teamplay value for emea t1

Potentially useless stat:

(Σ(teams assists)/ Σ(teams kills) )* 100 = %kills assisted

drawbacks:

• assists can stack, multiple assists for 1 kill, resulting in an inflated percentage.
• agents like skye and sage can inflate numbers due to heals.
• assists can also involve trades
• teamplay cant accurately be measured by assist values

1. Liquid: (68+61+48+38+25) / (60+68+95+86+130) = 54.6697%

2. Fnc: (54+51+51+18+18) / (62+80+82+79+92) = 48.6076%

3. FUT : (49+47+23+23+21+9) / (71+82+53+56+71+23) = 48.3146%

4. KOI: (47 +44 +34 +30 +19 ) / (84 +42 +99 +82 +83 ) = 44.6154%

5. VIT: (46+40+35+25+14) / (68+50+78+65+103) = 43.9560%

6. Navi (65+39+33+31+24)/(91+93+112+84+94) = 40.5063%

7. GIANTS: (36+29+29+29+26) / (93+79+91+79+57) = 37.3433%

8. KC: (33+28+23+18+12) / (60+52+76+57+72) = 35.9621%

9. TH: (29+21+19+10+8) / (49+50+42+34+67) = 35.9504%

10. BBL: (32+28+27+25+25) / (73+75+75+101+89) = 33.1719%

reflection:
liquid are so high mainly due to soulcas, very high assist numbers on kayo and sage.
small sample size and healing effects hurt the data.
but the general trend is the better the team is the higher the value

i aint readin allat

W analsis

This is cool

I read allat but you put too much weight on assist, it's pretty easy if you're a controller to get an insane amount of assist

can we please sign this guy to a franchised team

Liquid being 1st is quite accurate. Sayf got insane stats because of perfect support from teammates. Gotta give big W for emil

potentially useless? nah this is straight useless

1. GENG: 94 / 159 = 59.11%
2. RRQ: 76 / 136 = 55.88%
3. DFM: 82 / 156 = 52.56%
4. GES: 96 / 183 = 52.45%
5. DRX: 69 / 137 = 50.36%
6. TLN: 89 / 188 = 47.34%
7. TS: 98 / 210 = 46.66%
8. T1: 80 / 212 = 37.73%
9. ZETA: 40 / 107 = 37.38%
10. PRX: 52 / 153 = 33.98%

It may be due to the low sample size, but the Pacific region kind of ruins the trend. Just from the eye test alone, teams like RRQ/DFM/GES don't have very good teamplay (RRQ and GES both are known for communication issues) but place highly with this formula.

DRX is by far the best team when it comes to teamplay in the Pacific but are in the middle. Maybe the data will improve with a larger sample size like you said and it's still cool info.

Sprouts [#9]

1. GENG: 94 / 159 = 59.11%
2. RRQ: 76 / 136 = 55.88%
3. DFM: 82 / 156 = 52.56%
4. GES: 96 / 183 = 52.45%
5. DRX: 69 / 137 = 50.36%
6. TLN: 89 / 188 = 47.34%
7. TS: 98 / 210 = 46.66%
8. T1: 80 / 212 = 37.73%
9. ZETA: 40 / 107 = 37.38%
10. PRX: 52 / 153 = 33.98%

It may be due to the low sample size, but the Pacific region kind of ruins the trend. Just from the eye test alone, teams like RRQ/DFM/GES don't have very good teamplay (RRQ and GES both are known for communication issues) but place highly with this formula.

DRX is by far the best team when it comes to teamplay in the Pacific but are in the middle. Maybe the data will improve with a larger sample size like you said and it's still cool info.

I almost had a stroke when I saw RRQ is 2nd

Sprouts [#9]

1. GENG: 94 / 159 = 59.11%
2. RRQ: 76 / 136 = 55.88%
3. DFM: 82 / 156 = 52.56%
4. GES: 96 / 183 = 52.45%
5. DRX: 69 / 137 = 50.36%
6. TLN: 89 / 188 = 47.34%
7. TS: 98 / 210 = 46.66%
8. T1: 80 / 212 = 37.73%
9. ZETA: 40 / 107 = 37.38%
10. PRX: 52 / 153 = 33.98%

It may be due to the low sample size, but the Pacific region kind of ruins the trend. Just from the eye test alone, teams like RRQ/DFM/GES don't have very good teamplay (RRQ and GES both are known for communication issues) but place highly with this formula.

DRX is by far the best team when it comes to teamplay in the Pacific but are in the middle. Maybe the data will improve with a larger sample size like you said and it's still cool info.

yeah. its probably pretty comp dependent.
also drx probably stomping zeta didnt help.
id expect more buy rounds = more abilties which in turn allow more assist.

could be something to go back to after the split ends.

Common capital_d_colon interesting thread

hanafuuji [#5]

I read allat but you put too much weight on assist, it's pretty easy if you're a controller to get an insane amount of assist

yeah i do agree. i did realise but i thought i may aswell finish it
its pretty hard to gage some information of teams.
stats like trade%, true first deaths (first deaths that arent traded)
maybe also being able to differentiate between assists (flash assists vs dmg assists)