1.8 Points Per Inning and Still a Loss to 1.32: What the 3-Cushion Scoreboard Doesn't Tell the Bình Dương Crowd

In this season's national 3-cushion billiards championship semifinal, one player closed out the match with an average of 1.8 points per inning. His opponent finished with 1.32. The player lifting the trophy was the one with the lower average. There was no refereeing error, no inning wrongly called. The first player simply had 42 innings, while the second had only 18. The second player's average was lifted by three consecutive 7-plus innings in the decisive phase - but it took me three days sitting with the score sheets to see that. I am the one who stays with the numbers when the arena lights go out - you may call it a job, I call it a calling. In 3-cushion, the scoring average is the most familiar measure. The Union Mondiale de Billard (UMB) calculates it by dividing total points by innings played. A continental-class player typically holds between 1.2 and 1.5. Top players such as Dick Jaspers and Torbjörn Blomdahl can touch 1.8 to 2.0 in a small tournament and rarely sustain that level across an entire season. The problem is this: an average is just a fraction. The numerator is total points, the denominator is innings. The same 1.8 can come from two completely different portraits. Player A scores 1.8 over 50 innings with three long runs of 9-10 points. Player B also scores 1.8 over 20 innings with steady 4-5 point pickups. A viewer who only looks at the electronic board will think the two are alike. That is the first mistake. Before trusting a number, I ask where it was born. Just as one watches a player before passing judgment. In Vietnam, 3-cushion has no expensive software like football. National tournaments still use handwritten sheets, later transferred into spreadsheet files. But precisely because it is rudimentary, I get the chance to look at every inning, not just the aggregate figure. This season, I logged all 214 innings of the eight players who reached the semifinals and final. For the first three days I merely counted. On the fourth day I understood what I was counting. My dataset splits every inning into five groups: innings scoring 0-1 point (the blank innings), 2-4 points, 5-7 points, 8-10 points, and above 10 points. I then compared the share of each group across players. The result reversed my initial expectation. The champion had a blank-inning rate as high as 41%. The runner-up only 28%. On the surface, the runner-up looked more efficient. But when I separated blank innings into two types - blanks caused by a miss and blanks caused by the opponent locking the table - the story flipped. The champion had 33 of his 41 blank innings caused by table locks. He came to the table after the opponent had just scored, with the balls parked in an extremely difficult position. In those innings he accepted 0-1 point, used the cue to push the balls into safe zones, and waited. The runner-up had 9 of his 28 blank innings from his own misses, while 19 came from table locks where he still tried to score and failed. This is the key point the electronic board does not show: a blank inning can be passivity, or it can be a strategic move. In 3-cushion, viewers usually equate a blank inning with a mistake. But a master player can turn a blank inning into a weapon - he accepts losing points first to preserve the table position for the next inning. I added another metric called next-inning efficiency. After a blank inning caused by a table lock, the champion averaged 6.2 points on the immediately following inning. The runner-up averaged only 3.1. The difference lies not in the blank innings, but in how those blank innings were used. Another metric I tracked was the average run length in the final five innings. The champion averaged 5.4 points across the last five innings. The runner-up only 2.8. This is the data zone analysts call the clutch zone - where psychological pressure begins to eat into technique. In 3-cushion, the last five innings often account for 30-40% of a match's total points if the game goes to the deciding inning. A player can lead the whole match but lose the last five innings, and the electronic board will only display the final total that does not reflect it. Broadly, my 214-inning dataset reveals a pattern: players who advance deep into the tournament tend to have higher blank-inning rates than those who stop at the quarterfinals. The average blank-inning rate of the semifinal-and-above group is 37%, versus 26% for the quarterfinal-and-below group. This paradox resolves when we understand that strong players usually face strong defensive opponents, leading to more locked-table innings, while weaker players usually face opponents of similar level and enjoy more open innings. This is when I remember the lesson of World Cup 2026. That year I analyzed 64 matches and found Croatia was not the side with the highest xG, but was the team that covered the most distance at 112 km per match on average. The 2026 World Cup taught me the most important thing: numbers are only beautiful when they know which side of the story to stand on. The same holds in 3-cushion. The 1.8 is not beautiful if we don't know whether it came from 42 innings or 18. The 1.32 is not ugly if we understand it was scored in innings where the opponent was locking the table. What the 3-cushion scoreboard does not say is that the correlation between scoring average and tournament ranking is not as strong as many believe. When I plotted a scatter chart between scoring average and final ranking for the eight semifinalists this season, the correlation coefficient was only 0.41. Statistically, this is a weak correlation. There are two reasons. First, the innings denominator is shaped by the opponent and how they set the balls. A player facing a defensive opponent will have fewer scoring opportunities, but when opportunities come they tend to be easy positions. A player facing an attacking opponent will have more innings but more of them in difficult positions. The average, therefore, cannot measure inning quality; it only measures the final outcome. Second, the 3-cushion average suffers from large-inning magnification. One inning scoring 10-12 points can lift the entire match average by 0.1-0.2. With a low inning count (under 25), one big inning can dominate the whole metric. This is a familiar statistical phenomenon that in football people call small-sample bias. I once saw a player average 1.65 over six innings thanks to a single 9-point inning, then get hyped by the media as the tournament's sensation. Two days later, he met a tough opponent and averaged 0.9 over 22 innings. The first number was beautiful because of a small denominator, not because the player had changed class. Truth be told, in 3-cushion as in any sport with metrics, causality does not lie in the final number. The final number is the result of hundreds of small decisions the board does not record: choosing the position, choosing the speed, choosing the pre-set point. To understand a player, one must watch how he makes decisions in the innings where he does not score. People see an inning scoring 12 points, I see the twenty tense minutes in the blank innings before it that led to it. I do not deny the value of the scoring average. It remains a useful metric when comparing a player with himself across tournaments. But when used to rank two players within the same match, we are mistaking correlation for causation. The coming 3-cushion season will feature more international tournaments in which Vietnamese players will take part. I suggest organizers and analysts consider adding three new metrics alongside the scoring average: the rate of blank innings caused by table locks, the average points across the final five innings, and the average innings per match. These three metrics tell the story the plain average leaves out. They show how a player can win, why he loses, and where he stands in his career journey. I will keep sitting with the numbers when the arena lights go out. But this season, I want to invite a few coaches and players to sit with me. Because in the end, data does not belong to the analyst. It belongs to those holding the cue - the ones who need a correct number to believe in themselves.