Thoughts on CCSU: A Winless Non-Conference Season
Well folks, following Friday's 87-58 loss to Penn State, Central Connecticut has officially finished winless during the non-conference portion of its schedule. (Note: I will always pretend games against non-D1 teams do not actually exist)
The first two months of the season have been incredibly trying for everyone involved; players, coaches, and fans. Save for having a very young team, there really haven't been a ton of positives for this bunch, and the outlook for NEC play is bleak. The truth is, there aren't many weak teams in the Northeast Conference who will be ripe for the plucking.
I will say, however, that this year's team appears to be working hard, especially on defense. There are a bunch of good athletes who haven't given up despite the struggles. And they're improving, albeit slowly; according to Bart Torvik's Game Scores, the last two games were their third and fourth best performances of the season. But that doesn't tell the complete story; at Dartmouth, the game was tied with about 13 minutes left to play, and the Blue Devils essentially played Penn State even over the final 15 minutes. Little victories.
Anyhow, I have a bunch of thoughts as we head towards Christmas. NEC play starts next week!
1. The thing that jumps out at me every time I watch the Blue Devils play is the awful shot selection. If you've read anything I've written before (or follow me on Twitter), you know I can't stand long, two-point jump shots, and I've been critical of the Blue Devil teams in the past for that. But this season it's not even where the shots are being taken, it's who is taking them and when they come in the shot clock (as well as who else may be open).
The folks over at Dribble Handoff developed a proprietary metric called ShotQ that quantifies a team's offensive and defensive shot quality. So how does CCSU stack up in this metric? They currently rank 352nd (out of 353 D1 teams) in shot quality (and rank 342nd in eFG%). So, my eyes are not deceiving me; this team has a bunch of good shooters taking bad shots.
I could live with Tyler Kohl taking ill-advised jumpers last season, and I'm generally fine with Jamir Coleman taking them. And if I were fans of other NEC programs, I'd be more than happy to let guys like Adam Grant, Keith Braxton, Raiquan Clark, and Jahlil Jenkins take questionable shots from time-to-time. Sometimes when you have a great scorer, you have to let them do their thing. But this is not that. I mean, look at some of these shots:
Here Xavier Wilson, who has been one of the biggest offenders when it comes to questionable shot attempts, gets the ball in the post, with Myles Baker (one of CCSU's best shooters) wide open on the left wing. Maine ultimately double-teams Xavier, and Wilson fades away to his right with two defenders right in his face, with 21 seconds remaining on the shot-clock. That can't be more than a 15-20% shot, and should only be acceptable with the shot-clock winding down.
Stephane Ayangma has a tendency to take questionable shots early in the shot-clock. In the first screen-grab, there's 17 seconds on the shot-clock and he's taking a well-defended 16-footer. In the second, with 11 seconds on the shot-clock, he's taking a fade-away off his right foot with a taller defender in his face. Neither shot could have an eFG% better than 20%-25%.
Jamir Reed, who should have the green-light as much as anyone at this point, likes off-the-dribble two-point jumpers (similar to Ian Krishnan last season). It's not that these are bad shots, per se (he made the 2nd one), but he's passing up better shots by not starting that move a few feet further back. The expected point total (i.e. eFG%) on a 21-foot three-pointer and an 18-foot two-pointer aren't even close.
As I said earlier, I'm fine with Jamir Coleman taking mid-rangers (at least to a point). But he has a tendency to pass-up open three-pointers for more difficult two-pointers. Like here:
Coleman probably makes a step-in three-pointer at the top of the key, what, 35% of the time (conservatively)? That's an expected point total of 1.05 points. Instead, he takes this off-balance, well-defended runner from about 12-feet. That fact that he made the shot means little to me; over the long-run, the team scores more points if he takes those threes than if he passes them up (unless they're for layups).
I didn't capture screen shots of any, but Greg Outlaw has a penchant for taking low-percentage runners, and Zach Newkirk loves himself a mid-range jumper (he was 2-4 in the loss to Penn State, but is 5-14 on the season).
