep	1205800	775564	349	106377	5303	4863	2222	4.9	1.1	680.1	227.4	159.5	45.7
john_mullins	7668705	5440332	1987	620317	34592	31157	11919	7.8	1.1	741.8	221.7	174.6	38.3
kazz	396787	515294	213	66468	2173	2788	675	8.2	0.8	358.2	182.6	184.8	24.2
kep	165481	327887	131	41227	829	1724	130	9.9	0.5	240.8	199.6	190.2	7.5
kittens	5736464	4483278	1640	507628	27105	25214	7090	11.7	1.1	678.0	211.6	177.8	28.1
luna	396053	273448	129	39531	1955	1645	312	9.6	1.2	601.1	202.6	166.2	19.0
mean	566222	725457	262	83443	2695	3263	334	10.2	0.8	407.1	210.1	222.3	10.2
miles	2019706	2132441	835	258379	10176	11031	2201	9.3	0.9	469.0	198.5	193.3	20.0
mmbird	157465	301726	125	38091	810	1726	104	11.3	0.5	248.0	194.4	174.8	6.0
mooshu	3201933	3817601	1495	468700	16662	21235	4328	11.7	0.8	409.9	192.2	179.8	20.4
murkey	5162460	3652470	1394	436206	25358	18947	5701	13.8	1.3	710.1	203.6	192.8	30.1
nigel	3709980	3590852	1287	396576	18766	19263	4310	12.7	1.0	561.3	197.7	186.4	22.4
parcher	4539290	4486460	1557	489480	24202	20527	4045	12.4	1.2	556.4	187.6	218.6	19.7
pimp	390202	396227	176	57839	2045	2151	503	10.6	1.0	404.8	190.8	184.2	23.4
pingrage	382954	436595	185	57455	1921	2524	580	9.3	0.8	399.9	199.4	173.0	23.0
prowler	1083454	1658213	652	207762	5266	9087	1288	11.4	0.6	312.9	205.7	182.5	14.2
raiser	2616512	2022312	943	295281	11618	12049	4171	10.4	1.0	531.7	225.2	167.8	34.6
reker	1090325	1377576	576	181436	5389	7682	1412	13.2	0.7	360.6	202.3	179.3	18.4
risk	525165	358041	177	54173	2554	2050	1143	4.4	1.2	581.7	205.6	174.7	55.8
rob	1831373	1726792	759	235796	8867	9608	2750	8.0	0.9	466.0	206.5	179.7	28.6
robmaster	1510504	1715333	702	219581	7725	9064	1975	9.9	0.9	412.7	195.5	189.2	21.8
sem	408727	439692	148	46549	1959	1960	361	9.5	1.0	526.8	208.6	224.3	18.4
shaz	280546	307214	135	42500	1200	1645	259	6.6	0.7	396.1	233.8	186.8	15.7
silentstorm	1545721	1686973	763	231697	8139	9625	2608	10.3	0.8	400.3	189.9	175.3	27.1
souldriver	745345	1439832	544	170502	3854	7205	568	7.9	0.5	262.3	193.4	199.8	7.9
source	4435040	3148107	1361	424150	22609	17735	5893	11.3	1.3	627.4	196.2	177.5	33.2
spaztik	2912146	2933551	1293	409321	14344	17449	3368	9.7	0.8	426.9	203.0	168.1	19.3
tragic	499609	490437	191	60675	2665	2789	761	11.8	1.0	494.1	187.5	175.8	27.3
vacs	174223	343146	142	43269	918	1773	321	8.1	0.5	241.6	189.8	193.5	18.1
vein	403514	479507	170	50029	2080	2092	481	9.9	1.0	483.9	194.0	229.2	23.0
virkes	253606	318881	125	39166	1242	1632	276	11.4	0.8	388.5	204.2	195.4	16.9
vulki	199083	374750	149	46710	1056	1963	120	8.0	0.5	255.7	188.5	190.9	6.1
warrior	1348145	910865	423	133094	6591	5160	1635	12.6	1.3	607.8	204.5	176.5	31.7
xill	793760	1028430	346	110194	4279	4999	446	12.2	0.9	432.2	185.5	205.7	8.9
yeniceri	710238	1153541	476	147239	3816	5901	452	9.2	0.6	289.4	186.1	195.5	7.7
yyz	405931	471276	161	46150	2042	2252	548	9.5	0.9	527.8	198.8	209.3	24.3

KDR distribution

sns.set(rc={'figure.figsize':(12,1)})
sns.boxplot(data=tmp_averages[["KDR"]], orient="h", palette="Set2")

<AxesSubplot:>

Damange per Kill

sns.boxplot(data=tmp_averages[["DPK"]], orient="h", palette="Set2")

<AxesSubplot:>

Damage taken per death

sns.boxplot(data=tmp_averages[["DPD"]], orient="h", palette="Set2")

<AxesSubplot:>

Ratio of suicides to deaths

sns.boxplot(data=tmp_averages[["SPD%"]], orient="h", palette="Set2")

<AxesSubplot:>

Gameplay score = score - kills, TKs, suicides.

sns.boxplot(data=tmp_averages[["AdjScore"]], orient="h", palette="Set2")

<AxesSubplot:>