Thursday, March 27, 2008

#6 Illinois vs. #27 South Carolina

#6 Illinois
G Shaun Livingston G Dwayne Wade F Corey Maggette F Shawn Marion C Eddy Curry

#27 South Carolina
G Ray Felton G Anthony Johnson F Jermaine O'Neal F Kevin Garnett C Kwame Brown

Illinois has a team that is perhaps a bit disappointing for the #6 seed. Livingston has a degenerative knee condition that may never allow him to play again. Before the injury, he was a promising big point guard who made too many turnovers and didn't shoot well enough. It's scary to think so, but Wade might be on the decline already: he can't seem to stay healthy and his game is predicated on being able to get into the lane and taking contact. Maggette is another wing player who doesn't shoot well, but he's a very good post up player and one of the best in the league at creating free thow opportunities by abusing smaller players. It would be great to see him as the third wheel on a good team someday since he's been stuck on bad ones for his whole career. Marion's become invisible on offense since leaving Steve Nash, as many predicted. He can't create his own shot and the Heat don't have a point guard, and his odd shooting stroke means that he requires lots of space to get it off. He's still incredibly valuable as someone who can guard five positions. Curry is a dominant post scorer, but a famous whiner and a nonexistent defender and rebounder. Illinois will run offense through Wade, but teams will pack it in: no one wants to shoot threes except Marion, he doesn't shoot them too well, and the other four players score the vast majority of their points in the paint. Their spacing will be all screwed up, and somebody (Maggette? Livingston) is going to be missing a bunch of open jumpers.

Interesting South Carolina team that might give Illinois trouble, but they have one point guard too many and not enough wing guys. Felton has made himself into a solid starting point guard. Anthony Johnson is perhaps stretched as a starter at the point, but he's a better shooter than Felton. O'Neal has really faded on offense, but part of that might be due to the toxic situation in Indiana, and he's still an incredible defensive player. As is Kevin Garnett, my pick for league MVP this year. Brown might be the most disappointing number one overall pick in the history of the NBA, but at least he's still big and athletic, even if his heart and offensive game never came around. The offense should put O'Neal on the block and Garnett in the high post, with Felton and Johnson spotting up and Brown collecting garbage.

Team: Illinois
Wins: 570
PPG: 93.99
RPG: 41.09
APG: 17.98
TPG: 18.35

Team: South Carolina
Wins: 430
PPG: 91.289
RPG: 48.09
APG: 23.36
TPG: 19.363

This is pretty tight in the simulator. Illinois has the shooting advantage at about 50% to 45% overall, but S.C. makes up for that with a seven rebound per game advantage. South Carolina doesn' t really have any wing players; does that hurt them? I say no. They match up well with Illinois for two main reasons: 1) Felton can guard Wade. He's a quick, strong player who gives up a few inches, but it won't be the kind of mismatch that Wade can regularly exploit. 2) Maggette prefers to play on the block and doesn't scare you with his jump shot. O'Neal is a fantastic post defender; he can lay off of Maggette and make him take a bunch of jumpers that he doesn't want to take and make life difficult for him in the post. At the other end, O'Neal's size will force continual double teams by Illinois. This is a great matchup for South Carolina and I'm going to counter-act the simulator and say that they take it, on size advantage, shooting advantage, and K.G.'s heart.

Next: #7 Pennsylvania vs. #26 Alabama
Full Bracket Here.

Friday, March 21, 2008

#11 Georgia vs. #22 Wisconsin

#11 Georgia
G Louis Williams G Jarvis Hayes F Josh Smith F Shareef Abdur-Rahim C Dwight Howard

#22 Wisconsin
G Travis Diener G Devin Harris G Anthony Carter F Caron Butler C Chris Mihm

Georgia has a nice little team that could surprise some people. Williams might be the best player you've never heard of, an extremely quick slasher who has been a big part of the 76er resurgence this winter. Hayes is a nice piece, a great shooter who can defend wings well. Smith is an athletic enigma, who is capable of leading the league in blocks or steals, and could be a truly gifted rebounder, but who you don't really trust to score when you need a bucket and seems destined to be a good stats on a bad team guy for his whole career. Shareef is near the end, but he can still score and defend in the post. Howard, of course, is the next great big man. He's taken a massive leap forward this year, so he won't look quite as good in this tournament. Pick and rolls with Williams and Howard, with occasional dumps to Rahim in the post and Hayes in the corner three point position, will be the standard Georgia offense, with Smith cleaning up the garbage.

Wisconsin is undersized across the board. Diener is a good shooter, but tiny and can't get to the rim. Harris is a little overrated; he's a very good defender, but doesn't shoot as well as you'd like and struggles with quicker point guards. Carter is a classic pass-first point guard who can't shoot straight. If you're counting, that's three point guards so far for Wisconsin. Butler is having a great season, and is a borderline All-Star every year. He's above average at every skill but his only exceptional attributes are toughness and heart. Mihm nears the end, but he's still seven feet tall, and he still makes that twelve footer from the side.

Team: Georgia
Wins: 916
PPG: 102.11
RPG: 66.82
APG: 18.47
TPG: 21.87

Team: Wisconsin
Wins: 84
PPG: 82.05
RPG: 24.31
APG: 23.23
TPG: 15.24

Not a close result. The size advantage for Georgia is overwhelming, with a sixty-two rebound advantage. And unlike in the last game, Georgia has enough ballhandlers to defeat any gimmick defenses and end possessions with Howard dunks. Actually, a lot of possessions are going to end with Howard dunks, since he'll be too much in the post for Mihm. If Rahim has anything left, he's got a big size advantage over Butler and Georgia could also force double teams or get easy baskets by posting him up. Finally, Josh Smith is 6'10" and just as quick as any of the point guards from Wisconsin that will be trying to guard him. Georgia in a walk.

Georgia next plays on 4/3.
Next: #6 Illinois vs. #29 South Carolina
Full Bracket Here.

Wednesday, March 19, 2008

#14 Massachusetts/Arkansas vs. #19 Indiana

Sorry for the delays. We'll get back on course tomorrow with a double post.

#14 Massachusetts/Arkansas
G Will Blalock (M) G Derek Fisher (A) G Fred Jones (A) F Joe Johnson (A) F Corliss Williamson (A)

#19 Indiana
G Bonzi Wells F Jared Jeffries F Zach Randolph C Brad Miller C Alan Henderson

M/A are quite undersized. Blalock's NBA career consists of 14 subpar games for the Pistons last year; this year he's out of the league. He went 6/6 at the line last year; since I didn't want to hand him 100% made free throws across 1000 games, I used his college career mark of 65.3%. Fisher is still making good money at his gritty veteran shooter gig in Los Angeles. Jones is having a nondescript season as an eight/ninth guy for the Knicks; usually the ninth guy doesn't get any minutes, but since Isiah doesn't know what he's doing, they don't have a real rotation. His game is about as exciting as his name. Johnson here continues to fulfill his destiny as the star of bad teams. Williamson retired last fall; he's now an assistant coach at Arkansas Bapist College. It's hard to be a 6'7" post player in the NBA once you get into your thirties. Blalock and Williamson are among the worst players in the tournament. M/A have no size, but they'll also have problems with quick wing players on defense. On offense, whoever is supposed to be guarding Blalock will actually guard Johnson, who isn't good enough to fight through double teams consistently. As long as defenders stay home on Jones and Fisher, they'll be harmless.

