Duke vs. Texas Tech: What to make of Mark Williams and Trevor Keels?

When the NBA playoffs get going, and various teams look to make a championship push, it all boils down to matchups. (Health is a super important factor, too.) Having superstar talent atop the roster is a franchise’s ticket to the title dance; however, the ways in which a team is able to mold, shape-shift and counterpunch become crucial on a round-by-round basis.

The NCAA Tournament isn’t the same beast. It’s a single-elimination event, not a best-of-seven series. This leads to plenty of chaos and randomness, which can produce great theater. A 15-seed can defeat a 2-seed in a one-off. That upset outcome is less likely in a best-of series.

Despite the mayhem of March and April, matchups and tactics are pivotal. To survive and advance requires some weird combination of luck (shooting and matchups), talent and game-to-game adjustments.

During Duke’s Second Round win over Michigan State, Mike Krzyzewski’s team found ways to problem solve on the fly. With five minutes left to play, the Blue Devils trailed by five points; it would’ve been easy to fold. Instead, Duke unleashed some of its best basketball this season, which included a 20-to-6 run, and pushed through to the Sweet 16.

Now in San Francisco, Duke will meet a defensive bully: The Texas Tech Red Raiders. This matchup offers up a great deal of intrigue, including some good and some bad for the Blue Devils. Let’s dive in on the game’s most critical battlegrounds, including the postseason role of freshman guard Trevor Keels.


Texas Tech: No Middle, Man

Much like Virginia, Texas Tech is one of the premier defensive programs in the country, albeit with differing styles. The switch flipped when Chris Beard and Mark Adams arrived in Lubbock ahead of the 2016-17 season. That initial season, Texas Tech was a borderline top-50 defense, which is pretty good. Since then, though, the Red Raiders have posted a top-20 defense in four straight seasons, including three campaigns inside the top five.

Obviously, the trend continued even after Beard left for Texas in 2021. Adams, Beard’s former top assistant and the coordinator of Tech’s defense, has kept things humming. Currently, the Red Raiders rank No. 1 nationally in defensive efficiency.

Texas Tech brings a hard-nosed approach to defense, which is mixed with an elite scheme, a versatile roster and high-level execution. Adams runs a “no-middle” defense, which at face value isn’t that unique. There are a lot of college basketball teams that run no-middle in an effort to keep the ball on one side of the floor. However, Texas Tech’s commitment to that no-middle approach is extreme.

According to KenPom, opposing offenses average 17.8 seconds per possession vs. Texas Tech, which ranks No. 234 nationally. Opponents must run their offenses later into the shot clock, which can create all types of issues and inefficient play.

Secondly, Texas Tech has forced a turnover on 23.5 percent of its defensive possessions this season, which ranks No. 11 nationally. This marks the fifth straight season that TTU will post a defensive turnover rate above 21.5 percent.

Unsurprisingly, Texas Tech has a high steal rate: 11.9 percent (No. 26).

By aggressively forcing teams to one side of the floor — away from the middle — the Red Raiders greatly limit rim pressure. According to Bart Tovrik’s shot data, only 25.8 percent of opponent field goals attempts vs. Texas Tech’s defense are close 2-point attempts. This ranks No. 3 nationally.

Opponents are shooting just 53 percent on those 2-point shots around the basket, too, which is a top 30 number in the country. As a result, opponents of TTU have shot a measly 44 percent on 2-point attempts — with an effective shooting rate of 45.4 percent, both of which rank top 20 in Division I.


Flip The Switch: Roach and Wendell Moore Jr.

Texas Tech’s defensive numbers don’t come in a vacuum, though. That type of output requires scheme execution and strong defensive personnel.

To start off, Texas Tech switches a lot, 1-5. All five players on the floor must be willing and able to cover every assignment, 1-on-1. That includes big men Bryson Williams, Kevin Obanor and Marcus Santos-Silva, who are comfortable guarding in space vs. speedy ball handlers, like Notre Dame’s Blake Wesley.

Santos-Silva is in his fifth season of college basketball; he’s been a shot-blocking force at both VCU and now Texas Tech. The 6-foot-7 center is one of six high-major players with 7.0 percent block rate and 2.0 percent steal rate — along with Walker Kessler and Jesse Edwards.

