Flexner Stat Corner: A look back at GSTL 2011 Season 1
With the first season of the 2012 GSTL just getting underway, we thought it'd be a good time to look back on how the previous GSTL season played out - in numbers. ESFI World's Reece Flexner now introduces a new recurring feature, Flexner Stat Corner, which will delve into statistical StarCraft 2 analysis. Have ideas for future editions? Send them our way!
| Momentum |
|---|
Out of the 116 kill streaks, 64 (55%) of them ended at one game. On 55 of those 64 occasions, the player won one game and then lost the next. There were 22 two-kills, 16 three-kills, and a whooping 13 four-kills. Predictably, teams were much more likely to end a match on a long streak of wins as well. Out of 35 matches, 21 (60%) finished with a winning streak of three or more, while only 9 (26%) were ended by a player coming in for a single win.
Similarly, when players won, they tended to keep winning. After a win, players went on to win the next game 54% of the time; after winning two games, they won a third 63% of the time; and if they managed three in a row, they finished out with four games, and a match win, 62% of time.
| Race Representation |
|---|
As usual, Terrans dominated the first full season of the GSTL, being sent out 43% of the time as compared to 28% a piece for Protoss and Zerg players. SlayerS, of course, was the biggest offender here, with Terrans fighting on their side in 25 of 29 (86%) of games. However, Terrans were not much more successful than the other two races.
The win ratings for each match-up were fairly even, and Terrans accounted for just 44% of the tournament's player-games, only a bit above the percentage of time they were sent out (to clarify: “player-games” is the measure of the number of games played by the players of each race. For instance, a TvT would count as two player-games for Terran, as would a Terran beating a Protoss and then a Zerg.) Terran accounts for 14 of the 29 (48%) three-or-more-kill-streaks. Zerg scored 9 (31%) of these while Protoss lagged behind at 6 (21%), with three of those coming in the last few matches of the season.
| Strategy |
|---|
There is one common strategy that seems very strange in the GSTL, that of sending out a player of the same race as the one who was just beaten. The numbers bear out the obvious silliness of this tactic. As mentioned above, after winning any one game, a player tends to win the next 54% of the time. However, if the opposing team sends out the same race as was just defeated, that rate jumps to 69% of the time. That means that the likelihood of Zerg #2 triumphing where Zerg #1 failed is just 31%. (Note: these numbers only took into account immediately consecutive games. Perhaps I will look at sets of three games in the future.)
Reece Flexner
