2025 men's Final Four preview

LIVE FROM SAN ANTONIO: Armageddon, then more Armageddon

EDITOR’S NOTE: Hi! I am the editor. I am also the writer. I do everything around here. Anyway, if you’re in San Antonio and would like to say hello, let me know. I have some scattered-about plans but will be around starting Friday morning. I’m not currently planning on going to the games because the prices are $Texas (also, I am not credentialed…sorry!) but if you’re not going and want to watch, I will probably join for a bit. Anyway, previews.

---

Enough about the aspects of this game that may or may not foretell what is to come in college athletics. No more discussion of how boring it’s been to this point. Here, we all band together for one hope: that this game rocks as hard as everything suggests it should.

Ken Pomeroy’s site, using the term FanMatch, has assigned the rough equivalent of the lovely Fangraphs NERD metric to games for 15 seasons. In the average season, there’s about 6,000 games of Division I basketball. Even reducing that down to ~5,500 or so for the first few years of the database due to fewer D1 teams, this still means that over the last 15 years, Pomeroy’s site has assigned FanMatch ratings to around 87,000 different basketball games.

Of those 87,000 or so, this game is the third-best ever. The game that will follow, Houston-Duke, is merely the 22nd-best out of those 87,000. If you are lucky, you get one or the other of these two games once a year. You are getting both in the span of five total hours. Okay, maybe 5.5 with commercials, but still: two of the greatest games of the last 15 years, on paper, in one April evening.

A game does not get a high FanMatch rating without being projected to be very close and between two great teams to begin with. Each game offers both, and it’s to the point that you can use almost any ranking or sorting metric out there to pick your winner.

Want to go with KenPom’s picks? Get ready for a Duke/Florida title game. Prefer Torvik? Houston/Auburn. EvanMiya? Duke/Florida. The Massey Ratings consensus, covering 56 different ranking systems? Duke/Auburn. Results over the last 10 games only? Houston/Florida. Results over the last 20 games? Duke/Florida. Games against top-50 teams only? Houston/Auburn. Synergy Sports’ schedule-adjusted numbers? Duke/Auburn. Most future NBA players? Duke/Florida. Want the better defenses? Houston/Auburn. Want the better offenses? Duke/Florida. Best Shot Volume teams? Houston/Florida. Best records in close games? Houston/Auburn. Prefer games against non-conference competition only? Duke/Auburn. Prefer to pick based on which schools have a Popeyes closest nearby? Houston/Florida.

The point is this: even if I pick both favorites, there’s just a 30% chance I get these games right. That’s how close these are. By pure pregame combined spread of +2.62 points at KenPom, this is the closest we have ever had a Final Four in the modern era. The only one within a point is 2007, which came out at +2.9 on KenPom pregame. (Vegas has it at +7, which is the closest combined spread since the Wacko Tournament of 2011.)

So: regardless of what happens in either of these games, on paper, this is The Greatest Final Four in History. May we all be so lucky to actually see it live up to the title.

(1) Florida (-2.5) vs. (1) Auburn

6:09 PM ET, CBS

Look, you knew this already, but now that I don’t have to cover the team for 40 games a year, you’ve probably been made aware that my college basketball team of choice is Tennessee. Sue me. I went there, I have lived in Knoxville for 14 years now, I grew up listening to them lose to New Mexico and Nebraska and sometimes even George Mason. GEORGE MASON. DO YOU HEAR THAT, MASON FANATIC?

So, in one sense, this is a nightmare game. In the other, as someone who is a college basketball loyalist above any one team, this is one of the greatest games imaginable. I mentioned earlier the rareties this game has by pure FanMatch score at Ken Pomeroy’s site, but this is also just the fifth game between two top-four teams in the last four seasons. Any time that Ken’s projected score is 81-80 I think you’re in for a treat, perhaps doubly so given the circumstances.

Now, could this game suck? Of course. Any game could suck at any point in history. Last year’s title game, between #1 and #2, was pretty boring after halftime. The highest-rated game of 2022-23 was a Gonzaga/Kentucky game that Gonzaga led by 16 at halftime. But generally, games projected to be this good are this good; the average final score of games north of a 90+ FanMatch rating in history is 89-80, and only one of the seven had a final margin north of 11. If that’s the final score here I’d imagine most people will be happy.

