Addendum to Radiant Advantage

Ran some Radiant vs Dire numbers looking at the final kill differential. Basically I divide the games into Radiant wins and Dire wins and by games that end in less than 30 minutes (but longer than 15 to filter early abandons) and games that last longer than 30 minutes. The number represents the average number of kills the winning team is up by at the end of the game

Very High
under 30 minutes
Radiant — 19.13 || Dire — 11.76
over 30 minutes
Radiant — 14.44 || Dire — 13.92

High
under 30 minutes
Radiant — 19.41 || Dire — 11.22
over 30 minutes
Radiant — 15.30 || Dire — 14.27

Normal
under 30 minutes
Radiant — 17.89 || Dire — 12.02
over 30 minutes
Radiant — 15.38 || Dire — 15.02

Radiant appears to have a significant advantage in kills during the laning phase, but there’s still no clear answer as to why or whether it’s the kills themselves that are the source of the Radiant win advantage.

Re-created the test in two different ways. Both disagree markedly with the above results, and a bug was isolated. New results are:

Very High
under 30 minutes
Radiant — 19.13 || Dire — 18.50
over 30 minutes
Radiant — 14.44 || Dire — 15.28

Kill differential as a cause or symptom now appears to be a non-starter. Radiant win advantage in < 30 minute games remains.

Edit: Trying out average team creep score per minute

Very High
under 30 minutes
Radiant Win: Winner — 16.05 || Loser — 10.69
Dire Win: Winner — 15.71 || Loser — 10.97
over 30 minutes
Radiant Win: Winner — 16.94 || Loser — 13.80
Dire Win: Winner — 16.64 || Loser — 14.40

Minor shifts and nothing conclusive.

Advertisements

2 Responses to Addendum to Radiant Advantage

  1. shrddr says:

    so the cs numbers show that radiant have an advantage, be it or geographic availability of creeps or matchmaking bonus (stronger players)

    • phantasmal says:

      Maybe, but the numbers are close enough that I don’t feel comfortable saying anything definitive about it.

      One possibly interesting piece to the puzzle is looking at the differential

      Radiant Wins under 30 minutes: 16.05 – 10.97 = 5.08
      Dire Wins under 30 minutes: 15.71 – 10.69 = 5.02

      Radiant Wins over 30 minutes: 16.94 – 14.40 = 2.54
      Dire Wins over 30 minutes: 16.64 – 13.80 = 2.84

      In this light they look even closer, but it appears Radiant might have an advantage the higher the average CS in the game is. Whether that’s a cause or effect, or even noteworthy, is anyone’s guess, but it might be worth looking into in future samples.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: