Fast paced action games like DOTA 2 and League of Legends are often difficult to follow. Most people who watch them can spot trends and gain a relatively good idea of who will win a battle before it has even ended.
But there is actually a computer designed by teams from Germany, Denmark, and Sweden that review how players play and can determine who will win these games. The company eSports has released a model that reviews the behavior of each player in the game. The system will then break down the probability of each of these gamers being successful and based on the statistics, it will predict who the most likely winner will be in the game. Of course, this doesn’t always measure the exact winner.
Over the course of 400 games, the game was statistically accurate for the number of times players would win. However, real world events could still cause the game to turn out differently. An unexpected pause, someone talking to a player, and other experiences that could alter these results should be taken into account.
Through their research, players have the chance to better analyze their games and gain a deeper understanding of their tactics. It is with this information that they can begin to improve on their games and ultimately improve their areas of weakness to increase their probability of winning.
While the use of this system is limited to a couple of games now, the computer model is expected to roll out to cover additional games. This will allow players of fighting and similar combat strategy games to review what they are doing and move forward with a depth of information. Ultimately, professional gamers will have the chance to improve their skill and to help win at real world tournaments.