At this point I have to wonder out loud; is the CCSU staff generally fine with these shot attempts, or are the players simply not listening to their coaches? These shots are essentially turnovers, and are exactly what the defense wants.
Ultimately, I feel comfortable saying that CCSU will never have anything close to an efficient offense (this season or in the future) if the shot selection doesn't see a major improvement. This isn't a situation where Central is running good offensive sets, then just missing open shots.
2. Along the same lines as the offensive shot selection, it seems CCSU gives up an inordinate amount of three-pointers. While watching the Dartmouth game and getting frustrated at the number of open looks the Big Green was getting from the perimeter, I did some research. Since Donyell Marshall took over prior to the 2016-17 season, CCSU has allowed 902 made three-pointers (in 2,464 attempts), while they've only made 609 (in 1,830 attempts). That's a delta of 879 points!
How does that compare to other NEC programs? Not well. Central has both made and attempted the fewest three-pointers in the Northeast Conference since the beginning of the 2016-17 season (Bryant is the next closest with 626 makes in 1,882 attempts). However, they've also allowed the 9th most three-pointers (out of 10 teams), as Saint Francis has allowed 913 in the last 3+ seasons.
Putting that all together, the 293 three-point difference between their makes and their opponents makes is easily the widest in the Northeast Conference during that period. Sacred Heart is the next closest with a negative 109 three-point differential, and Fairleigh Dickinson is third at negative 69.
I think what makes this worse is that CCSU has actually shot the ball fairly well over this period (33.3%, which is 6th best), but they've allowed opponents to shoot a whopping 36.6% from three, which again is worst in the league.
Why is this the case? I hesitate to suggest that the CCSU staff does not properly value the three-pointer (on either side of the ball), but what else could explain this large of a gap? Again using the Dribble Handoff data, CCSU's defensive shot quality ranks 345th nationally (322nd in opponents eFG%) this season. To me, the biggest issue is how the Central Connecticut defense collapses the paint anytime an opponent even threatens to penetrate. Look at these examples:
The man in the corner who ultimately makes a three-pointer is senior Ian Sistare, who shot 43.4% from three last season (just 23% this year), while three players try to defend the paint.
In the next screen-shot, #1 in the left corner, Taurus Samuels, is a career 30% three-point shooter and four defenders are collapsing on the driver. A 30% shooter wide-open in the corner makes him, what, a 50% shooter?
The fact that they seem content with taking mid-range jumpers while allowing their opponents to shoot open three-pointers at this magnitude is very concerning.
3. On a positive note, CCSU has done a much better job at limiting turnovers. Check out this graph (courtesy of Bart Torvik):
In their first four games, the Blue Devils turned it over on 28.5% of their possessions. Over their past 7 games (prior to the Penn State loss), however, that number is just 18.5%, and just 16.3% against mid-majors (excluding the loss to Boston College).
I'd suggest that part of it has to do with slowing the tempo; those first 4 games averaged 76 possessions, while the last 7 games averaged just 70 possessions.
There are obviously problems scoring the ball, but it appears the turnover issue have at least temporarily been curbed (a 16.3% turnover rate would have been the best in the NEC last season). Progress!
(Editor's note: I put this together prior to the Penn State game, but the fact that they only turned it over 20.6% of possessions against a top 20ish Big Ten game further supports the improvement).
4. I'm not sure what it is, but it's been very difficult for CCSU to find even a league average point guard since Malcolm McMillan got injured two games into the 2015 season, then transferred to Canisius for his Redshirt Senior campaign. I mean, look at these numbers:
Note: BPM stands for "Box Plus/Minus", which is an advanced stat intended to measure a player's total contribution. Negative 2.0 is considered "replacement level".
Perhaps it's the "curse of Kyle Vinales".