Now, a minor problem: we did already see this game this season. Florida defeated Auburn 90-81 on February 8 at Neville Arena, leading by as much as 21 before a late-game scoring burst made the margin a bit closer. Some reasonably notable things did happen to swing that outcome: Florida shot 39% from three and 83% from the FT line, while Auburn went 32% and 61%. Auburn actually won the Shot Volume battle, getting off the equivalent of 81 shot attempts to UF’s 73 (also producing UF’s third-worst REB/TO margin of the year), but Florida generating 15 more points from jumpers swung the game.

Plus, Alijah Martin wasn’t available that game for Florida. The Gators are now at full strength, and it’s Auburn who has caught a bit of an injury bug. Johni Broome looked to potentially pick up a season-ending elbow injury (not to mention a minor knee issue) during the Elite Eight win over Michigan State, yet returned to grimace through making a three. Both Pearls (Bruce and Steven) have erred on the side of caution, stating that he won’t practice until he feels rehabbed enough to do it. I’m going to assume it would require a true disaster for Broome to not play, but he’s unlikely to be 100%.

The problem with that is that Broome hasn’t been at 100% since he got injured in mid-January and he’s still been one of the best players in America, albeit significantly less of a world-beater than he was prior. (FWIW, Chad Baker-Mazara seems to get injured every single game, but as usual I’ll assume he’s his equivalent of 100% until he’s not.) So: more or less, the game is what the game is. Let’s basketball.

When Auburn has the ball

Expect a lot of actions through the post. This Auburn team has actually played quite a bit slower post-Broome injury in January. While the tempo itself is roughly the same, the number of possessions lasting longer than 15 seconds has gone from 48% in non-conference play to 54% over the last 10 games. Not nothing, obviously, but they’re a little more deliberate, which has led to less true transition action and a greater emphasis on their tried-and-true post actions.

Obviously, the first, second, and third option in the post is Broome. He carries a matchup advantage over pretty much everyone he can potentially face left in the field, save for maybe a hypothetical 1-on-1 deal with JoJo Tugler in an Auburn/Houston title game. (Yes, technically, Cooper Flagg would be his matchup in an Auburn/Duke game, but in the earlier version of that game in December, it was Maliq Brown and Khaman Maluach who mostly traded off defending his post-ups.)

In the first matchup between these two, Florida refused to double the post, which can be a death sentence (as Michigan State found out) but can also give you life elsewhere on defense. Not doubling the post dampens the chance of the inside-out play Auburn loves so much. The Tigers don’t actually use that inside-out stuff to create a ton of catch-and-shoot threes, but rather use the attention Broome draws to create cuts like these from the perimeter to create even better shots.

It’s not to say that Florida never doubles or helps in the post, just that their personnel are beefy and large and pretty strong. They usually do not need to help, to the point that I have the Gators with the fifth-lowest Post Aggression Index, i.e. how frequently you are marked as doubling versus your total number of post-up possessions allowed. (On the other end, Auburn is 8th-lowest. Houston, meanwhile: 8th-highest. If you want a true Styles Make Fights match down low you probably want Houston to win the other game.) Basically, this is a question that Florida is going to ask Broome often in this game: can you beat us 1-on-1? If so, we’ll take it, because your two points is fewer than three, and we don’t expect you to get those two points every time down.

One potential X-factor, however, would be the usage of Chaney Johnson as a post presence instead. Whereas Broome is a known quantity, it’s actually Johnson who’s been far more efficient in 1-on-1 post battles, albeit on far fewer opportunities. Johnson is usually a recipient of the numerous dump-offs and cuts Broome’s gravity creates down low, but on true 1-on-1 post-ups this year, Johnson sits at 1.06 PPP (84th-percentile and is outstanding when allowed to get to his left shoulder.

Chaney Johnson post-ups over left shoulder: 39 points on 32 possessions (1.22 PPP)
Chaney Johnson post-ups, right shoulder or facing up: 16 points on 20 possessions (0.8 PPP)

If Johnson can get down low and to his left shoulder, it could be a secret area of growth for Auburn in a game where they need everything they’ve got. Johnson usually comes in for Dylan Cardwell around four or five minutes into the game to allow Broome to play at the 5. Florida usually matches Thomas Haugh and Micah Handlogten together at this time to give the starting frontcourt of Condon/Chinyelu a breather. The one area I noted Florida has had genuine trouble in this year in the Texas Tech preview was 1-on-1 postups…which Texas Tech proceeded to exploit to 11 points in 1-on-1 matchups, Florida’s third-highest number allowed this year.