Williams is an excellent two-way player for Texas Tech as well. He’s much improved from the start of the season, and he’s capable of switching across a variety of position types.

This will put Duke’s primary ball handlers — Jeremy Roach, Wendell Moore Jr. and Trevor Keels — in situations where they’ll get to go at an opposing frontcourt player.

In recent weeks, Roach has taken his play to another level. Over the last 10 games, Roach has averaged 10.8 points, 2.1 assists and 2.0 rebounds per game. During this stretch the sophomore guard has shot 52.1 percent from the floor, 40.0 percent on 3-point attempts (3.5 3PA per game) and 83.3 percent from the line.

This has come with 20.4 percent usage (22.5 percent usage over the last five games) and a true shooting rate of 65.5 percent. Roach is using more possessions than ever before and playing the most efficient basketball of his college career.

One of the big question marks with Roach before the season centered on his ability to get downhill and pressure the rim. That’s been an up-and-down quest for much of the season, but recently Roach has shown real pop while driving the basketball.

Going back to the loss vs. North Carolina, Roach is 11-of-18 on attempts around the rim (61.1 FG%). That may not seem like much, but it’s efficient finishing. Plus, those drives to the rim have a cascading effect. Rim attacks force defensive rotations and create opening for other offensive players.

Roach drove the ball with determination and craft vs. Michigan State, especially as Duke repeatedly attacked Max Christie in space.

That initial switch is only part of the recipe for Texas Tech, though. It can be an absolute bear to create driving lanes against this defense.

Once the offense moves the ball to one side of the floor, the on-ball defender will angle his body to funnel the ball handler away from the middle — towards the baseline and an army of help defenders.

Texas Tech floods the lane and has a help defender meet the ball handler at the edge of the paint. Essentially, four defenders are on one side of the rim, while one back-side help defender sinks in, closer to the opposite side of the lane.

From the Notre Dame game: When Prentiss Hubb drives to his left vs. Arms, all four off-ball defenders swarm and shrink the paint. Texas Tech’s defense walls up perfectly, too. With Kevin Obanor showing help for Arms, Williams has Paul Atkinson covered in the dunker spot and McCullar makes sure that Dane Goodwin won’t receive an easy kick-out 3. Shannon is in free safety mode on the back side.

Cormac Ryan does a nice job of cutting through the defense, but it’s too late and Hubb doesn’t see the cut.

The goal of the switch and exhaustive no-middle approach is to take away the base of an opposing offense: sets, actions and reads. Stuff that an offense is used to gets is wiped off the floor.

These measures disrupt the flow of an offense, which can lead to turnovers or a variety of inefficient, late-clock shots at the basket.


Skip It Good

Given how Texas Tech sells out on no-middle, there are other opportunities for an offense to pursue. Basically: can you put Texas Tech’s defense into rotation and generate quality 3-point looks?

Currently, Texas Tech ranks No. 344 nationally in opponent 3-point attempt rate: 45.7 percent of opponent field goal attempts come from beyond the arc. That’s the highest rate for any TTU team coached by either Beard or Adams.

If the offense can skip the basketball — flip sides of the floor — and play quickly while TTU’s defense rotates around, then it has a chance to create 3-point shots or get into gaps.

Two weeks ago, Kansas did a nice job attacking early in the possession, not letting Texas Tech get completely comfortable.

(I also thought the Jayhawks did a nice job of utilizing dummy motion on offense — weave action, handoffs, etc. — before launching quickly into their sets.)

Ball handlers can take what the defense gives them — the baseline — and work off that space. By driving baseline and then snaking back to the middle, Trey Wertz (with Williams switched on him) gets a piece of the paint and creates an open 3 for Ryan.

The help rotations are coming, though. Texas Tech will close the window in a heartbeat.

Notre Dame’s offense was held to under 0.85 points per possessions, while shooting 8-of-24 on 2-point attempts and 9-of-28 on 3-point attempts.

Overall, opponents shot just 31.4 percent from downtown this season vs. Texas Tech (No. 62 nationally) The Red Raiders closeout hard and contest everything.