Indiana have no point guard and they're going to try to play three post players at once. It might actually work out, because Randolph likes to play down low, while Miller prefers the high post, and Henderson is a terrible offensive player so it doesn't matter if he gets touches where he likes it. One problem is that Bonzi also prefers the low block, and Jeffries can't shoot. No one on this team can shoot, actually: Randolph, almost certainly for the first time in his life, leads his team in three point percentage. Anyone who plays them will play zone (well, as much zone as the NBA allows). The simulator isn't going to know how to deal with this team. Bonus: Bonzi and Zach are among the league's finest sulkers and pouters. Defensively, they'll struggle to defend quick point guards, but Henderson is a solid post defender and Wells and Jeffries excellent wing defenders. Randolph, of course, is famously terrible.

Team: Arkansas/Massachusetts
Wins: 5
PPG: 70.75
RPG: 15.86
APG: 15.49
TPG: 16.7

Team: Indiana
Wins: 995
PPG: 100.05
RPG: 76.29
APG: 18.24
TPG: 24.36

There's no way that the matchup is this lopsided. This is a problem with the simulation: it's designed for teams that have the basic composition of an average NBA team and these teams don't have that composition. One of these teams is playing three guards, a wing forward, and an undersized power forward, and the other is playing two wing guys and three post players. The simulation overvalues the size of Indiana against the quickness of M/A. So, we should basically toss out the wins and losses and look at matchups and statistics. A key to the game is: who has the advantage on offense, the undersized team or the team with no ballhandlers? When M/A have the ball, they'll be able to easily get down the floor into their offense. If I were Indiana, I'd play a triangle and two (or however close to that they could get without violating the NBA rules). Have Jeffries and Wells chase Johnson and Fisher, and put the big guys in the zone. The reason is that there's no way that the post guys can match up with all of M/A's perimeter players, and you can't play a 2/3 or a 1/3/1 because then the shooters for M/A kill you. This way, Indiana can harass M/A's best offensive players with excellent individual defenders and hide the lack of quickness of the big guys. The problem is Fred Jones: he's going to get a ton of open looks this way, and if he knocks them down, it could be a long night for Indiana. Jones shoots three pointers at about 30%; that's a sort of awful number, but it will surely go up with as many open shots as he'll see. Another weakness is Williamson: he's comfortable shooting from fifteen feet and there will be holes in the zone that he can exploit.
When Indiana has the ball, M/A must press full court after every made basket. Indiana has no one who is even an average ball handler, and they are all very slow-footed. Full court pressure from M/A's quickness will force many turnovers and create easy baskets--many more than the simulation knows about, since it doesn't consider defense. M/A should vary their halfcourt defense, playing mostly an aggressive ballhawking and trapping man to man, to attempt to force bad decisions from Indiana's ballhandlers and more turnovers, but they should sometimes fall back into a zone, to bait Indiana's lousy jump shooters into taking bad shots. Once Indiana gets the ball into the post, it's basically all over: it will be a parade of dunks and free throws for Indiana's bigs over M/A's tiny defenders. M/A should double team as often as possible when the ball arrives into the post, because no one on the Indiana side has a reliable jump shot save Brad Miller, and it's difficult to score against a post double team, even when the defenders are undersized. It would take a heady commitment and all out effort by all five M/A guys to make this work, and probably an outstanding defensive coach as well.
I'll be honest: I have no idea who would win this game. When in doubt, go with the big guys. Indiana is the pick.

Indiana next plays on 4/2 against Southern California.
Tomorrow: #11 Georgia vs. #22 Wisconsin.
Full Bracket Here.

Monday, March 17, 2008

#3 Southern California vs. #30 Iowa/Wyoming

#3 Southern California
G Andre Miller G Baron Davis F RIchard Jefferson F Craig Smith C Jason Collins

#30 Iowa/Wyoming
G Kirk Hinrich G Adam Morrison F Nick Collison F Raef LaFrentz C Patrick O'Bryant

SoCal has some weapons, but I'd say that the team is a bit disappointing for the #3 seed. First, Miller and Davis don't fit well together in the backcourt. They both need to dominate the ball to be effective; Davis is much better creating his own shot than he is spotting up, while Miller isn't much of a shooter at all. Jefferson has had some injury problems and seems like he's lost a step or two from his peak; he's not a top shelf shooter either. Things are worse in the frontcourt. Smith is a high energy player, but he'll never be a star or even someone you're happy starting since he is undersized. Collins is perhaps the worst offense player in the NBA. He's in the league for his defense, obviously, but even that has taken a step back in recent years. I worry that this team will play too much one on one with Davis and Jefferson, with Smith and Collins cleaning up the garbage--but that leaves MIller to stand in the corner and miss a bunch of jumpers. So, they could give Miller the ball and let him get into the lane and draw fouls and create shots for others--but are Smith and Collins going to finish easy opportunities? Is Baron going to be happy standing around on the perimeter. This team just doesn't seem to hang together very nicely.

Iowa/Wyoming aren't good enough to challenge them, probably. Hinrich has totally collapsed this year, but last year he was a strong defensive point guard with a decent jump shot. Morrison was very, very bad last year. He did not shoot the ball well, his defense was pitiful and he rebounded like a point guard. Collison and LaFrentz are at opposite ends of the tall slow soft jump shooting white guy career path. O'Bryant is very big and very raw and can't get on the court. This another low seeded team that will be forced to launch many bad jumpers because they don't have anyone who can play offense in the post or get to the rim. At least these guys rebound a bit better than Oregon...

Team: SoCal
Wins: 592
PPG: 85.78
RPG: 48.19
APG: 19.64
TPG: 19.48

Team: Iowa/Wyoming
Wins: 408
PPG: 82.591
RPG: 51.38
APG: 13.64
TPG: 17.79

This turned out to be a lot closer than I would have guessed. SoCal shot much better on two point attempts, but slightly worse from three point, were slightly outrebounded, and committed a few more turnovers per game. Though they attempted more free throws, Iowa/Wyoming shot a better percentage. I think that the simulation understates the advantage for SoCal. When they are on offense, who guards Jefferson? Morrison is terrible, and Collison and LaFrentz are lousy and too big. Hinrich matches up ok with Davis, but then who guards Miller? It has to be Morrison, since he at least has a chance to stay in front of him, but that leaves Nick Collison trying to guard Jefferson, and that isn't going to work. Look for a steady stream of isolations for Jefferson, with Iowa/Wyoming being forced to double team constantly. On their offense, Iowa/Wyoming won't be able to take advantage of Jefferson because LaFrentz isn't a strong post player, and Collison and O'Bryant can be handled by Smith and Collins. This is a solid win for Southern California.