Johnson does not match up the exact same way, but it’s similar enough - he and Darrion Williams are both ‘playmaking bigs’ per Synergy, after all. Haugh has actually been Florida’s best post-up defender this year, but the sample size is pretty small, and on the whole, he’s not as good defensively as Alex Condon. If you want to find a true ‘area of opportunity’ here for Auburn, it might be this, which is something that for better or worse I haven’t really seen discussed out and about just yet.

Beyond that, one other major note from the first battle between these two: Auburn became unusually P&R-heavy against Florida’s varying coverage. While the Gators opted to play in drop coverage as their base, they mixed in some hedges as well as switching based on lineup personnel. This led to a relative oddity based on Auburn’s season on the whole: the third-highest number of field goal attempts by the ball-handler in a pick-and-roll for Auburn all year, and the second-highest usage of P&R overall (Creighton, Round of 32).

Given that Florida’s P&R coverage is very good and really hasn’t been beaten by anyone since…well, playing Alabama twice, it could seem a fool’s errand for Auburn to go back to this same strategy, especially since they’ve been tremendous all year at creating great looks from off-ball screens and cuts. (Pearl’s specialty as a coach, after all!) Still, they’ll have to go to some of it because everyone does. If this is a day where Tahaad Pettiford looks like an undeniable top-20 pick in the next Draft, I think Auburn’s scoring too much for Florida to keep up. If it’s a day where Auburn gets bogged down, misses jumpers, and Broome isn’t the Player of the Year in the post…well, hey, still a great season. Right?

When Florida has the ball

They’ll start the same way they’ve started all year: attack, attack, attack, all through the backcourt. The only team left in the field with a comparably amazing offensive backcourt actually might be Houston, which is insane to say but I feel like I could defend it pretty well. (Look, how many backcourt in America are STARTING three guys all shooting 42% or better from deep?)

As has been the standard for some time now, this will revolve heavily around what magic Walter Clayton can work out of the pick-and-roll, dribble handoffs, or even spot-up threes. Yes, somehow in April 2025, people are still allowing Walter Freaking Clayton to take catch-and-shoot threes. (In Texas Tech’s defense, they held Clayton to zero points from catch-and-shoots for the first time since a March 5 battle with Alabama, so they did a good job.) While Auburn tracks as a slightly less aggressive P&R defense than Florida, they mix in more drop than UF does with more switching.

I don’t think that drop coverage is going to be the way to win here, obviously, but against Clayton there are no perfect answers. UConn arguably came closest of anyone this entire Tournament and he dropped 23 on them with seemingly six straight amazing shots to close it out. In that one, UConn heavily upped their rate of hard hedges to simply get the ball out of Clayton’s hands. Their theory was pretty simple and kind of the reverse of Golden’s defensive one: we will have anyone other than your best player beat us. He still beat them, but it wasn’t easy, and it took about 30 minutes to really break it.

Auburn’s problem: they kinda tried that already and still lost. Auburn forced the exact same percentage of ‘single’ coverage at Synergy (i.e., drop/at-the-level coverage), and at times, they extended very far out into a true trap of the P&R, almost like a Houston. Didn’t matter; Clayton nailed a cross-court pass to an open Urban Klavzar to beat it.

Therein lies the problem with playing against Walter Clayton: you can do a lot of things right and still lose, as he is Walter Clayton and you are not. Dropping against him, even in a high P&R, can be a death wish; Clayton has shot 43% on his 80 top-of-key threes this year and is an elite shooter from the wings. Florida’s fortunes here center pretty heavily around what he’s able to do.

I figure Auburn’s plan will be similar to UConn’s, which was (for 30 minutes) similar to Texas Tech’s: we want the ball out of Clayton’s hands, and if Alijah Martin or Will Richard or Thomas Haugh beat us, we’ve gotta tip the cap. But while Clayton was indeed a superstar in the first battle and has been for months, he didn’t do it alone. He hasn’t in general.

