Why won't BGMI zone prediction algorithms work?

Getting to the play zone is a tricky job (Image via Krafton)
Getting to the play zone is a tricky job (Image via Krafton)

Battle royale games like BGMI strongly depend on the play zones to make the game chaotic and exciting at the same time.

Play zones are unpredictable and random. There have been multiple attempts by BGMI analysts, content creators and esports players to formulate an algorithm for zone prediction. Most of it is based solely on observations and experience.

This article explains why these algorithms won't help in predicting the zones.


Zone deciding mechanisms in BGMI

There are many zone determining algorithms used by millions of BGMI fans and players. The following are the most probable ones:

Random center zone

Humans can come up with random numbers quite easily, but computers aren't intelligent enough to do the same. They use an algorithm to find/generate random numbers. The trick lies in finding an initial number (also known as seed) and performing mathematical operations on them repeatedly to find random numbers.

For example: Let the initial number be 121. It is then multiplied by itself, i.e., 121*121 = 14,641. The digits at that center are 464. This gives the first random number. Then the above steps are repeated with 464, which gives 2,15,296. The middle digits are 529, making it the second random number. Thus, the random numbers entirely depend on the seed and the mathematical procedures. Computers generally use time, temperature, humidity and the like as the seed.

BGMI can also use a similar algorithm with complex mathematical operations and complicated seeds to generate the center of play zones. Without good knowledge of the seed and mathematical operations, none of the algorithms given by experts will work.


Cluster center approach

If the BGMI algorithm for zone shift is designed for fair play and to reduce randomness, it might be linked to the player distribution across the map. This algorithm will shift the zone center to the cluster center of the players. In simple terms, the zone will shift to a place with a higher number of players.

The math is as follows:

The algorithm can take the position of each player as the Cartesian coordinates. According to these Cartesian coordinates, the center point of all players can be found using a simple mathematical equation. It is similar to finding the center of a circle or a triangle. After finding the center, the new BGMI zone will be formed with a reduced radius.

The initial position of players (Image via Medium/Karan Gandhi)
The initial position of players (Image via Medium/Karan Gandhi)
Cluster center of players' positions depicted as a star (Image via Medium/Karan Gandhi)
Cluster center of players' positions depicted as a star (Image via Medium/Karan Gandhi)

Inverse cluster center algorithm

This is similar to the cluster center algorithm but with a twist. Instead of taking the player position in BGMI, this algorithm will take the position where players are absent and shift the zone center to the cluster center of those points. In simple terms, the zone will shift to a place with fewer players.


Combined algorithm

There is a probability of a combined algorithm where the BGMI zone shift will either be randomly decided or by using the cluster algorithm.


Machine learning in zone predictions

One can understand the true nature of zone shifts and their patterns by deploying a machine learning algorithm.

youtube-cover

Zone prediction poses a great challenge to esports players and analysts, especially BGMI players. If the underlying pattern is recognized, only then can one propose a crude algorithm for zone shift.

The author has an extensive background in mathematics and computer programming.


Disclaimer: The article reflects the views of the author.

Quick Links