Southern California next play 4/2.
Tomorrow: #14 Massachusetts/Arkansas vs. #19 Indiana
Full Bracket Here.

#4 Northern California vs. #29 Oregon

Before I get to the contest, a few administrative issues: 1) This contest was supposed to be played on Friday, but I had a laptop power adapter failure that left me computerless for the weekend. I'll post Monday's contest later tonight. 2) Everyone in America is using bracketmaker.com to create NCAA tournament brackets this week, it seems, and their servers are overwhelmed, so you might have trouble loading the full bracket in the near term. I expect things to be very busy until the tournament starts on Thursday, and hopefully things will settle down. I might have trouble loading the full bracket myself, so if you do get it loaded, I might not have been able to update it. If anyone has a possible different hosting solution for the bracket, suggestions are hereby solicited.

#4 Northern California
G Jason Kidd G Paul Pierce F Ray Allen F Drew Gooden C Tyson Chandler

A strong team that should be a contender for an overall title. Kidd is a future Hall of Famer whose play has declined considerably this year, but since I'm using last years stats for the simulation, that won't hurt the team. Pierce and Allen complement each other nicely, just like they do in real life on the Celtic wing. Both are perennial All-Stars who will have borderline Hall of Fame cases. Gooden is a solid fourth offensive option, who has a nice jump shot to eighteen feet and can be effective on the block; I'm really sorry that the Cavs traded him. Chandler is strong and athletic, with a weaker jump shot but excellent post defense and toughness. This might be the best rebounding team in the tournament: Gooden and Chandler are each excellent, Kidd has long been one of the best rebounding guards in basketball, and Pierce and Allen are adequate. This will be a running team, since every player is a superior athlete for their position and no one runs the fast break like Kidd. But they'll still have a lot of ways to hurt you: pick and roll with some combination of Kidd/Pierce and Gooden/Chandler; Pierce and Gooden on isolations; Pierce or Gooden or Kidd in the post. They'd be really fun to watch.

#29 Oregon
G Damon Stoudamire G Ronnie Brewer G Salim Stoudamire F Luke Jackson F Ime Udoka

This team doesn't have any post players; this is a terrible matchup for them. Damon is on the downside of what has been a surprisingly long and adequate career for a short point guard. Brewer has good size and strength for his position; he's more of a scorer than a shooter, but he's probably not up to carrying the load for this offense. Stoudamire has been in and out of the rotation in Atlanta. He's the sort of player that does everything sort of ok, but most teams prefer players on the bench who have an exceptional skill. Jackson is a bust who has fallen out of the league. Udoka has had some bursts of quality play with the Spurs, who need to get younger, but he's not a top shelf ball handler or athlete and is basically a three point specialists. With five guys who want to play on the perimeter, it's hard to know how the offense will work. Jackson is tall but soft and prefers shooting threes; the coach will have to make him play down low, along with Brewer, who is tough but severely undersized. They're going to be jacking up a bunch of contested three pointers (each player save Brewer has a two point rate of around 60%) and they're not tall enough to rebound the misses

Team: Northern California
Wins: 990
PPG: 112.72
RPG: 78.78
APG: 21.95
TPG: 20.618

Team: Oregon
Wins: 10
PPG: 78.7
RPG: 18.13
APG: 16.35
TPG: 12.737

It's the biggest blowout so far, says the software. Basketball rule of thumb: if you outrebound the other team by 60, you're probably going to win. It's all about the incredible size advantage that Northern California has at every position: Northern California goes 6'4", 6'6", 6'7", 6'9", 7'. Oregon goes 5'11", 6'1", 6'5", 6'7", 6'10". The simulation might overstate the California advantage a little bit, since Gooden and Chandler will have to try to guard a perimeter players at the defensive end, and that might turn out badly. But it can't turn out nearly as badly as at the other end, where Jackson and someone small will just get endlessly dunked on by the California bigs. That's before we even mention the five inch advantage that Kidd and Allen will have on whoever tries to guard them. When matchups like these actually happen on the court, everyone on the offense clears out to the other side while the little guy playing defense screams for help and then either fouls or gives up an easy basket; you see it once or twice a game. Northern California has that kind of advantage at four positions. This one wouldn't be too fun to watch.

Northern California next plays on 4/1 against #13 New Jersey.
Later today: #3 Southern California vs. #30 Iowa/Wyoming.
Full Bracket Here.

Thursday, March 13, 2008

#13 New Jersey vs. #20 Tennessee

#13 New Jersey
G Brevin Knight G Mike Bibby G J.R. Smith F Troy Murphy C Shaquille O'Neal

#20 Tennessee
G J.J. Redick G Rodney Carney F Shawne Williams F Lorenzen Wright C David Harrison

Jersey has sort of a strange team. They've got Andrew Bynum on the bench; if I was using this year's statistics instead of last years, he'd be starting. Knight and Bibby have both slipped; Knight from solid starter to end of the bench guy, and Bibby from borderline All-Star to solid starter. Smith continues to be a talented enigma as his offcourt troubles and dysfunctional relationship with George Karl slide him up and down the Denver depth chart from game to game. Troy Murphy has one of the most ungainly jump shots in the league; it kind of looks like a shot put coming off his hand and he way jumps forward instead of straight up, but he's very effective with it. Shaq was actually very good last year when he was healthy-he just wasn't healthy all that often. The offense will consist in throwing it to Shaq while the other four guys stand around and wait to shoot a jump shot. Not too exciting, but it could be effective.

Tennessee is missing a point guard; Jersey has an extra one; perhaps they could work something out. As it stands, Redick can't really run an NBA offense, and Carney and Williams are poor ballhandlers. This team should commit tons of turnovers. Williams and Carney are athletes in search of jump shots and basketball intelligence. It's hard to believe that Lorenzen Wright is still in the league, isn't it? You can't teach size. Harrison has struggled with drugs and inconsistent effort and has been buried on the Pacer bench for most of this season. There's no star power for Tennesse, and no one really in their prime: Wright is just about done and the other guys are very young. It's hard to figure out how this team will score. Nobody on the wing can get his own shot regularly. Maybe they can try posting up Harrison? He's shown flashes of talent on the block in the past.