Clayton may be the straw that stirs the drink, but the second and third-most important players are not Martin or Richard, at least against top-end competition. They’ve been Haugh and Alex Condon, who have been astonishing against Florida’s toughest tasks. Haugh has a 134 ORtg on 21% USG against top-25 competition; Condon, a mere 114 ORtg on 22% USG while shooting 59% on twos and posting an 11.4% OREB%.

Against the best competition they’ve played, Florida’s transition rate has been cut by about 30%, but their usage of frontcourt actions - post-ups, cuts, rolls, even inside-out play - has risen by about 20% to compensate. Much like Florida’s defense, Auburn basically never doubles in the post, which won’t be a concern against Chinyelu (an awful finisher/hook shooter for his size) but could be a problem against Condon, Florida’s best offensive piece in the post. Plus, if Auburn gets overly aggressive in getting the ball out of Clayton’s hands, it might turn into a 4-on-3 possession for the Gators if Auburn can’t force a turnover, and Haugh in particular has been excellent as a roller in Florida’s P&R sets.

Thoughts, predictions, etc.

THING TO CTRL+F THE BOXSCORE FOR: Each team’s two-point percentage. Auburn has not lost a game all season when they’ve held their opponent below 47% from two. Florida hasn’t lost a single game this year when their opponent has shot 46% or worse from two. The problem, as much as it exists: Florida has been held below 47% five times all year - twice by Tennessee, only once (UConn) in the last two months - and Auburn below 46% just thrice, two of which were with Broome out hurt. Someone has to crack, but if it’s neither and this is just point after point, it’s going to be a blast.

THING TO LOOK AT THE SCOREBOARD FOR: The halftime score. Yes, I’m being serious. Auburn’s just 4-11 against the first half spread in their last 15 games. Now, this comes with the note that the only team with a higher second half power rating at TeamRankings this year is Duke, so if they can just keep it close (or even lead) they’re in business. It also comes with the note that Auburn shoots 48% 2PT/31% 3PT in the first half post-Broome injury, and if you do that against Florida you’re gonna be down by 5+ at the break.

X-FACTOR, AUBURN: …is it somehow Chad Baker-Mazara? The Troll of Trolls has played his very best in the last three months against Auburn’s best opponents, shooting 44% from deep in 13 games against top-25 competition. If this is a Good CBM Day, watch out. If this is a Bad CBM Day, watch out if you’re literally anyone else on the floor.

X-FACTOR, FLORIDA: Condon or Haugh, for reasons listed above, but the always-lurking deep sleeper of any game is Will Richard. It is at least somewhat noteworthy to me that of Florida’s seven sub-80 point games this year, Richard scored single-digit points in three of them. A Richard no-show or a bad day from three limits Florida’s paths a bit.

THE STRUCTURAL REQUIREMENTS OF SPORTSWRITING FORCE ME TO DECLARE: That this game probably comes down to who has the better-functioning half-court offense against a really good half-court defense. NCAA Tournament games are routinely slower by about 2-3 possessions versus the season-long average as the stakes get higher, the anxiety rises, and coaches feel the need to step in with their best designs of the season. This is projected at 70 possessions at KenPom, but based on Tournament averages it’ll be more like 68-69.

You can go in any direction you want based on that. This year, Auburn has the superior half-court offense (against a tougher schedule, too) to the tune of about 5 points per 100 possessions. The two defenses are almost dead-even in half-court D. Auburn’s inability to fully limit transition opportunities would worry me if I’m a fan of theirs, but Florida has not been elite at it, either. If this game does get up-and-down it certainly favors Florida, who has the significantly better two-way transition game and would likely expose Auburn’s inability to stay out of foul trouble.

I ended up being swayed to Auburn for two reasons: it does not seem like the Broome injury is going to meaningfully limit his minutes or what he does, and while Florida has been the slightly superior team overall (to the tune of a 0.8 point advantage per 100 possessions at KenPom), I do feel a little more favorability with the team that has executed a slight touch better each way in half-court play this year.

PREDICTION: My best guess is Auburn 81, Florida 79. My only other note: of the two games, I have a harder time seeing this one end with a margin of 10+ either way for the sole reason it’s less reliant on jump shooting.

