Some Possession-based Stats From the U18 Tournament

Some possession-based stat from the U18 tournament in Formosa, Argentina. Three games across 2 separated brackets may wreak havoc on notions of Baysean connecting, but the contrast between the top and bottom teams is pretty stark. Going in there was a consensus the USA team would do well (the Americans tend to dominate tournaments of younger players, but fall into the middle of the pack when playing tournaments with older players...), but the separation between the teams at the top of each bracket (Argentina and USA) is very clear. As the possession-based offensive stats below indicate...

		Four Factors
Team	Pace	eFG%	TO %	OReb %	FTRate	ORtg	PPWS
USA (B)	86.7	50.8	20.4	47.2	30.8	116.5	1.10
Argentina (A)	81.0	55.4	23.0	39.8	36.1	110.3	1.13
Puerto Rico (B)	89.5	42.0	16.8	33.1	37.1	95.0	0.91
Canada (A)	78.6	44.0	25.0	37.8	36.6	92.4	0.97
Venezuela (B)	92.4	47.8	29.6	32.5	61.5	84.5	1.01
Mexico (A)	81.4	40.0	17.2	15.9	17.6	75.8	0.83
Uruguay (A)	78.3	34.6	24.7	31.3	53.1	72.8	0.78
Bahamas (B)	87.2	34.2	25.2	32.7	23.6	68.8	0.72

The Americans and Argentines have offensive ratings that are at least 13% better than the next tier. Note they dominated their respective brackets in the crucial offensive categories, including shooting (eFG%), rebounding (OR% -- offensive rebound rate) and scoring (PPWS). Those two team used their possessions to score far more efficiently (the ORtgs) than their bracketmates. The semi-final games (whose stats are not included in these tables), which matched Argentina against Venezuela and the USA with Canada were the virtual walk-overs, these tables reflect. The first four in the table were matched in the last round, an interesting predicted outcome. I included Pace (average possessions per game, raw) because frankly, the stat surprised me when I collected data on the first few games. The pace resembles AAU or the NBA far more than the typical D1 basketball played in the NCAA these days. The average for D1 last season (as computed by Ken Pomeroy) was 67.0. The "most deliberate" team in the tournament, Uruguay (78.3) had a pace very close to Texas State (see link), the #1 ranked team for possessions (79.8) in D1 last season. The highest paced team, Venezuela (92.4), works at a pace similar to an NBA team (though considerably less efficient...).

The defensive numbers lay out in 4 very neat tiers -- note the DRtg column (second to last on the right)...

	Four Factors
Team	eFG%	TO%	OR%	FTRate	DRtg	PPWS
USA (B)	34.1	23.8	28.3	38.7	72.3	0.78
Argentina (A)	39.5	22.2	21.4	31.9	76.1	0.87
Canada (A)	40.4	20.8	26.7	30.3	82.7	0.86
Venezuela (B)	41.5	21.3	31.2	27.5	86.6	0.88
Uruguay (A)	46.5	21.7	28.7	35.8	90.7	0.97
Puerto Rico (B)	46.6	25.3	39.1	30.6	92.0	0.97
Mexico (A)	48.5	25.0	45.4	42.4	102.0	1.02
Bahamas (B)	51.9	21.8	49.2	52.1	113.6	1.11

The teams could have been ordered by eFG% or PPWS and would have ordered the teams almost exactly as they are by DRtg -- shot defense matters pilgrims. While defense is not an element of the game typically mastered (or even practiced...) by AAU teams, the USA and Argentine teams clearly limited scoring by their opponents, ie -- they played defense. This was suggested by player post practice interviews during the week after Independence Day. The order, no matter how clearly delineated, did not hold in medal round play.

A few notes...
1. The tables include data only from the bracket play, conducted Monday through Wednesday..
2. I used a modified version of Dean Oliver's possession calculation to estimate the possessions in each game. Oliver's original calculation, FGAs + TOs - ORs + (.475 * FTAs) modified over the years by Pomeroy who substituted Olivers's .4 (used in the NBA for .475 for D1, reflecting a slightly increased frequency with which a team's possession ends on the free throw line). I used a modified version, FGAs + (.45 * FTAs) - 1.07 * (OR%/(OR% + opponent's OR%)) - (FGA - FGM) + TO. The calculation, vetted some over at , provides a slightly more "accurate estimate" of the number of second chance FGAs. I have used both during the 2007-08 season, the modified calculation tends to raise the possesion rate for most teams. I have used .45 as the constant to estimate the rate by which FTAs are possession-ending events for international games. I confess I am arbitrary here, looking for a constant that reflects the general nature of the international game (somewhere between the D1 game and the NBA game...). The international game however, offers a variety of styles and strategies that is even wider than that found in the D1 game.
3. I have noted the assigned bracket (A or B) in parenthesis to the right of each team.

I will most likely post some individual player stats when the tournament is over.

Oh, and Canada just beat Puerto Rico, 83-68, to take the bronze medal.