Team: New Jersey
Wins: 700
PPG: 97.25
RPG: 45.83
APG: 21.88
TPG: 15.86

Team: Tennessee
Wins: 300
PPG: 89.517
RPG: 48.51
APG: 11.235
TPG: 20.643

New Jersey has a solid lead in the simulation. Check out that assist to turnover ratio for Tennessee; there's really no one getting easy baskets for anyone else and even that number is understated because the simulation doesn't know that one of these guys has to try to run the offense. Bibby has to try to guard Carney; that's a size advantage, but I'm not sure that Carney is talented enough to get a significant edge-especially in the post. Shaq is a nightmare matchup for Harrison; too much size and strength, and Harrison doesn't have the quickness to take advantage. Even a 75% Shaq should be enough here. Big advantages at every position on offense for New Jersey. Redick will probably try to guard Knight, but he's nowhere near quick enough. Smith and Bibby should be mostly neutralized by the length and quickness of Williams and Carney, but Murphy is a big problem for Wright because Murphy loves to hang out on the perimeter. It's hard to know what Shaq does against Harrison; healthy, motivated Shaq kills him, but we don't see that guy too often anymore. I was surprised by the Tennessee rebounding advantage; looking more closely at the numbers, it turns out that both Bibby and Smith are terrible rebounders for their size and position, and that isn't compensated for by the averageness of the other players. That's going to hurt New Jersey going forward, most likely, but not today.

New Jersey plays next on 4/1
Tomorrow: #4 Northern California vs. #29 Oregon
Full Bracket Here.

Wednesday, March 12, 2008

#12 North Carolina vs. #21 Maryland

#12 North Carolina
G Chris Paul G Jerry Stackhouse F Josh Howard F Cedric Simmons C Chris Wilcox

#21 Maryland
G Sam Cassell G Jarret Jack G Rudy Gay F Mike Sweetney C Josh Boone

Terps against Tar Heels? Should be a close matchup. On the North Carolina side, Paul is the next great point guard in the league and a dark horse MVP candidate in 07-08, though he's made significant strides from last year to this. Stackhouse is a little overripe; he's grown brittle and lost a step or two, but he's still a shooter and a heady vet. You don't hear much about Josh Howard these days, but he's settled in as Dirk's sidekick in Dallas at just below All-Star level. Simmons is a stiff who has begun the itinerant portion of his career; he's big and young, so he'll keep getting chances. Wilcox is a decent rebounding and defense big guy. You want to run pick and rolls with Paul as your bread and butter if you've got him, but Simmons and Wilcox don't move well and they are not strong jump shooters. The swingmen have high basketball IQs and can find ways to affect the game without being the focus of the offense.

Opposing them is a Maryland team that seems to lack an identity. Cassell is 38; while some people think his move to Boston puts them over the top, it's not clear that he's more valuable than Rondo at this point. Jack and Gay are similar; young, long swingmen who enjoy shooting jump shots and are a bit underathletic. Gay has a little more upside, but Jack is a more finished product: he's a better shooter and defender right now. Sweetney has eaten himself out of the league between last year and this. Boone is interesting; he doesn't really have an offensive game, but he's very long, runs the floor well for a big man, finishes his opportunities and doesn't make mistakes. Maryland doesn't have a star to bail them out at the end of the shot clock (unless you think Sam I Am can still get it done...maybe he can). They'll have to vary their offense to take advantage of matchups, but eventually they're going to run into a team that just has better players. That time might be today...

Team: North Carolina
Wins: 538
PPG: 90.905
RPG: 41.678
APG: 15.98
TPG: 15.21

Team: Maryland
Wins: 462
PPG: 90.118
RPG: 52.9
APG: 17.19
TPG: 20.38

Boy, this is just about as close as could be. Both teams shot nearly identical field goal percentages overall, and Maryland's rebounding advantage was negated by its turnover disadvantage and North Carolina's superior three point shooting. Let's take a close look at defense and matchups. First, there isn't a single Maryland player who is quicker than his North Carolina counterpart; Wilcox and Boone are a wash, and every other position favors Carolina. Cassell has a size and strength advantage over Paul; Sammy is a canny operator on the block, and Maryland would be well served by taking him into the post regularly, especially since it's important to make Paul work on defense. Otherwise, there's not much there for Maryland to exploit on offense-they just have to hope that their jump shots are falling. On the North Carolina side, Paul can get into the lane against Cassell anytime he wants; this will be a huge problem for Maryland, especially since neither Maryland big man is particularly imposing on the defensive end. Howard is bigger and stronger than Gay and Jack; whoever guards him will be vulnerable on isolations, but Howard is not used to being a focal point; his performance if placed in those situations against NBA-level defense is something of an unknown quantity. I can't emphasize enough what a fat waste of talent Sweetney is, but it's unclear whether Simmons is good enough to do anything against him. Stack might break down with so many minutes, but so could Cassell and Sweetney. I don't see anything to budge the prediction of the simulator much one way or another; perhaps just a few more games in North Carolina's direction because of Cassell trying to guard Paul. It's still damned close though; if the Jack/Gay jumpers are falling and Sammy's feeling good, Maryland would win a single game between these two squads.

North Carolina next plays on 3/31 against Florida.
Tomorrow: #13 New Jersey vs. #20 Tennessee
Full Bracket Here.

#5 Florida vs. #28 Louisiana

Sorry for the delay. I was having trouble with blogger yesterday, so you get a two for one today. Tune in this evening for Maryland/North Carolina.

#5 Florida
G Gilbert Arenas G Vince Carter F Tracy McGrady F Udonis Haslem F Amare Stoudemire

#28 Louisiana
G Chris Duhon G Danny Granger F Rashard Lewis F Antawn Jamison C Tyrus Thomas

Florida has to be the pre-tourney favorite. They've got four of the top 25 players in the league and a solid power forward in Udonis Haslem, not to mention Trevor Ariza and Reggie Evans on the bench. If your fourth scoring option is a little Vinsanity, you've got a pretty good team, I'd say. Chemistry could be a problem for this team. Gil and Stoudemire are famously impressed with their own abilities to the point where their teammates hate playing with them; and Vince is a sulky whiner who might well sabotage things without enough shots coming his way. McGrady and Haslem are by all accounts good guys who should be happy to make sacrifices to make things work. The offense will probably be Gil and Stoudemire running pick and rolls; McGrady spotting up (which is a waste of his talents), Haslem doing the dirty work, and Vince pouting and not going to the basket enough.

Louisiana is a nice team, especially for #28. It's a bummer that they have to run into this juggernaut so early, because they're just a little bit worse than Florida at almost every position. It's uncanny, actually; Duhon is a nice starting point guard; Granger and Lewis are excellent jump shooters with athleticism and size, Thomas is an incredible athlete who could dominate the league if he figures out how to play. Jamison does many things well at power forward; he's a scorer, not a shooter, but he's big and creates matchup problems with his size and athleticism. But it's hard to see how they'll match up with a bunch of guys who are All-NBA every year.