(1) Duke (-4.5) vs. (1) Houston

8:49 PM ET, CBS

A short history of 1 vs. 2 matchups in KenPom history, as this is #1 Duke playing #2 Houston:

• November 9, 2024: (2) Auburn 74, (1) Houston 69

• April 8, 2024: (1) UConn 75, (2) Purdue 60

• April 5, 2021: (2) Baylor 86, (1) Gonzaga 70

• February 22, 2020: (1) Kansas 64, (2) Baylor 61

• November 5, 2019: (2) Kentucky 69, (1) Michigan State 62

• January 19, 2019: (2) Duke 72, (1) Virginia 70

• April 4, 2016: (1) Villanova 77, (2) North Carolina 74

• November 17, 2015: (2) Kentucky 74, (1) Duke 63

• April 7, 2008: (1) Kansas 75, (2) Memphis 68 (overtime)

• April 4, 2005: (1) North Carolina 75, (2) Illinois 70

• April 3, 2004: (2) UConn 79, (1) Duke 78

• December 21, 2000: (2) Stanford 84, (1) Duke 83

• March 27, 1999: (1) Duke 68, (2) Michigan State 62

• March 8, 1998: (2) North Carolina 83, (1) Duke 68

• February 28, 1998: (1) Duke 77, (2) North Carolina 75

• February 5, 1998: (2) North Carolina 97, (1) Duke 73

Now, if you’re a Duke fan, you may want to look away from this list for a couple of reasons. For one, the #1 team went 7-9 in these matchups; for two, Duke themselves went 3-5. The other problem here is another piece of history: the opponent getting to play an in-state Final Four game. Duke fans will almost certainly outnumber Houston’s because of the sheer size disparity amongst the fanbases, but teams playing a Final Four game in their home state went 7-1.

Am I writing all of this because, personally, I think it would be the coolest story the sport has produced in many moons if Houston were to be your national champion? Maybe. Or perhaps I am simply a little too attuned to history and what it means for overly obvious favorites to win it all. After all, nine of the last 16 odds-on favorites entering the sport’s final weekend failed to bring home the trophy. Maybe this isn’t just a coronation for Cooper Flagg.

Alternately, maybe it is. Duke sits as the second-best team in KenPom history at the time of writing, second to only the 1998-99 version of themselves, meaning you’re witnessing the best college basketball team to touch a court in 25+ years. They have the best offense in KenPom history and have added themselves to a list of just seven teams in the last 25 years to have a top-5 offense and defense at KenPom. The last three - 2008 Kansas, 2010 Duke, 2019 Virginia - all won the title.

But the first three of those (2002 Duke, 2004 Duke, 2007 UNC) did not, and even the champions all required some great strokes of fortune to win it all. It’s no lock. If it’s Houston’s time, it would be beyond well-deserved for them, too. They’ve been inside the KenPom top 5 for the entire season, started out #1 overall, improved heavily by pure Net Rating from that preseason projection, and only finished #2 because Duke went supernova behind one of the best freshmen we’ve ever seen. It’s the best offense playing the best defense, and arguably, it feels like the ‘real’ national title game.

And yet: it isn’t. Whoever survives this however they do, they must move on to play another 40-or-more-minute-war just 48 hours later at the same location. Even the actual Battle of the Alamo only lasted 90 minutes total.

When Duke has the ball

Well, I’ll assume you have heard of one Cooper Flagg. We are in agreement that he is the Matchup Problem in college basketball, right? Right. Alright, with that in mind, let’s figure out how Duke’s going to use him, particularly in a weird matchup he’s seen a pretty small amount of this year.

Over the last two months, Duke and Houston have both been a bit better in first halves than second, which should mean we get to see the best of each to start before fatigue sets in a little bit each way. With that in mind, Duke has played a bit differently in first halves than second halves offensively. The first half is much more perimeter-based and features a ton of transition play with attack-and-kick actions; the second half sees rim attacks and drives rise by nearly 40% as kickout threes and transition play fall accordingly.

Flagg can score from pretty much anywhere in pretty much any action, but his top action this year has unsurprisingly been himself as the ball-handler in Duke’s numerous ball screen sets. We mostly see Flagg dribble off of the pick, as just 10% of the time he rejects it. This will really be put to the test here. No team in the entire sport has a more aggressive perimeter coverage than Houston, who blitzes ball screens more often than anyone and plays the lowest amount of drop coverage out there.