It's really not close. I've said it before and I'll say it again; I truly hate being critical of individual players, as I recognize how hard it is to be a Division 1 athlete, and even the worst players on NEC rosters are really good freaking players. With that said, sophomore Zach Newkirk is last in every metric that one would use to measure point guard play, and only Bryant and Long Island currently have a below-replacement level '1', though both of Bryant's guys (Green and Ndugba) have flashed major potential this season, while Agosto has the ability to put up points in a hurry (21 against North Florida, 15 against Delaware St., 14 at George Mason).
Newkirk is currently playing 26.1 mpg, and I'm just not sure how long you can give significant playing time to a guy with a 40.7% eFG% and 0.8 A/TO ratio (and just 7-12 from the free throw line). In my last "Thoughts" column, I opined that I would start playing Segwai more, as I think he's been better overall (both provide solid defense, neither provides perimeter shooting, but Segwai seems more aggressive in attacking the paint), and while he's played 34 minutes over the past three games, it's still significantly less than Newkirk. Sure, Newkirk probably had his best game Friday night, as he looked more comfortable getting into the paint via the dribble and knocked down a couple of mid-range jumpers. But we need to see more of that.
Also, Marshall seems to like playing Segwai and Newkirk together, and I'm not quite sure why; I'm going to assume it's because they're both good defensively, but when they are on the floor at the same time it allows the defense to sag the middle and dare one of them to make an outside shot (Newkirk hasn't made a three-pointer in over a month, and Segwai hasn't made one all season).
If I were Marshall, who obviously knows his personnel and the gave of basketball significantly better than I do, I would roll with a combination of Myles Baker, Jamir Reed, and Greg Outlaw as my primary ball-handlers for a large portion of the game. No, none of those guys are likely to run an offense as well as Newkirk can, and it could result in increased turnovers, but the trade-off of putting more dynamic offensive players on the floor at the same time is well worth it, at least in my opinion.
5. Compounding the point guard issue is the fact that this Central Connecticut team does not have a player who can consistently create offense, which is largely the name of the game at this level. Give it to your best playmaker (like Braxton, Grant, Jenkins, Parker, Raiquan Clark) and have them get a bucket (for themselves or someone else) is a pretty sound offensive game-plan when all else fails.
If you watch CCSU play, it certainly feels like Jamir Coleman should be that guy. And he has been on occasion. After the 6'7" senior went off for 27 points on 16 shot attempts in the close loss to UMass-Lowell on December 1st, this is what I wrote:
"Here's the thing about CCSU's loss on Sunday; the only reason they were even in that game was because Jamir Coleman shot the ball out of his mind; 12-16 from the field, including 11-13 from two (3-5 from the mid-range), and he got to the rim 8 times. Is that sustainable? Coleman had a 143 O-Rating on 29% usage on Sunday. Before that, his best O-Rating with a usage rate north of 25% was 125 last February in a loss to Wagner, where he scored 18 points on 6-11 shooting. What's more, Coleman had struggled through his team's first five games with an eFG% of 35.5%. What's more likely; i) Coleman is going to become a league POY candidate going forward; or ii) this game was an outlier? I'd lean Option B here."
Since then? Coleman was having a great game against Maine (4-4 from the field, 13 points) before leaving with what looked like a potential concussion. And in the three games since? He's just 4-19 in 49 minutes, with 11 rebounds, 1 assist and 7 turnovers, which continues an up-and-down career in which he's looked like an All-NEC player at times, and merely a role player at others.
But the other issue is neither Coleman, nor anyone on this roster, consistently makes plays for others. Central doesn't have a single player averaging more than 1.7 assists per game (Newkirk), nor does any player currently have more assists than turnovers. On December 7th, with CCSU down two points to Maine in the closing seconds, Marshall opted to run a play for freshman Myles Baker, who ultimately attempted a step-back three-pointer that was blocked.
Baker probably wasn't the right guy for that moment (he's converting on just 32% of his shots at the rim, and he's 0-13 from the mid-range), but honestly who was? Greg Outlaw is the one guy who has shown an ability to get the ball to the rim with the dribble, but he has a knack for taking bad shots (as I mentioned earlier), and Coleman was out due to the aforementioned injury.