Team: Florida
Wins: 699
PPG: 100.78
RPG: 54.86
APG: 16.04
TPG: 16.62

Team: Louisiana
Wins: 301
PPG: 92.31
RPG: 40.52
APG: 15.324
TPG: 16.45

Solid showing by Louisiana against a great team in Florida. The difference was on the boards for Florida: otherwise, the team statistics were very close. Unfortunately, the simulation understates the advantage for Florida, because four of their guys would ordinarily draw double-teams, and Louisiana is going to mostly have to play them straight-up. All of the Louisiana players are good defenders except Lewis, but that's enough. He's going to have to guard either McGrady or Carter, and he can't do it: Florida will spend all day running isolations on him. At least, they would if they were smart, but there's no guarantee that Arenas, running the show from the point, won't spend all game jacking up bad threes, or that Stoudemire won't bitch and moan until a disproportionate number of shots come his way. On the Louisiana offensive end, Duhon won't be capable of much against Arenas' pressure defense, but Vince doesn't work hard on defense anymore and can be taken advantage of by either Granger or Lewis. The Florida coach would probably stick him on Lewis, since he attacks the basket less, but Vince isn't the dominating athlete that he was in his youth and Lousiana would do well to find ways to attack him. Stoudemire is not hard worker on defense either and tends to get into foul trouble, but Jamison is more of a perimeter player and Thomas doesn't have a refined offensive game in the post. Florida takes this one, and the offense of their four stars gives good reason to be optimistic about their title hopes, but defense and chemistry might cost them a close game down the line. It might have been better, from a team perspective, for Florida to take Ariza over Carter for defense and chemistry. However, I tried to pick the best five players first, without considering team makeup beyond position, which I think is the right move-if Florida should lose, I'll feel more confident in the results with Ariza on the bench than I would if Vince were there.

Florida plays next on 3/31.
Later today: #12 North Carolina vs. #21 Maryland
Full Bracket Here.

Monday, March 10, 2008

#8 Ohio vs. #25 Colorado/South Dakota

#8 Ohio
G Kevin Martin G Michael Redd F LeBron James F Ruben Patterson C Calvin Booth

#25 Colordao/South Dakota
G Chauncey Billups (C) G Darrick Martin (C) F Mike Miller (SD) F Michael Ruffin (C) C Jared Reiner (SD)

Ohio has the reigning best player in the world (though he wasn't quite at that level last year), along with what I consider to be the ideal backcourt to supplement his talents. Lots of people think that James doesn't need a point guard because of his domination of the ball and his exceptional court vision; I'm one of them. Martin and Redd are tall, athletic shooting guards who are solid defenders and rebounders, and are capable of creating their own shots but also have exceptional catch-and-shoot strokes. They don't need plays run for them to contribute, but they are capable of carrying a team offensively for stretches. Every team that Ohio faces is going to have at least one serious matchup problem due to the three of them. It's absolutely a championship level backcourt; it's a shame that the frontcourt doesn't pull its weight. Ruben Patterson is nearing the end of a solid career as a gritty, undersized rebounder and defender, and Booth is a stiff who is only in the league because of his height. This team can hurt you in a variety of ways, none of which involve Calvin Booth: pick-and-rolls with James and Patterson; isolations and post-ups for whichever wing player has the most advantageous matchup; Martin and Redd on curls for jump shots; dunks and wide open jump shots for everyone when LeBron gets into the lane.

Opposing them today is an overmatched Colorado/South Dakota team. Billups is a top five NBA point guard and Miller a very good wing scorer, but he's a lousy defender, Reiner is out of the league, and Ruffin and Martin are well on their way. If this matchup is close, there's something wrong with the simulator. Let's take a look...

Team: Ohio
Wins: 866
PPG: 100.04
RPG: 46.91
APG: 17.231
TPG: 14.44

Team: Colorado/South Dakota
Wins: 134
PPG: 83.23
RPG: 49.22
APG: 19.95
TPG: 22.96

Overwhelming victory for the Ohio side. The question, as always with any team he plays against, starts with who guards LeBron James. He's far too big for Martin and too athletic for Ruffin and Miller. That leaves Billups, who is a good defender but gives up five inches and fifty pounds and doesn't have a quickness advantage, and also is the guy who is supposed to be a primary scorer at the other end. Darrick Martin is 5'11"; who does he guard: the 6'6" Redd or the 6'7" Kevin Martin? Miller has the size to handle one of these guys, but not the raw defensive ability. All this adds up to bushels and bushels of points from the Ohio wing players. Oh yeah: Michael Ruffin isn't athletic enough to guard Patterson either. Ohio will have an incredible matchup advantage at two positions and a distinct matchup advantage at two positions and a wash at center. And they can't play zone because of the shooting ability of Martin and Redd. All the C/SD wing guys would foul out in about a quarter and a half. On the C/SD offensive end, Martin is tall and quick enough to make Billups' life difficult, while Redd and James can trade working hard on Miller with taking it easy by guarding the smaller Martin. Thus, I think that the simulation understates the Ohio advantage by about 100 games, even with the already strong 866 win number posted. Reason to worry for Ohio going forward: they were outrebounded by these guys by three boards a game. The front court is going to kill them; it's a matter of when, not if.

Ohio plays next on 3/28 against #24 West Virginia/Connecticut
Tomorrow: #5 Florida vs. #28 Louisiana
Full Bracket Here.

Friday, March 7, 2008

#9 Michigan vs. #24 West Virginia/Connecticut

#9 Michigan
G Morris Peterson G Jason Richardson F Shane Battier F Chris Webber C Chris Kaman

#24 West Virginia/Connecticut
G Deron Williams (WV) G Jason Williams (WV) F Ryan Gomes (C) F Marcus Camby (C) C Dwayne Jones (WV)

Michigan is missing a point guard, and Peterson and Richardson and Battier are not strong ball handlers. Webber is pretty much finished with that leg injury, though Kaman is a solid scoring and rebounding center. Who runs the offense? Probably Webber from the high post sometimes, and sometimes Richardson and Webber on pick and rolls, with Peterson and Battier spotting up for threes and Kaman down low.

On the other side, Deron Williams is a rising star at point guard, while Jason is pretty much a three point shooter at this stage of his career. Gomes is underrated, and does many things well, including shooting the basketball. Camby is the league's reigning defensive player of the year (though many think that he wasn't the best choice) and is one of the strongest rebounders in the world. He's also an efficient offense player, with range to fifteen feet. Jones is a stiff lay-up-and-dunks-only seven footer, though he does rebound and defend; he's only getting minutes with the Cavs this year because of their injuries. This would be an extremely interesting matchup.

