Reply to thread

Message: [QUOTE="caw, post: 4609279, member: 563"] I disagree with this assessment of the situation. Sample size is small regardless of pace of a singular basketball game when talking shooting percentages and shot attempts. Correct me if I am wrong but your talking about shooting percentages, no? Then you are statistically speaking correct about percentages but that does not change what the poster said, necessarily. Since we are talking pace and shooting percentages and shot attempts. A slower paced, poor offensive team may have more variability to percentages game to game throughout the year due to smaller sample size during a game but that is only variability from their average. If their average shooting is 34% from three and a team only takes 10 threes a game (making 3.4 per game on average) then there is a wide variability statistically when looking at the percentages by making one, or two, extra shots. That is true. They go 5 of 10 instead of 3 of 10 and they are shooting twenty percent better against your team. Looks bad but it's only a six point difference. However, a good offensive team who has a high pace and shoots 40% on the year from three has a lower variability if they are putting up 25 threes a game and making on average 10. Each three attempt is worth a lower percentage to the shooting percentage, especially if you look at the season average as their number of shots will stack much higher over the year. All that is true, but you have a flaw in your assessment of the situation. A better offense with more shots per game (lower variable to percentage) team is actually more dangerous here. If they shoot 30% from three they only convert 8 threes (making it a whole number for realistic in game shooting), if they shoot 50% from three they convert 13 threes (rounding up as with the 30% for a realistic made number). Same 20% difference but that is now a fifteen point difference instead of a mere six points. So there is actually more variability to the actual score with the higher paced better offensive team even if the percentages have less variability. Now the counter would be that team B has less of a likelihood to have a 30% game or a 50% game but that is not really true as 25 shots in a single game is a statistically small sample size those percentages should fall within a 95% confidence interval to the average for the population (didn't do the math out fully but quick mental calculation says it's fine or off by a percentage point or two, which doesn't make a difference in the overall point). [/QUOTE]

Verification: First name of men's bb coach