The only full-blitz coverage Duke has drawn all year of real decency was Wake Forest, who Duke tangled with twice. The first time around, Duke generated just five points from transition opportunities (tied for a season-low) and saw Flagg commit seven turnovers. The second time, Duke adjusted fully, dropped 93 points, and Flagg lit Wake up for 28 points on 16 shots with just one turnover. The difference-maker: Duke settled a bit less for deep threes, got out faster off of misses, and flexed their muscles on the boards.

It’s going to depend which version of Flagg, and which version of Duke, we end up getting. A deferential Flagg who looks a bit puzzled at first by the most aggressive and best defense he’ll play against in his one-year college career could make some really, really bad plays.

Alternately, as he did later on against this defense, he could use the aggressiveness as a foil, choosing to re-use or reject the screen entirely. He’s an incredibly smart player with a great skill set, and hey, maybe he just makes Houston look foolish. They haven’t faced anything like this, either.

Beyond Flagg, we’ll get to see a lot of other stars-in-training, too. Flagg’s gravity will create kickout threes for Kon Knueppel and Tyrese Proctor, and in general, Houston will give up kickout threes rather than straight-line drives to the rim even if it requires a double team. In general, the Houston structure will lead to a lot of three-point attempts from the Blue Devils, and against the only two defenses in Houston’s stratosphere they’ve played this year (Kansas and Auburn), Duke has averaged taking 45.3% of their shots from downtown. Against Wake, the most similar defensive structure on their schedule, that number shot up to 54%.

But beyond all the kickout threes, off-ball screens for Knueppel/Proctor, and even Flagg post-ups (he’ll get doubled heavily here), there is one guy that feels like the potential decider: Khaman Maluach. Maluach is a tremendous leaper and sets fantastic screens in Duke’s P&R offense. Lately, Flagg has begun to just lob it up somewhere in the direction of the rim because he knows Maluach can get to it and slam it down.

I think Houston can win in a lot of areas defensively, but the two somewhat similar guys Houston’s gone up against who can do similar things are Cliff Omoruyi of Alabama and Flory Bidunga of Kansas. Big Cliff was held down by Houston, but Flory went for 19 almost entirely on dump-offs or lobs from the Kansas backcourt, was fantastic on the boards, and was almost the reason Kansas won that game before their historical home meltdown.

Also, Houston got a lot of credit for smoking Tennessee and should, but it’s worth noting that Tennessee used off-ball screens in that game to create 15 field goal attempts, over double their previous highest amount allowed this season. Tennessee only hit four of these, but the majority of them were open looks. If Houston gives up 15 open threes again…well, they better hope for good 3PT% luck.

When Houston has the ball

Let’s get chaotic. Houston thrives when the game turns into a volleyball match that’s somehow mixed with tackle football. If this game becomes determined by which team is tougher and better on the boards and it’s Houston, I think they’ve got the edge. They’re likely to have the edge when it comes to turnovers. The problem for Houston is the same that it usually is: if they can’t own shot volume, their paths slim out pretty fast. Houston went 7-4 when they had a combined rebound/turnover differential of +2 or lower this year, which isn’t bad, but an average of 70.1 PPG that required three overtime games to even get that high is pretty ugly.

But we’re focusing here on when the ball is in their hands, not as much as when it’s a free-for-all. Houston’s half-court offense revolves heavily around what Milos Uzan can and will do in their ball screen sets, and I think hay can reasonably be made here. The Duke defense is gigantic, the literal tallest team in the nation, but deterring Houston away from the rim is not really tough to do to begin with. They generate the fewest rim attempts in the sport and love the midrange, which can actually play in their favor in the Moreyball era if they’re hitting from their spots.

Teams have generally looked to force Uzan into being a shooter when he’s had the ball in his hands as opposed to letting him create for others, which frankly has not gone well for the opponents. Uzan has turned into one of the very best tough-shot hitters in the sport this year out of nowhere, shooting 41% on midrange jumpers, 46% on runners, and 44% from three, around half of which were pull-up threes. Duke has defended basically everything in the P&R realm well this year, alternating between a drop and a hard hedge, but their usual path of forcing a lot of runners or pull-up jumpers is probably a bad idea against Uzan in particular.

When Uzan has to get the ball out of his hands, he’s been responsible for 79 of the team’s 224 assisted threes, the lion’s share of which have gone to LJ Cryer and Emanuel Sharp. His drives draw in a lot of gravity…usually. It’s worth noting that Houston has had one of the lowest kickout rates of any team in America this year, which plays into Duke being a team that prefers to let the ball-handler do the talking as opposed to overcommitting and creating open threes on the perimeter.