Not all 3-point shots are created equal. Given how TTU forces teams to get stuck on one side of the floor and run offense late into the clock, they produce plenty of panicked shots — ones that come just before the clock expires or off-dribble heaves against a contest.

Regardless, it behooves Duke to deploy AJ Griffin as a critical floor spacer.

Griffin, who is shooting 45.5 percent from downtown, has an effective shooting rate of 61.5 percent on spot-up attempts.

The shooting gravity of Griffin could widen Texas Tech’s defense. If that happens, Duke can look to cut into pockets of space for finishes. It’s important to cut with purpose vs. Texas Tech, especially when the Red Raiders show help at the nail or double in the post.

Griffin, Moore and Banchero are excellent cutters. Moore moves aggressively without the ball. (This is an under-discussed aspect of his game, in my opinion.)

When things bog down with Duke’s half-court offense, Moore’s off-ball movements can help grease the wheels and encourage additional player/ball movement.


Trevor Keels: Is he Duke’s X-Factor?

As Duke looks to mount one last run at a national championship under Coach K, you pretty much know what to expect with this team. Over the last four months of basketball, Duke’s key guys have cultivated set roles.

Banchero is an incredible a hub of offense and the team’s No. 1 option. Griffin is a lower-usage floor-spacer who amplifies Duke’s attack with his range shooting. Roach is a steady ball-handler who gives the Blue Devils some zip in the half court. Williams is the pressure point at the rim — on both sides of the floor. Finally, Moore fills in all of the gaps in between.

This leaves Trevor Keels as a bit of a wildcard. His role on defense is defined clearly: heat up the basketball (2.4 percent steal rate) and cause problems at the point of attack.

Throughout this season, Duke’s offense worked with a multi-pronged approach to half-court playmaking. Duke is the only ACC team this season with four players who have 17.0 percent usage and 15.0 percent assist rates: Moore, Keels, Banchero and Roach.

This isn’t a heliocentric offense — with one player working as the offensive kickstarter. On a night-to-night basis, the Blue Devils can alternate which player will lead the dance. In general, that can be a strength. The more playmakers on the floor at the same time, the more dynamic an offense can be, assuming there’s synergy and those guys also do stuff without the ball that adds value.

To that end: as Roach takes on more ownership of Duke’s on-ball creation, Keels slides to (slightly) more of a tertiary role.

Now, there are still plenty of on-ball reps for Keels; the Blue Devils aren’t going to stop running slot pick-and-roll for their power guard. They need his downhill drive game. (Once again, there goes a Duke guard blowing by Christie.)

However, this presents an opportunity for Keels to display what he can do without the ball, too. That matters for Duke (now) and his outlook as a prospect (future).

At times this season, Keels has shown he’s more willing to move without the ball; he’ll relocate for 3-pointers or cut along the back side. Duke has even leveraged Keels’ power and used him to post-up smaller guards.

One of his best games this season came in Duke’s win at Clemson.

During that contest, Keels was active without the ball. He cut with a purpose and got on the offensive glass. As a result, Keels scored 25 points (9-of-13 FGA, 6 FTA), which at the time tied a career high.

Here, Duke comes out in its stacked Floppy action. Keels throws the entry pass to Theo John and instantly cuts off a flare screen from Banchero. Chase Hunter is right there for Clemson, but Keels has a step and finishes deftly with his inside hand — thanks to a nice pass from John.

That effort from Keels raced into my mind during the second half of the Michigan State game, when Keels found life playing off the gravity of Banchero.

On this possession, Duke starts in a Horns alignment and runs 4-5 pick-and-roll with Banchero and Williams, something its done dozens of times this season. As Michigan State loads up on Banchero — showing early help at each elbow — the possessions starts to grind down. Banchero is a brilliant playmaker, but he needs movement to unlock his passing, which is exactly what he receives from Keels.

Keels and Roach perform an off-ball exchange, which Michigan State switches. As Banchero drives left, though, Tyson Walker (now on Keels) shows help. Keels wisely senses this opening — with Walker occupied — and darts to the rim for two points.