Team: Michigan
Wins: 469
PPG: 92.75
RPG: 44.42
APG: 15.13
TPG: 14.72

Team: West Virginia/ Connecticut
Wins: 531
PPG: 94.05
RPG: 53.32
APG: 23.89
TPG: 18.718

The simulation has this at a virtual dead heat, with WV/C coming out slightly ahead. Remember that the simulation doesn't know that Michigan has no point guard; though they turned the ball over about four times per game less than WV/C in the simulation, almost certainly those turnovers would shoot way up with Richardson or Peterson, mediocre ball handlers even for wing players, trying to run the offense. As for defense, no one on WV/C could guard a vintage Chris Webber, but this version can't run or jump, so Dwayne Jones can handle him easily. Deron Williams has excellent size and strength for a point guard so he can't be taken advantage of defensively, but Jason Williams is undersized and a poor defender. He'll have to try to guard Peterson, and likely the Michgan offense would involve those players posting up J. Williams at every opportunity. This will force WV/U to double team at times, creating open shots for other players. However, Peterson is not an experienced or particularly skilled operator in the post--he usually just sits in a corner and waits for a three point shot to fall into his lap. Also, WV/C plays three excellent defenders alongside J. Williams, including a fantastic weak-side help defender in Camby, and also Jones, who is at least huge. So I think that much of the time, WV/C could get away with playing Peterson one on one (Aside: this discussion is pretty amusing; surely no one has double teamed Morris Peterson since he was playing at Michigan State eight years ago, and maybe not even then, since Mateen Cleaves was the star of those teams).
On the WV/C offensive end, both Williams men are quicker than anyone in the Michigan backcourt, and could be in the lane as much as they like, getting to the line and wreaking havoc, especially since no one in the Michigan front court is an imposing shot blocker. Webber surely can't give you 48 high level minutes at this point, and wasn't a strong defensive player even in his prime; luckily for him, Dwayne Jones' 7.5 usage rate means that he shoots less than nearly anyone in the tournament. All told, I think that the simulation actually understates the West Virginia/Connecticut advantage by 100 games or so, mostly due to the talents of Deron Williams and the lack of Michigan ballhandling. I think that this would be the first tournament game that I'd be really excited to see. It's also the first upset. Remember that NBA States is off for the weekend; I'll hope you'll tune in next week.

West Virginia/Connecticut next plays on 3/28.
Monday: #8 Ohio vs. #25 Colorado/South Dakota
Full Bracket Here.

Thursday, March 6, 2008

#16 Idaho/Minnesota vs. #17 Nevada/Utah

#16 Idaho/Minnesota
G Luke Ridnour (I) G Alan Anderson (M) F Devean George (M) F Kris Humphries (M) C Joel Pryzbilla (M)

#17 Nevada/Utah
G Marcus Banks (N) G Ricky Davis (N) F Devin Brown (U) F Pat Garrity (N) C Scot Pollard (U)

This is a pretty interesting matchup. I/M have a better frontcourt, and N/U have a better backcourt. Who scores for I/M? Ridnour is a decent distributing point guard who doesn't look for his own shot and Anderson has fallen out of the league between last year and this year. The big guys are not strong offensive players (though they are efficient); that leaves (gulp) Devean George, who has been in the news recently for refusing to be traded for Jason Kidd and not for any basketball-related abilities of note. Still, he's a tall, athletic guy who shoots ok; that makes him the star of his team.
On the other end, Davis is a solid scoring wing (though he's completely, unambigously crazy, which is something else that isn't modeled in the simulation). Banks has been buried on the Phoenix bench and has actually played pretty well in Miami since the Shaq trade. Brown is an interesting role player who does everything pretty well but nothing really well; he'll have to do more on this team. Garrity and Pollard are both pretty much finished, unfortunately.

Team: Idaho/Minnesota
Wins: 601
PPG: 86.293
RPG: 59.67
APG: 13.801
TPG: 25.5

Team: Nevada/Utah
Wins: 399
PPG: 82.64
RPG: 36.13
APG: 15.52
TPG: 16.96

I/M wins 60% of the simulated games. They shot the ball and rebounded a litle bit better than N/U. There's reason to think that a game between these two teams might be even closer. Banks is taller, bigger, and stronger than Ridnour, who is not a strong defender. Anderson must guard Brown; he's not a strong defensive player either. N/U could have success attacking the defense of those two players with isolations and post-up plays. Pryzbilla and Humphries are bigger and stronger than the N/U inside players, but they are not talented enough offensively to take advantage. Davis is the best player on the floor and the only one who might command a double team (I'm not sure whether George could handle him one on one). I/M doesn't have a player like that. I think that the simulation, in short, understates that backcourt advantage that I see for N/U, and overstates the frontcourt advantage of the offensively unagressive Pryzbilla and Humphries. I think that this game would be closer to a toss-up than the simulation allows.
However, I/M are stronger rebounders at every position, and when they shoot the ball, it goes in more often. I'm picking Idaho/Minnesota as the winner--narrowly. Turnovers will be a problem for this team going forward: 25.5 is an awfully high per game average.

Idaho/Minnesota plays Texas on 3/27.
Tomorrow: #9 Michigan vs. #24 West Virginia/ Connecticut
Full Bracket Here.

Wednesday, March 5, 2008

#1 Texas vs. #32 Kansas

#1 Texas
G T.J. Ford G Stephen Jackson F Mike Dunleavy F Chris Bosh C Jeff Foster

#32 Kansas
G Earl Watson G Maurice Evans G Adrian Griffin F Matt Freije F Wayne Simien

Texas has the best five of these ten players; this matchup shouldn't be close. The Texas team is well balanced with a play-making point guard in Ford, three point shooting and solid wing play in Jackson and Dunleavy, Foster provides rebounding and defense, and a highly efficient All-Star power forward named Chris Bosh will score from the high post. On the Kansas side, only Watson and Evans belong in an NBA rotation, and they really belong at the tail end, though Watson currently plays a lot of minutes for a bad Seattle team. Let's run the numbers...

Team: Texas
Wins: 981
PPG: 102.0
RPG: 74.47
APG: 22.07
TPG: 20.97

Team: Kansas
Wins: 19
PPG: 72.178
RPG: 25.205
APG: 15.91
TPG: 13.07

Wow. The undersized Simien and Freije trying to rebound against great rebounders like Foster and Bosh turns out to be sort of like a kindergardener rebounding against you and me--Texas averaged more rebounds than Kansas did points! Texas shot 45% against Kansas's 39%; shot three times as many free throws, and outscored Kansas by, on average, 30 points per game. This Texas team isn't just out to blow away teams full of guys who are barely hanging on; they're out to win this thing, as the #1 seed. And on that note, there are a few reasons to worry going forward. There isn't a true superstar on this team--someone to make something from nothing as the shot clock winds down; 20 turnovers per game is more than you'd like; Dunleavy is a lousy defender, and Ford a lousy shooter. But there's no questioning that they're winners today.

