Loaded Brackets
02.23.06
Ken,
I just reviewed Joe Lunardi’s 2/20/06 bracket projection on ESPN.com. In general those projections usually look like a fair comparison to how the committee would seed the teams at that particular point in time. I think that’s again the case with his 2/20/06 bracket. However, I have to laugh when I see that Duke is seeded #1 in the Atlanta bracket. While that’s great for Duke in terms of geography, I’m not so sure they wouldn’t rather be out West. The Atlanta bracket includes 7 of the top 13 pythag winning% teams (Duke, Kansas, Ohio St, Pitt, North Carolina, Washington, and LSU). I’m not a Duke fan, but I’ve got think Coach K would rather swap brackets with Memphis who’s slated to be in the Oakland bracket.
It seems like a way around this would be to either (1) have you slot the tournament teams, or (2) have the committee rank the teams 1 thru 65, and each team selects their slot (highest ranked team picking first). On the other hand, if you’re a Memphis fan you probably like where things are headed. Bring on Gonzaga on more time.
Kevin
Wow. That would be the most loaded region of all time. But this is an example of why you shouldn’t get too invested in season-ended-now projections. Last night, Ohio State, UNC, and LSU all got big road wins, and if Joe did his projection today, I am sure those wins would be worth a bump up in seed and a realignment of the brackets.
One of the recurring themes of this season has been how traditional powers may not make the tournament. One of the themes developing in my own crazy head is how traditional powers Kansas and UNC will be the most under-seeded teams in the dance. The common perception among fans is that teams like that will always get the benefit of the doubt due to TV concerns or other conspiracies.
It’s possible Kansas and UNC could still play their way into their deserved seed. I doubt either will win out, but if they did, they would be looking at 2-seeds even despite the early hiccups. More than likely they lose one or two, and stay as an extremely dangerous 4 or 5. How would a KU/UNC sweet sixteen game sound? [Note: Duh, this would be a second round game] Come to think of it, that’s exactly what the folks at CBS Sports would want.
In an unrelated note, I was disturbed by this tidbit from a recent piece by Andy Katz:
Here’s something we didn’t know: Apparently the committee also will look at the number of points the missing player averages and look how many points the team lost a game by to see whether that player’s production could have made up for the loss.
In my best John McEnroe: You can’t be serious!!
Let’s list a couple of big fallacies here. For one thing, there are 15 ppg scorers that are offensive drains on their team, and there are 5 ppg scorers who contribute much more than 5 points to the offense. The other thing I love is how the defensive value of a player is totally ignored by this idea. I like the old method of just evaluating how a team plays with and without the injured player. The committee needs to stay far away from any statistical analysis.
Line o’ the Night
FG 3pt FT Reb
Min M-A M-A M-A O-T A F S TO BLK Pts
John Bowler 38 6-10 0-1 17-24 6-15 0 4 0 1 0 29
Result: Loss. Toledo 70, Eastern Michigan 59.
Scorekeeper’s Bias
02.16.06
Ken- Love your website. I visit it regularly, even though I’m on the women’s side of college basketball. That said, I saw an amazing box score last night. Florida’s women’s team scored 63 points, won by 15… yet only had 3 assists. I thought that’d make an interesting scavenger hunt for your readers… what is the highest point total a team has ever scored with 3 or fewer assists? I can’t imagine it topping the 63 by the Lady Gators. Regards, Andy Demetra Broadcaster University of South Carolina women’s basketball
This was a seemingly innocuous question, but it turned up an interesting find. The two worst assist outputs of the season occurred in the SWAC, and they both occurred at Jackson State. But it’s not a coincidence.
Counting stats at a basketball game is to some degree subjective. Arguably, the most subjective element is how to record assists. There has always been some notion that certain scorekeepers are less charitable than others about doling out assists. However, I have never seen the bias that exists at Jackson State.
Let’s just say if you are a member of the visiting team playing JSU, you won’t get your assists properly counted. Here is how SWAC teams have fared against the Jaguars in home and away games. Figures for each game are assists/field goals made.
Home Away Alabama A&M 3/25 Alabama St. 5/26 Alcorn St. 12/21 4/22 Southern 17/32 4/28 Prairie View 11/29 2/29 Texas Southern 18/28 8/22 Grambling 2/28 MVSU 14/20 Ark-Pine Bluff 15/19 Total A/FGM .584 .156
It’s obvious that the Jackson State scorer’s table doesn’t have any interest in recording opponents’ assists. They don’t do a very good job with the home team, but at least they try. Jackson State has a .328 A/FGM ratio in conference home games compared to .510 on the road.