Later in the game, as Duke started to summon its comeback, Keels once again made an important off-ball play. Out of a timeout: the Blue Devils went to their 21 pick-and-roll action: Roach hands off to Moore, who quickly dribbles off a screen from Williams. As the Spartan cover it up, the ball changes sides and winds up in the hands of Banchero on the block. Banchero goes to work on Joey Hauser; he draws a second defender (AJ Hoggard), the man who was guarding Keels. Keels threw the entry pass to Banchero and never stopped moving, though. He slides left into an open window of space for a P5 kick-out pass.

Once Keels catches the ball, it’s too late for Hoggard. Keels isn’t where he last saw him and there’s too much ground to cover on the closeout.

Duke will need this constant movement from Keels vs. Texas Tech.

Another way for Keels to “star in his role” comes on the defensive side of the floor. Texas Tech is a team that’s stocked at the wing position; however, there’s no pure point guard on the roster, save for maybe Mylik Wilson.

The “lead guard by committee” approach has mostly worked. With some combination of Arms, McCullar, Shannon and Davion Warren running the show, Adams and Texas Tech have created a top-50 offense this season. The Red Raiders pound the offensive glass (33.2 percent offensive rebound rate) and get to the line (35.8 percent FTA rate).

With that said, defenses can speed this offense up and create turnovers.

Texas Tech’s offense turned the ball over on 20 percent of its possessions this season: No. 282 nationally in turnover rate. This presents a major opportunity for Keels. Any live-ball turnover forced will be worth its weight in gold: get in transition and try to score before Tech can set its defense. (This is another area where Moore excels. He’s a terrific link-up passer when Duke transitions from defense to offense.)

The Blue Devils could look to turn Keels loose at the point of attack in this game. Keels can absolutely be a wrecker on the basketball, but he doesn’t have the consistent impact his profile seems to offer.

Well, there’s never been a better time to deliver.


Mark Williams / Ruby Gobert Factor

The flip side of Texas Tech switching 1-5 pick-and-rolls is that it will place a guard defender on the opposing center. In this case, that’s All-ACC big man Mark Williams.

This raises the question: What will Mark Williams be able to do vs. Tech’s switch?

Williams can be a dominant force. The 7-footer is shooting 76.8 percent at the basket. He also has 87 dunks to his name, which is the second most among all Division I players.

Going back to the 2007-08 season, only two other ACC players have recorded 85+ dunks in one season: Duke’s Marvin Bagley III and Doral Moore of Wake Forest.

Williams works hard as a dive man in Duke’s pick-and-roll action. He’s also opportunistic as a play-finisher while lurking around in the dunker spot — floating along the baseline, just outside of the lane.

This is how Williams generates the majority of his interior finishes: rolls to the rim, cuts or put-back attempts. According to Synergy Sports, Williams shot 74.7 percent on non-post-up rim field goal attempts this season. That ranks No. 4 nationally among players with 100+ attempts.

Those types of looks, however, come when another player creates an advantage, forces the defense to help/rotate and finds Williams at the rim for an easy 2.

Williams has expanded his game as a post-up target (54.8 FG% on post-ups), although that’s not what he does best.

Duke will throw it into Williams in the post, but his optimal activity with his back to the basket comes off quick maneuvers: seal, catch and go.

Despite his size, Williams can get knocked off his spots while finishing through contact in the post — even if he has the touch to complete those plays.

What Williams does less of is punish switches in the post. That’s just not really his game, which is fine; Williams has serious upside as an archetypal modern center. He will roll to the basket, protect the rim and work as a drop pick-and-roll defender.

However, if Williams catches a mismatch in the post vs. Texas Tech, will he look to duck-in or call for the ball? When TTU fronts the post on the switch, can Williams carve out enough space with a seal for a lob finish?

Of course, the Red Raiders assuage some of their own concerns by having a deep rotation that’s almost entirely comprised of big wings or versatile hybrid forwards/centers.

Texas Tech’s bench accounts for 36.8 percent of the team’s overall minutes this season, per KenPom, which ranks 58th nationally. No one in the rotation is under 6-foot-3 and only one player is above 6-foot-8: freshman center Daniel Batcho, who stands 6-foot-11. This means everyone else in TTU’s rotation is between 6-foot-3 and and 6-foot-8.