Tomorrow: #17 Nevada/Utah vs #16 Idaho/Minnesota; Texas plays the winner of that contest on 3/27.
Full Bracket Here.

Tuesday, March 4, 2008

#32 Kansas vs. #33 Washington D.C.

#32 Kansas
G Earl Watson G Maurice Evans G Adrian Griffin F Matt Freije F Wayne Simien

#33 Washington D.C.
G Delonte West G Donnell Taylor G Roger Mason G DeMarr Johnson F James White

This looks like an appropriate play in game: these teams are pretty terrible. I think they're terrible, anyway: I don't know much about a lot of these guys. These teams are pretty comparable in a few ways: West and Watson are both decent third guards who are the stars of these teams. Evans, Taylor, Griffin, and Mason are defensive specialists who have trouble scoring enough to stay in the league; Evans is the best of the bunch, but Griffin is just about finished (well, all of these guys are close to finished except Evans). Simien and White are both undersized and underathletic forwards who are already out of the league between last year and this one. The big difference is between Freije and Johnson: Johnson is an athletic lottery pick bust on the wing who never learned to shoot; Freije is a basically useless forward who is tall, but doesn't defend or rebound and shoots poorly. Kansas has a definite size advantage, but their big guys don't rebound well so much of that might be negated. I don't see much to pick between these teams before I run the simulation.

Results
Team: Kansas
Wins: 687
PPG: 86.22
RPG: 64.72
APG: 18.96
TPG: 15.68

Team: Washington D.C.
Wins: 313
PPG: 78.99
RPG: 46.7
APG: 13.63
TPG: 13.78

Kansas only shot 39% as a team and won significantly more games. The advantage on the boards turned out to be huge--around fifteen per game. DeMarr Johnson killed D.C., throwing up 16 shots a game at terrible 32% overall-he took too many threes, and he shoots those very poorly at 21%. Kansas got much better shot distribution: Watson and Evans and Griffin-the strongest shooters-took the bulk of the shots, while D.C. distributed its shots among its weaker and stronger shooters.

I don't see any reason that defense would swing the balance of the game in favor of D.C.: if anything, a more accurate simulation would swing further in the Kansas direction, to reflect the inability of a smaller D.C. team to handle the size of Freije and Simien. D.C.'s perimeter players aren't talented enough offensively to exploit mismatches on the other end consistently. I don't know who wins the Freije/Johnson matchup--most likely nobody, and certainly not the fans watching the game. Similarly, if either team would be hurt by foul trouble, it would be D.C. guarding the bigger Kansas players inside. Finally, the Kansas team features a better mix of guards, wing players, and post players, while the D.C. team is heavily guard-oriented, almost like a mid-major college team in the NCAA tournament. I think Kansas is the clear winner today--but they shouldn't feel too optimistic after shooting 39% with Texas lurking.

Tomorrow: #1 Texas vs. #32 Kansas.
Full Bracket Here.

Monday, March 3, 2008

Technical Post: How will the blog work? How does the simulation work?

How will the blog work?

Each weekday, I'll begin by listing the roster for the two competing teams. During the first round, I'll write a bit about the players on each team, and the way in which these players fit together. I'll make a note of players that were left off, or if there was a difficult decision between one player or another in the construction of a team.

I'll then run a simulation of 1,000 games between the two teams, using the software that I've written. This happens in about twenty seconds, once the data for the players has been entered. The software will keep track of team and individual statistics for points, assists, rebounds, turnovers, free throws, three pointers, and shooting, three point, and free throw percentages for the simulated games--available on a total and a per game basis. I'll present to you the wins and losses for each side, and team statistics for points per game, rebounds per game, assists per game, and turnovers per game during every post. I'll mention other team statistics and individual statistics as they seem relevant in my discussion of the results.

Using the simulation results as a starting point, I'll pick a tentative winner of the game: the team that won more of the 1,000 simulated games. But I'll then try to figure out whether the results of the simulation are misleading. All simulations are limited; I'll be up front about the limitations of this simulation below. Perhaps obviously, the closer that the 1,000 games are split between the two teams, the more likely it will be that this discussion will counteract the result of the simulation. I'll pick a final winner of the matchup and update the bracket, and leave comments open so that you can discuss the results.

How does the simulation work?
The simulation starts with nine different metrics for each player. I'll name the metric, explain it, and present the high number, low number, and median for the 165 players in the sample. Each of these metrics is either available or derivable from information presented at espn.com. I used players and metrics from last season: the 2006-07 season. Note that you won't see many very good players at the extremes here; very good players tend to have a broad range of skills; these guys are generally either specialists, on the high end, or guys on their way out of the league, on the low end.

Two point percentage: The percentage of the player's two point shots that he makes. Players who take many jump shots will have a lower two point percentage; players who shoot almost exclusively layups and dunks will have a higher percentage.
High: Alan Henderson, Indiana, .642
Low: Michael Ruffin, .278, Colorado/South Dakota
Median: .475

Three point percentage: The percentage of the player's three point shots that he makes.
High: Jason Terry, .438, Washington
Low: Several players in the sample made no three point shots in 06-07
Median: .308

Free throw percentage: The percentage of the player's free throws that he makes.
High: Lindsey Hunter, Mississippi, .909
Low: Lorenzen Wright, Tennessee, .287 (!)
Median: .763

Two point rate: The percentage of this player's shots are two point shots, as opposed to three point shots. Many big men take two point shots at a rate of 90% and up; a two point rate in the 70s or 80s is normal for a guard. Percentages in the sixties or lower indicate that the player is a three point specialist.
High: 100%; several players in the sample have never taken a three point shot in the NBA
Low: Travis Diener, Wisconsin, .375
Median: .814

Free Throw Rate: Free Throw Attempts Divided By (Field Goal Attempts Times 2). This number is used to determine the times when a player is fouled while shooting. If you're a hulking lay-up and dunk only type shooter, this number will be high; if you take many shots and they're all open threes, it will be lower.
High: Dwayne Jones, West Virginia/Connecticut, .698
Low: Roger Mason, Washington D.C., .045
Median: .158

**Interlude**
These last four statistics were created by John Hollinger, who now works for espn.com. The stats are defined here(Remember that I'm using last year's stats, so the stats that you see there won't match up with what I'm using). They measure rebounding, passing, turnovers, and possession usage in two ways that are important for simulation purposes:
1) NBA statistics are traditionally measured in terms of "x per game": assists per game, points per game, rebounds per game. No one knows that Kobe Bryant and Manu Ginobli are nearly equally valuable players on the court, in part because Kobe plays so many more minutes than Ginobli, so his per game averages are higher. These statistics are minutes independent; they allow me to compare players who played many minutes with those who played few.
2) A team like Golden State runs up and down the floor and takes many quick shots; a team like Houston walks the ball up and down the court. A player on Golden State therefore has many more opportunities to score points, make assists, and gather rebounds than a player on Houston, which clouds the usefulness of per game statistics. These statistics are "pace" neutral; they don't penalize a player who happens to play on a team that plays slowly, or reward players who play on teams that use possessions as quickly as possible.
**End Interlude**