Aside from that, I think you know where else this typically goes for Houston: the post. The Cougars have used post-ups at a high rate this year, with J’Wan Roberts being their go-to down low for seemingly nine years now. Roberts is not actually all that efficient, but the gravity he creates (and the numerous fouls he draws) has allowed him to create the second-most assisted rim attempts behind Uzan and third-most assisted threes.

Duke has normally not doubled all that often on post-ups, but if Roberts hits a couple early or gets Maluach in foul trouble, they may not have a choice. Roberts is abnormally good at hunting the open man, and it could certainly lead to better/more looks for the three main reasons why Houston is here and not at home: Cryer, Sharp, and Uzan.

For all the good Duke has done this year, Synergy has them as allowing 1.14 PPP on passes out of the post, which is bizarrely poor. I’d chalk at least some of that up to sample size, but it might be a path for Houston to create more/better shots.

I almost see this as a pretty simple game to decipher. Based on the mass amounts of jumpers forced each way, on an average day, I favor Duke to hit a few more than I do Houston. But the Cougars are simply tremendous at both generating and hitting their jumper attempts, and this feels like a titanic battle between the two best teams of this excellent season. If this is a day where Houston can hit 10+ threes, they’re in business. If this isn’t, well, there’s no shame in a Final Four run.

Thoughts, predictions, etc.

THING TO CTRL+F THE BOXSCORE FOR: The rebound margin. We make fun of this stat because it is really, really simple and gets way more importance in broadcasts or team analysis than it should. Yet if you’re just looking at ESPN or StatBroadcast and don’t want to do the hard work, it’s been quite predictive for each team. All four of Houston’s losses have seen them go +2 or lower in rebound margin. Duke’s three losses: -13, -6, and +1 in rebound margin. The boards are going to be a huge factor.

THING TO LOOK AT THE SCOREBOARD FOR: Houston’s threes. Simple as that. They went 24-1 this year when hitting 7 or more. I maintain they need no fewer than 10 in this game.

X-FACTOR, DUKE: It’s Maluach for me. Duke’s 16-1 this year when he scores 10+, but far more important to me is that Duke has averaged 89.3 PPG in those games. I highly doubt Duke is touching 89 on Houston, but if they are, Maluach has put together a top-5 pick performance.

X-FACTOR, HOUSTON: JoJo Tugler means a ton to this game. If he effectively stunts post-ups, drives to the rim, and lobs in this game, Houston not only has a real chance, they’d be favored to win. Tugler is the best defender left in the Tournament, and it feels like a must for Houston to get 28+ minutes out of him here (he averages 21).

THE STRUCTURAL REQUIREMENTS OF SPORTSWRITING FORCE ME TO DECLARE: That Duke has a fundamentally higher ceiling but Houston might have the higher floor. When Duke explodes, they explode in a way no other team in the nation can or has in many years. This is the best offense I’ve seen in my lifetime as a college basketball fan, and to pair it with a top-five defense feels unfair. They’re favored here for a reason.

But Houston can win this game. Of course they can. This is the best and toughest team Duke has drawn all year, one who could give Duke a taste of something they really haven’t been given outside of the Clemson game: a pounding on the boards. If Houston is getting back their misses and restricting Duke on the other end, they can turn this game into the exact kind of sludge Houston thrives in. A Houston win here almost certainly requires holding Duke to 65 points or less, and as extreme as that sounds, a game played at 60 possessions makes that possible.

Moreso than game one, jump shooting variance is going to be a major factor here. Houston doesn’t generate much of anything truly at the rim; Duke’s 3PA/FGA has been over 50% against the only Houston-like defense they’ve played and nearly 50% against the actual two toughest. This game feels like it’ll be close but it being a game of two-way runs also feels appropriate if jump shooting in a football stadium takes a little time to get used to.

This is the very best transition defense (and overall defense) in the sport against the most lethal offense college basketball has seen in a long, long time. Hopefully it lives up to the billing. But most likely, I think it lives up to the expected outcome.

PREDICTION: Duke 70, Houston 66. I think Houston hangs and hangs and hangs but can’t find either the stop or the shot to get them over the top. That sets up Duke/Auburn - a game with the two best players in the sport - on Monday night.