That makes this roster ideal for switching: the plethora of strong, like-sized players allow for seamless switches and nullify some of the more glaring mismatches. In this discrete sense, Texas Tech feels like Florida State: switch schemes with wing-heavy rosters.

Terrence Shannon Jr., Adonis Arms and Kevin McCullar are all 6-foot-5 or 6-foot-6 and hover around 210 pounds. Shannon is already NBA-caliber defender. There are few guys on the roster to poke at in a switch, especially this trio.

When Texas Tech gets a wing switched on an opposing big man, the Red Raiders will front the post and flood in back-side help. Tech rotates on a string and help defenders come packing a punch. These efforts can even offset designed quick-hitter hi-lo sets.

Again, those post-ups need to come quick, before the help arrives. This defense will crash hard in an effort to make life miserable at the rim.

If the post touches comes further away from the basket, the Red Raiders will look to double the ball, too. This can create kick-out opportunities for spot-up looks.

He won’t be confused for Nikola Jokic anytime soon, but Williams has shown some passing ability — simple stuff, like kicking to weak-side shooter from the post.

However, Williams can also have trouble in passing situations when you show him a double team.

One of the others ways Williams can take advantage of his size vs. the switch is on the offensive glass.

Williams is incredibly long (7-foot-7 wingspan) and a quick leaper off the floor. He can be especially explosive with his second jump.

Through the first 36 games of the season, Williams posted an offensive rebound rate of 13.6 percent, a top-50 number in the country. Williams shot just under 58 percent on put-back attempts, per Synergy.

This is an area where Texas Tech is slightly vulnerable, too. The Red Raider rank 84th nationally in defensive rebound rate: grabbing 76 percent of opponent missed field goals. That’s pretty darn good, but it’s not an elite number and a tradeoff that comes with so much switching.

If a 6-foot-3 guy is trying to box-out Williams, Duke’s center should be able to play over the top and create second-chance opportunities.

Williams doesn’t even need to turn every contested rebound opportunity into a ball he catches. Similar to former NBA center Tyson Chandler, who played for Coach K with Team USA, Williams has a knack for back-tapping loose ball in offensive rebound situations.

These efforts help extend possessions and can jolt the Blue Devils into short-space advantage situations vs. a scrambled defense.


What does Williams do defensively?

According to CBB Analytics, only 22.2 percent of opponent field goal attempts vs. Duke this season have come at the rim. Opponents are shooting 63 percent on those attempts, which is below the national average (64.8 FG%).

While he has a tendency to over-help in the pursuit of blocking shots, Williams is an incredible deterrent at the basket: 5.0 blocks per 40 minutes.

During the Michigan State game, Williams was brilliant as a drop pick-and-roll defender.

The 7-footer puts a lid on the rim and pushes opposing attacks to take tougher, contested shots from less efficient zones on the floor. Per CBB Analytics, 28.5 percent of opponent field goal attempts come in the paint but outside of the restricted area — above the national average of 23.6 percent. Opponent shoot a chilly 39.5 percent on these attempts.

As impactful as he is on defense, Williams comes with some limitations, too. He can be put in precarious situations or played off the floor at times when an opponent has the ability to play 5-out offense.

When opponents run pick-and-roll action, Williams wants to sink into the paint and close off the rim. However, that’s more of a challenge when the opposing center can pick-and-pop.

This created problems for Duke in both matchups with Miami this season.

When Duke and Miami met in the semifinals of the ACC Tournament, the Hurricanes scored 1.13 points per possession. It was trouble for the Blue Devils when they were put in rotation guarding pick-and-pop actions.

Eventually, Duke started to switch Williams on the screen actions — asking the 7-footer to guard in space against smaller ball handlers.

Williams is agile enough to get this done in a pinch, but it takes Duke out of its base defense, which can cause lots of problems.

Unfortunately for Duke, Texas Tech offers similar issues.

6-foot-8 Bryson Williams gives up several inches to Mark Williams; however, he’s a very good pick-and-pop player. Williams averages 2.6 3-point attempts per game, and he’s shooting 40.9 percent from beyond the arc.

Duke must figure out what it wants to do with this action. Dropping Williams may not be a great option. Putting two on the ball will still leave the pop man open for a 3 — even with a weak-side stunt.

