Do the Predator style sets of 4 actually lead to more upsets than one long race? I took a look at the data from the 172 matches played so far

kingwang

AzB Silver Member
Silver Member
A lot of unsubstantiated claims have been getting thrown around regarding Predator's set format for the World 10-ball Championship, so I decided to take a look at the actual data and find out the truth. The first claim I wanted to tackle: the set format leads to a lot more upsets than a single race.

I painstakingly looked at all the matches played so far in this tournament. The official website did not make this task easy.

The results
  • 172 matches were played
  • In 3 matches (1.7%), the winner won one fewer game than the loser
  • In 169 matches, the winner won the same number of games or more than the loser
The three matches
  • Konrad Juszczyszyn (8) v Sun Yi Hsuan (9)
  • Konrad Juszczyszyn (8) v Edwin Gamas (9)
  • Duong Quoc Hoang (16) v Aloysius Yapp (15)
My take on this? I think the set format is nice. The matches stay exciting, neither player is never really "out of it", and yet the results are still 98%+ the same as having one long race.
 
... My take on this? I think the set format is nice. The matches stay exciting, neither player is never really "out of it", and yet the results are still 98%+ the same as having one long race.
It is easy enough to calculate the chance of an upset for each format. Mike Page gave an example of this. For any particular set race, there are single race lengths that will be better and others that will be worse at deciding the better player.

It turns out that if you set a maximum number of games for the match, a single race gives more advantage to the better player. That's another way to say "fewer upsets".

One disadvantage of the set format is that the length has more variation. Race to three sets of races to four could end in 12 games (4-0 three time) or it could take 35 (4-3 five times) counting a shoot out as a game, if that is used. That's a problem for scheduling.

Carom billiards changed to sets from single games. In that case it was race to three with games to 15 points. The championships had been single games to 60 points. They have since changed back to single games with the chance of a shootout as the non-breaker is allowed a chance to tie in "regulation" in some matches.
 
It is easy enough to calculate the chance of an upset for each format. Mike Page gave an example of this. For any particular set race, there are single race lengths that will be better and others that will be worse at deciding the better player.

Yes, but both you and I know that there are too many people on this forum that lose their minds the second they see the words "calculate" or "probability". I figured gathering some real world data would be helpful.
 
A lot of unsubstantiated claims have been getting thrown around regarding Predator's set format for the World 10-ball Championship, so I decided to take a look at the actual data and find out the truth. The first claim I wanted to tackle: the set format leads to a lot more upsets than a single race.

I painstakingly looked at all the matches played so far in this tournament. The official website did not make this task easy.

The results
  • 172 matches were played
  • In 3 matches (1.7%), the winner won one fewer game than the loser
  • In 169 matches, the winner won the same number of games or more than the loser
The three matches
  • Konrad Juszczyszyn (8) v Sun Yi Hsuan (9)
  • Konrad Juszczyszyn (8) v Edwin Gamas (9)
  • Duong Quoc Hoang (16) v Aloysius Yapp (15)
My take on this? I think the set format is nice. The matches stay exciting, neither player is never really "out of it", and yet the results are still 98%+ the same as having one long race.
I think this is way too simplified. The fact of having won more games or not and won the match, isn’t the reason why people think there are upsets.
It’s more the point that , specifically in best of 3 early rounds, the weaker player has a stronger chance to make an upset than in a long race.
For example in a long race it could quite possibly be 4-2 because of winning the lag, breaking first, getting a lucky roll or the opponent missing a final ball. However the stronger player will then usually come back and get to the finish like first. Not the case in this format is you would now be one set down and your back is against the wall.
 
Back
Top