To answer Andy’s question, both Prairie View and Grambling cashed in for 77 points with their two assists.
Line o’ the Night
When I dreamt up the Line o’ the Night feature, I expected to have multiple intriguing lines to choose from on most nights. That hasn’t usually been the case, though. But wow, Wednesday night was the most interesting of the season from a statistical standpoint.
FG 3pt FT Reb
Min M-A M-A M-A O-T A F S TO BLK Pts
Tyler Hansbrough 32 13-17 0-0 14-19 6-10 1 4 1 1 0 40
Result: Win. North Carolina 82, Georgia Tech 75.
Hansbrough gets the official LotN. This game brings up two thoughts. First, if Hansbrough comes back for his sophomore season, he’s going to rack up some tremendously efficient offensive numbers. He’s already one of the best high usage players in terms of free throw rate. Second, nobody is going to predict that UNC will successfully defend their title, but they are going to be a tougher out in the postseason than people think. It would be an incredible accomplishment for them to reach the Final Four, but that wouldn’t be an unreasonable goal, either.
A couple other lines are worth publication.
FG 3pt FT Reb
Min M-A M-A M-A O-T A F S TO BLK Pts
Paul Millsap 31 4-12 0-0 3-6 11-28 3 4 1 4 3 11
Result: Win. Louisiana Tech 74, San Jose St. 63.
FG 3pt FT Reb
Min M-A M-A M-A O-T A F S TO BLK Pts
Shawn James 31 7-18 1-4 2-2 6-14 1 2 0 4 11 17
Result: Win. Northeastern 77, James Madison 55.
But that’s not all. Did you see the night LeBron had? No, not that LeBron. I’m talking about Hofstra’s Jorge Lebron, and his five fouls in one official minute of action against UNC Wilmington. The NCAA lists the record for quickest to foul out at 1:11. King Maims used up his five fouls in a 46 second span, but it’s unclear exactly how long he was on the floor.
All of this makes the 12 turnovers committed by UMass point guard Chris Lowe less than noteworthy.
Finally, on a team note, Utah State busted out a 37-0 run over a 13+ minute span against Idaho. Idaho vandalized their home rims with 24 consecutive bricks during the drought.
E-Mail Thursday
02.09.06
Maybe I’ll make this a regular Thursday feature. It’s an effective way to maintain my laziness and continue to churn out content.
First up, we have someone who can navigate the NCAA web site better than me:
Ken,
It’s been three years since the last 50-point game. Here are all the D-I 50-point games from the last four years (from the NCAA web site):
54 Michael Watson, UMKC vs. Oral Roberts, 02/22/2003
53 Antawn Dobie, Long Island vs. St. Francis (N.Y.), 02/22/2003
52 Ron Williamson, Howard vs. N.C. A&T, 01/21/2003
50 Desmond Cambridge, Alabama A&M vs. Texas Southern, 02/25/2002Chuck
Thanks, Chuck! It’s a little creepy that the last 50 point games occurred on the same day.
Ken,
Earlier you posted your preference for the Pythagorean winning pct. over your ratings. Now that the season is a bit further along and the ratings seem to have stabilized a bit, how do you feel?
A couple of notable disparities as of today (2/08):
Team Pom Rating Pythag Rating ******************************************* Memphis 4 11 George Wash. 14 48 Arizona 41 24Dave
I still think the Pomeroy ratings are done after this season. Sure Pythagoras craps the bed every now and then, but have you seen the teams rising to the top of the power ratings this season? It hasn’t been pretty, and the latest victim was Florida, who was #2 before last night’s loss to South Carolina. Even before last night, I don’t think too many people would have bought that Florida was better than UConn, Villanova, or Texas.
Next season, I will make an adjustment to the efficiency ratings to calculate winning percentage based on something similar to Dean Oliver’s bell curve concept rather than the Pythagorean equation. The benefit of the bell curve is to penalize good teams that are inconsistent and reward good teams that are consistent. This would knock Kansas down a few pegs and bring Gonzaga up a little. While those two - rightly, I believe - look very underrated/overrated compared to national perception, the values coming out from Pythagoras are a little extreme.
Line o’ the Night
FG 3pt FT Reb
Min M-A M-A M-A O-T A F S TO BLK Pts
Jamon Gordon 42 9-15 0-1 5-6 3-16 4 3 4 3 1 21
Result: Win. Virginia Tech 75, Clemson 74. (OT)
Hey, the guy is only 6’-3”.