I had a request to examine the gamble in twitch chat and so I wanted to do a short article about it!
With round two of the 2023 Root Winter Tournament wrapping up, I once again wanted to look at the twitch stat. How good is twitch chat a bettering on the winner? So, let’s talk about the gamble. What is it? Why does it happen? How easy is it to predict the winner? And how good is chat at doing so?
Let’s start with the basics. What is the gamble? Typically, we have a gamble every game where viewers can bet their GuerricBux on the faction they believe will win. The gamble happens at the end of a round after a player has reached 15 points.
Round two saw the game average go down to 7.8 rounds per game. The end of round 5 was still a consistent time for the bet to occur.
We still saw some bets happening in round 6 and there were more times that the bet happened at the end of round 4.
The two times that the bet happened during round 3 was when someone declared dominance. These are edge cases but the bet would happen immediately since the landscape of the game was just drastically changed.
Now we got over the what, let’s get onto the why. Whenever you watch the GuerricSamplesGame stream on twitch, you will earn GuerricBux just for watching. You can always use them for highlighting messages or modifying emotes, but you can also use them for gambling on your favored faction!
What will this get you? Potentially more GuerricBux! It’s a never-ending cycle. But since you can bet any number of GuerricBux on a faction, having more can mean showing more support or making a riskier play.
But can statistics tell us which faction will be good to bet on? Short answer, no. I looked at two different variables. One – in what position was the faction at the time of the bet. Were they in first (winning) or in fourth (losing). Two – Was there a correlation in their scores at the time of the bet.
In terms of position, there was a surprising spread. 46% of winners were winning at the time of the bet but 31% of them were in the second position, 4% were in the third position, and 19% were in the fourth position. No one was in the fifth position at the time of the bet if a five player game was happening.
If points were tied at the time of the bet, the tie breaker was determined by seat order. While the first position has the highest percentage, I don’t think it shows much of an advantage.
And then we get to the scores, and they are all over the place! Some winners were only at 6 points at the time of the bet while others were at 21 points! It’s truly an interesting scattering of a range. Being at 16 or 17 points seems to have the highest chance of winning in the range of scores.
The last stat I wanted to look at is the percent ranges for the coins bet and the people bet. To accurately compare one game to another, I had to standardize the units. 14 people can bet in one game and 4 people can bet in another but those number might make up the same percentage of the vote depending on the game. That is why I standardized by percent.
I found that people were more confident with their money they were in uniting with each other. I found it interesting how many people would bet so many coins and so few coins (I see those hedge bets). Yet, the coins didn’t reflect the number of people betting. A few people with a lot of coins would heavily tip the scales. Or even a lot of people with only a few coins.
Overall people were less confident in their bets (either not as many coins or not as many people) in round two than they were in round one.
The tournament round three is coming up so keep watching and keep betting!