Ian Krishnan, who will certainly give this roster a boost, wasn't a play-maker last season; he averaged just 1.2 assists per game, and 209 of his 355 field goal attempts (59%) came from behind the three-point line. He's a sharpshooter (and a really good one), and unlikely to solve this problem for Central.
6. Coming into the season, we all knew this team would be young (they're currently the 4th least experienced team in Division 1 according to Kenpom), but at we could dream on some of the young players. So how do the CCSU freshmen stack up compared to the rest of the Northeast Conference? Courtesy of Bart Torvik, sorted by BPM:
For some reason Torvik's system has Elijah Ford and Zach Newkirk as freshmen (they're sophomores), so ignore them.
Greg Outlaw has been more than solid overall, but the rest? Jamir Reed and Trey Tennyson have been better of late, while Myles Baker has really struggled (plus he missed two games for unspecified reasons), as has Xavier Wilson.
However, the "glass half full" look at this is that, while the freshmen have largely been inconsistent, each has shown flashes of major potential.
Outlaw scored 18 points (and grabbed 7 rebounds) at Boston College and has scored in double figures 8 times. Plus, he's really shown an ability to get to the rim (39 for 69, 23 more attempts than any other CCSU player) and he's a lock-down perimeter defender (2.8% steal rate, which is 12th in the league).
Reed was averaging 11.7 ppg over the six games prior to the Penn State loss, and has shown himself to be a strong rebounder (his 13.7% DR% is 4th on the team). He's making 32% from three, 67% from the free throw line, and anecdotally seems the most comfortable offensively.
Tennyson really struggled in the backup point guard role, but since being moved to (almost) strictly off the ball, he's played more like the guy I expected; in his last three games he's averaging a little over 10 ppg thanks to shooting 5-13 from three (42% on the season), and hasn't turned it over nearly as much as he did earlier in the season. He's long, athletic, and can really shoot it, and I think there's a lot of upside here.
Baker is shooting an abysmal 20% from inside the three-point line, but beyond it he's been really good (37%). Anecdotally, I think he's trying to do too much (or being asked to do too much); he's taking over 27% of the team's shots when he's on the floor (which is the highest on the team). If he was allowed to be more of a 3&D guy while the rest of his game develops, we'd be talking about him the way we talked about Ian Krishnan last season.
Big men often take more time to settle in at this level than guards do, and Wilson is certainly no exception. We've already talked about his poor shot selection, and he really needs to get better on the boards (though he did grab 8 rebounds at Maine, and had 7 in the loss to UMass-Lowell). But he's 6'7" and a solid 210 lbs. (I'd bet he's closer to 225), has blocked 10 shots, and has shown an ability to stretch the defense (4 made three-pointers).
7. In addition to struggling to contain three-point shooting, CCSU has really had an issue protecting the paint; opponents are shooting 66.1% on "close twos", which is 42nd highest nationally and worst in the NEC. Part of that has been the injury to Karrington Wallace, who is certainly the team's best rim protector, but part of that is the lack of depth at the '5'; Stephane Ayangma and Coleman are both undersized for the position (even in the NEC) and Wilson is only a freshman.
Similar to the point guard situation, I think it's imperative Marshall tries to bring in someone who can be more of a rim protector for next season, especially since Wallace is more of a "help guy" as opposed to a guy who can defend opposing bigs in the post. 6'9" freshman Ola Olamuyiwa appears to be a project (he's essentially only played garbage minutes this year, though had a nice showing at Penn State), and I'm not sure going into next season with Ayangma and Wallace as your starting bigs will provide enough frontcourt defense.
8. Over the past few games, Marshall has toyed with a couple of different full-court press' (man-to-man, as well as what appeared to be a 2-1-2 zone). And I have to say, I like it! I don't think it needs to become part of their defensive DNA, but it gives the opposing team something to game-plan for, and it gives the staff an option in games where they are falling behind and need a quick spurt. I think right now three press will struggle without a rim protector at the back of the defense, but once Karrington Wallace is back he should help in that regard.