Rebounding Rate: Hollinger metric: the percentage of missed shots that happen while a player is on the floor that he rebounds.
High: Tyson Chandler, Northern California; David Lee, Missouri; 20.7
Low: Darrick Martin, Colorado/South Dakota, 3.2
Median: 9.3

Assist Ratio: Hollinger metric: the percentage of a player's possessions that end in an assist.
High: Anthony Carter, Wisconsin, 38.4
Low: Alonzo Mourning, Virginia/Delaware, 2.5
Median: 14.6

Turnover Ration: Hollinger metric: the percentage of a player's possessions that end in a turnover.
High: Joel Pryzbilla, Minnesota, 28.2
Low: James White, Washington D.C., 4.4
Median: 11.2

Usage Rate: Hollinger metric: the number of possessions the player uses per 40 minutes. Basically, this stat tells us how often does the player does something noteworthy with the basketball. This is an important statistic for player evaluation purposes because of the NBA shot clock--because time is so limited, a team needs players who can create shots from nothing. The best players tend to have high usage rates--though the correlation is not ironclad.
High: Tracy McGrady, Florida, 32.9
Low: Michael Ruffin, Colorado/South Dakota, 5.8
Median: 18.5

*****

I'll chart briefly now how the simulation works. In short, the simulation generates many random numbers and compares them with player metrics to determine outcomes. Each basketball possession must end with someone doing one of two things: either shooting or turning the ball over. The simulation uses Usage Rate to determine who does something, and Turnover ratio to determine whether that something is a turnover--in which case, possession switches to the other team. Otherwise, it's a shot, and the simulation uses the Two Point Rate of the player taking the shot to determine whether it is a two point or three point shot, and the relevant percentage to determine whether the shot goes in, and the Free Throw Rate to determine whether the shooter was fouled. If the shot is made, the simulation uses the Assist Ratio and Usage Rate of other players on the team to determine whether the basket was assisted. If the shooter was fouled, the shooter's free throw percentage determines how many of the relevant number of free throws the player makes. If the shot misses, or if the last free throw misses, the Rebound Ratios for each player, modified with a penalty for the offensive team to reflect the difficulty of offensive rebounding, determines which team and player get the rebound. That's probably enough detail; I'll make a further technical post somewhere along the line if there is interest.

Limitations of the Simulation

There are three major issues:
1) Defense is not considered. As you might know, metrics that measure the value of defense in basketball range from terrible to less terrible. As Hollinger said in his 2003 basketball statistics annual, "It's doubly hard to talk about defense in basketball, however, because the numbers aren't there to support a discussion and the ones that do exist mislead or confuse us...at least as often as they represent an honest portrait of a player's defensive skills." So, I ignored defense entirely when I wrote the simulation software in the spring of 2004. My sense is that the metrics have not appreciably improved in the ensuing four years; I'll continue to ignore defense.

2) Substitutions and fouls and injuries. Some players commit many more fouls than others; some players get hurt a lot; some players are constantly out of shape. These players hurt their teams by being unable to play. But there is no notion of 'unable to play' in my simulation. The reason for this is that I'm not sure how to implement artificial intelligence to make substitution decisions that don't cloud the result of the picture. This is my limitation, not a limitation with the available data. So, I stick to a simplified five on five game. Since many states can barely field five players, this is a forgivable sin for this project.

3) Team composition. The simulation doesn't know that a team full of Shaqs wouldn't be able to bring the ball up the floor. It doesn't know that a team full of three point specialists wouldn't shoot at their usual percentages, because no one would draw the double teams that leave them open. For these sorts of reasons, teams that are playing many players out of position may behave oddly in the simulation.

I'll mostly be thinking about these three issues when I think about the limitations of the simulation. Your comments are invited; tune in on Tuesday for the first game in the tournament: the play-in game between Washington D.C. and Kansas.

Why We're Here

The idea for this blog stems from an incredibly interesting map of the birthplaces of NBA players that I stumbled across here. This map appealed to two of my passions: the NBA, and amateur data analysis. I've composed, with the help of this map, the best starting five of NBA players possible from each state and constructed a March Madness-style single elimination playoff between those teams. On this blog, together we'll go through the tournament, matchup by matchup, each weekday from today until April 15th, to pick a "state champion." We'll do this with the help of basketball simulation software that I created for my senior research project in college. This introductory post describes how the teams and bracket were constructed; the following technical post will say a bit about the the simulator, and describe the way in which I see the blog unfolding; then this evening I'll present the first matchup, between Kansas and Washington D.C.

While there were 22 states that produced enough players for a full squad, some states can call no current NBA player their own. Apologies to Alaska, Arizona (somewhat surprisingly), Hawaii, Maine, Montana, Nebraska, New Mexico, North Dakota, Rhode Island, and Vermont. Sometimes, the 'best' five was easy to determine; sometimes, I had to make tough decisions about which players to include, as some states have produced quite a few players. I informed my judgments with statistics, but I also tried, insofar as it was possible, to create a balanced team by position. I'll make a note of a tough decision between players where it seems relevant.

There were 18 states that had a few NBA players born there, but less than a full starting five. I partnered each of those 18 states together, creating 9 full squads. This process was somewhat arbitrary: I tried to first match numbers: if one state had three players, I matched it with a state that had two. Beyond that, I looked at positions: if there was a state with four players and no point guard, I matched it with a state that produced one point guard, for example. I didn't consider the quality of the players at all when I made these decisions; it might be the case that a different arrangement of partnered states would change the results of this tournament significantly.

I had at this point a running total of 31 squads. I made two executive decisions. First, the District of Columbia produced enough players for a full squad, so I added them in to make the total 32. Second, the state of California has so many more people than other states and produced so many good players that I decided to cut the state in two: Northern California and Southern California produce separate squads, for a final total of 33 squads.

This number lends itself to an NCAA bracket-style competition--almost. Since 32 is a power of 2, we have one team too many to make a clean bracket, so we'll have a play-in game. I've seeded the teams by total state population, or in the case of the combined teams, by the populations of the two combined states (For California, I divided up the major metropolitan areas between northern and southern and assumed that rural areas come out in the wash, which is accurate enough for my purposes). The two teams with the lowest population, Kansas and Washington D.C., meet tomorrow, with the winner facing top seeded Texas on Wednesday. The full bracket, with seedings and dates for contests, can be found here. I'm not thrilled with the presentation at Bracket Maker, but it seems to me to be the best available that is free and Mac-compatible. If anyone has a better solution, please leave a comment on this post. Thanks for stopping by and read on for more details!