How are we able to replicate real life scores in sports games like FIFA, Madden and 2K in something like one tenth of the time?
In: 4
10% isn’t a particularly high number. A coin flip for who wins would be right 50% of the time and that takes no data. The games, like Vegas odds, have a lot of data input to approximate the strengths of the teams. So we’re able to make a much better guess than just random odds. For who will win, that means more than 50%. But scores are a different beast.
With FIFA the most common score ranges are 4 goals or less per side. That means there’s less than 10 likely outcomes in scores. So again, random chance is going to get us 10% and we’re working with more data than that. With Madden there’s more variability to scores but it’s still very unlikely both teams score over 35 and uncommon that either team doesn’t manage to score. So once again you have a relatively small number of likely scores since safety and missed PAT are uncommon enough. Using more data to know which teams are likely to go for a 4th down vs kick a FG you can make educated guesses.
So, 10% isn’t a big number, the total odds of getting the right score aren’t that tough, and we’re working with a lot of data to make best guesses.
Latest Answers