Additionally, Marshall has started mixing up the defenses more, going with zone at times. In the loss to Dartmouth, it was clear that the Big Green offense was thrown off a bit by the different defenses, at least until they started knocking down open perimeter shots.
If Marshall is going to turn this program around over the next couple of seasons, it's going to have to happen on the defensive side of the ball; absent one or two major junior college additions (or major developments with some players on the current roster), I don't see Central Connecticut suddenly having one of the more efficient offenses in the Northeast Conference. However, the quickest way to increased wins is often improvement on the defensive side of the ball (Bryant has gone from having the worst defense in the league last season to the best this year, so that kind of improvement is possible).
As of this writing, CCSU has the 344th ranked defense in the nation (per Kenpom), and in Marshall's first three seasons they were: 324th, 296th, and 313th. As we talked about earlier, they've consistently struggled to defend the three-ball but have been decent at forcing turnovers (69th in 2017, 196th in 2019) and rebounding (157th in 2018, 284th in 2019). If they could somehow curb their opponents' ability to make three-pointers at a high rate, they could really have something here defensively.
9. Prior to the season, I predicted that the 2019-20 version of Central Connecticut would follow a similar path as the 2018-19 Mount St. Mary's team. So, let's compare:
The Mount had just one non-conference win last season, a 56-55 road win over American which finished at #200 in Kenpom. Meanwhile, CCSU lost two one-possession games (at Maine and vs. UMass-Lowell).
However, the comparisons kind of end there. Despite having an even younger team last season, the Mount was able to avoid blowout losses, outside of a 105-55 season opening loss to N.C. State. They fell to #24 Maryland by 15, #88 St. John's by 14, and #93 Hofstra by 18. CCSU, on the other hand, lost by 29+ points to #24 Penn State, #80 Vermont, #84 St. John's, #88 Arizona St., #131 Boston College, #194 UMass, and #215 Columbia. The records look similar, but the performances were not, and is captured by their average Game Scores (according to Bart Torvik).
Of course, it's worth noting that Central has gotten just 5 games from Karrington Wallace (ankle injury) and zero from Ian Krishnan, while Myles Baker missed two games and Jamir Coleman missed one full game and about half of two others.
Despite having players miss time, it's clear that, as of right now, CCSU's rebuild is behind Mount St. Mary's' at the same time last year, and it's not like the Mount has exactly taken off in Dan Engelstad's second year; they're currently 2-9 against D1 competition and are ranked #276 at Kenpom, though Vado Morse has been battling a knee injury that kept him out of a couple games and has him clearly hobbled.
10. Kenpom projects CCSU to finish 3-15 in the Northeast Conference, while Bart Torvik is even more bearish (2-16), with both systems projecting Central Connecticut as an underdog in each of their remaining games. So if we set the over/under at 2.5, what are you taking?
On one hand, the NEC is void of truly bad teams this season; last year there were 4 teams rated #314 or worse at Kenpom. This year CCSU is the lone team ranked below #310, and Central only plays St. Francis-Brooklyn (#310) just once this season (as well as Mount St. Mary's). On the other hand, there has to be at least a few games in which teams shoot poorly in Detrick Gym while Jamir Coleman and/or Ian Krishnan shoot the lights out, right? RIGHT?!
I'm going to say that CCSU will win at least three league games this season, only because the alternative is super depressing. Can they make a push and qualify for the NEC Tournament? The last four #8 seeds in the NEC Tournament averaged 7.5 wins per season, so let's go best-case scenario and said that CCSU needs to finish 7-11 in the league to qualify. Bart Torvik gives the Blue Devils just a 1.3% chance of winning 7 games (and a 0.4% chance of finishing 8-10). Not likely.
*insert Jim Carrey/Dumb & Dumber/SO YOU'RE TELLING ME THERE'S A CHANCE gif*.