Not only that, but Williams also has the ability to attack a closeout.

As Notre Dame “ices” this ball screen, Nate Laszewski drops into the paint. Hubb does a nice job with a hard stunt (Arms doesn’t cut), which buys Laszewski time to recover. Williams is ready and willing to attack that advantage, though.

If you Ice those looks and give Williams the middle in space, Texas Tech will also use this to flow into a dribble handoff and attack a rotating defense.

Similar to the Miami game in Brooklyn, Duke could look to switch Williams out on these screening action. Texas Tech has good guards, but they aren’t nearly as skilled or shifty as Isaiah Wong, Kam McGusty and Charlie Moore. Duke’s staff may feel more comfortable switching Williams on perimeter guys, if needed.

Theo John could have a role, too. He allows Duke to keep a true center on the floor — one that can protect the rim — and switch with a little more vigor.

Williams isn’t the only TTU frontcourt player that adds some stretch to the frontcourt. So, too, does Oral Roberts transfer Kevin Obanor: 46-of-139 from deep (33.1 3P%).

That adds another layer to the offense, which can be tough for opposing defenses, especially when Tech mixes in roll-replace action.

Obanor and Williams can dive to the rim or pop into space. Both guys are also comfortable handling the ball or posting up against a smaller defender. Adams does a nice job leveraging these components and playing them off one another — moving players into empty space and causing confusion.

Santos-Silva and Batcho, the other two big guys in the rotation, don’t offer much in terms of stretch: a combined 0-of-2 on 3-point attempts. However, they’re frisky rim finishers, both shooting above 70 percent around the basket this season.


Small Ball: What’s the call?

One of the other cards Duke can play vs. 5-out/pick-and-pop action is the team’s small-ball alignment. When Griffin emerged this season, Duke unlocked a killer small-ball contingent with Banchero at center. Griffin is the de facto 4 in these lineups, along with Keels, Roach and Moore, primarily.

At times this season, these configurations have been awesome. With Williams and John off the floor, Banchero can attack the rim with no one else in the paint or at the dunker spot. That added spacing is key.

According to Pivot Analysis, Duke is +46 in 169 minutes with Banchero on the floor and Williams and John on the bench. The Blue Devils have scored 120 points per 100 possessions with these lineups on the floor.

With the lineup of Roach, Keels, Moore, Griffin and Banchero on the floor together, Duke is +32 in 80 minutes, scoring 1.29 points per possession, according to Pivot Analysis.

It’s a nice tool to have, and it’s definitely helped swing some games for the Blue Devils. (The December win over Virginia Tech really stands out in this regard.) However, it’s not exactly a panacea, either.

The Blue Devils have lost just three games since the start of February. In two of those more recent defeats — North Carolina and Virginia Tech — Banchero at the 5 lineups haven’t hit, mostly on the defensive side of the court.

During the loss to VT in the ACC title game, Duke played 16 minutes with Banchero as the one true big on the floor. The Blue Devils were outscored by five points in this stretch and allowed 1.11 points per possession.

One week prior, when UNC started to carve up Duke’s pick-and-roll defense with Williams on the floor, Krzyzewski flipped the switch to his small-ball lineup. Once again, this group flopped as Duke melted down: Blue Devils were -9 in five minutes with this group on the floor (1.85 points per possession allowed on defense.)

In theory, these lineups allow Duke to switch more easily 1-5 with Banchero at center. Banchero is a smooth mover, who can guard laterally in space; however, RJ Davis of North Carolina gave him issues.

Duke has yet to go back to these lineups in the NCAA Tournament, although neither matchup in the first two round necessitated the move. Texas Tech could force Duke’s hand in the Sweet 16. If that happens, there will be a tradeoff.

On one hand, Duke would be able to coax Tech’s defense into switch smaller players on Banchero, which could lead to good half-court offense.

Duke’s wings must be ready to meet TTU’s physicality on the glass and in the paint, though, Griffin is strong as hell, but his post defense has been inconsistent this season.


Read More on Trevor Keels and the NCAA Tournament

Inside The Playbook: How Miami shutdown future Top-5 NBA Draft pick Jabari Smith