St. John's Needs to Play Ruben Prey More
a lineup analysis, where I attempt to uncover the team's most optimized lineup
Dear Rick Pitino,
The best long-term prospect on your team is Ruben Prey. He has done nothing but produce. Please triple his minutes per game, to both improve your team’s outlook and to give him the platform to deservedly make the NBA.
Sincerely,
100guaranteed.
This is analysis divided into two parts. The first revolves around my contention that Ruben Prey is the best prospect on St. John’s Red Storm. The second is moreso on effectively utilizing the existent talent on this team.
As discussed in my thread yesterday, Prey has a pretty impressive combination of cognition, physicality, size, and production. The query above is a simple intersection of A:TO and stocks, an indication of possession value. Just 10 players have qualified, but it’s impressive that Prey
St. Johns has underperformed this year, and much of its lackluster start has been placed on their guard play.
In the summer, Pitino made headlines by declaring that the point guard position is “no more”, and that these players are typically scoring liabilities. His offseason acquisitions reflecting this philosophy, as 3 of his 4 guard transfers posted a career assist rate under 10%. All 4 had a career 3P rate over 40, which was likely a product of Pitino emphasizing a previous lack of shooting: 2024 St. John’s ranked 340th in 3P%, by far the worst rank of Pitino’s lengthy coaching tenure.
Readers are likely familiar with my emphasis on cognition and mass. Death of guards is an unfortunately common expression as of late, but it ignores the position’s importance towards maintaining and extending possessions. This is primarily why guards have the highest cognitive burden, and why I’ve consistently relied on some combination of assist rate, A:TO, and steal rate to estimate cognition. I’ve tweeted before about how player “hold time” and a product of hands-on-ball, steal% and A:TO have matched up reasonably well, and it’s all part of this cognition paradigm.
Adding 3 players with prototypical guard size (under 6’6), low assist to turnover rate, and sub 1.5% steal rate indicates a horrific cognition-for-mass. Sellers’ career 1.5 assist to turnover rate indicated some upside, but Ian Jackson and Joson Sanon’s combination of horrible assist rate, steal rate, and A:TO were basic but high fidelity signals that they were generally unlikely to meet the value proposition of guards. Pitino’s misunderstanding of the cognitive value proposition of guards is interesting given his career-long success in forcing defensive turnovers.
Nonetheless, this has gone exactly how we would have predicted. St. John’s has dropped from 48th in offensive turnover rate to 193rd. And all due respect to RJ Luis, but that was a team that played RJ Luis and his 27% usage+ 0.7 A:TO pretty heavily. St. John’s takes lots of dunks and midrange 2s, so their turnover rate should theoretically trend lower.
Take a wild guess at the Red Storm’s worst two man pairing by far (luck adjusted, vs top 200, minimum 100 possessions).
This is just horrific. It gets worse, though:
With the duo on, St. John’s falls to the 2nd percentile in “attack & kick” plays. It isn’t that surprising that this duo is especially horrific at making passing plays, but it’s notable that rim attacks AND perimeter cuts are also inefficient. Contrary to popular opinion, playing two guards with high 3Pr does not inherently improve offensive flow. Poor cognition at the guard position certainly leads to ineffective passing impact, but it reasonably has downstream effects not limited to a mitigation of rim efficiency.
With the duo off, the Red Storm are just unbelievably good, and it implies that Jackson/Sanon’s backups are edges hiding in plain sight. The turnover rate difference is just jarring.
Note how “attack and kick” frequency went from 2nd percentile to 25th percentile. And note how efficient the team is at scoring on rim attacks and perimeter cuts. The entire offensive process is much more fruitful!
It doesn’t take much detective work to come to the conclusion that Dylan Darling is the answer at point guard.
I don’t want to spend too much time on this given how clearly obvious it is. I do want to note some things about Darling.
Darling played a handful of minutes for Washington State as a freshman in 2023, and then medically redshirted in 2024. His teammates during this span included Mo Gueye, Jaylen Wells, Isaac Jones, and Oscar Cluff. Pretty unbelievable level of talent looking back on it.
Upon transferring down to Idaho State for the 2025 NCAA season, Darling exploded. He comfortably won Big Sky Player of the Year as he put up a wildly efficient season in nearly every possible facet.
Still, there were some clear holes in Darling’s profile. His eFG% ranks 14th out of 17 here, with 48% 2P and 36% 3P. We had a huge sample of FTs, and while 81% is quite good, it’s not good enough for us to ever bill Darling as a sharpshooter. 48% 2P in the Big Sky is a fairly big red flag, even if just 8% of his rim attempts were assisted. Coupled with zero dunks and a 6’1/175 pound frame, Darling’s inside-the-arc projection wasn’t great.
And it has sort of continued to be an issue.
Darling is shooting just 56% at the rim this year, but he’s 2/5 at the rim in 5 games versus top 100 competition. His rim frequency falls off strongly in these games, meaning his poor rim efficiency has largely come against lackluster competition. This combination of low rim frequency and poor rim conversion is likely hard for Pitino to stomach, as is his 25% mark from 3. The ball simply isn’t going in.
With all of his guards seemingly duds, Pitino recently lamented “that’s our fault as a staff for not having a big-time point guard that makes people better”.
I sympathize with Pitino, but this is an objectively incorrect statement. Darling makes his teammates better even if the ball isn’t going in right now. He backpacked a poor Idaho State team to incredible heights last year, and he’s supercharged Red Storm lineups this year. Darling has only played 42% of team minutes this year, and just 34% against top 100 competition. I strongly believe that proportion should be closer to 80%.
In 5 games against top 100 competition, Pitino has played a total of 4 lineups that have accumulated over 20 possessions. Upon garbage and luck adjustments:
A Dylan Darling - Oziyah Sellers backcourt is clearly in the Red Storm’s best interest. The team has actually fared decently well with its most frequent 5 man lineup (net 20.2), but it’s losing games because of the badness of its other lineups. The strength of the team is in its frontcourt, but the underlying strength of the Red Storm’s lineups actually comes down to which 2 or 3 of Darling/Sellers/Sanon/Jackson that Pitino decides to play.
In non-Darling minutes, I would advise a Sellers/Jackson backcourt. That lineup has a micro 11.6% turnover rate, with a monstrous 86% of team shots coming at the rim or from 3. My worry is this backcourt will not force defensive turnovers consistently enough (see: 15% dTOV rate for the lineup, as well as career steal rates under 1.5% for both Sellers and Jackson). In hopes of avoiding an identity crisis for Pitino, I would primarily suggest a backcourt with at least one defensive turnover generator, and that is Darling by default.
By far the three best 4-man combinations for St. John’s in these top 100 games were the following (the next 3 combinations had net rating of 61, 52, and 48).
In context of the Red Storm’s turnover + 2P% woes, these lineups are even more absurd. And they consist of 6 players: Darling, Sellers, Hopkins, Mitchell, Ejiofor, and Ruben Prey. If rotations are to be kept short (a decision that may not be as straightforward given the taxing nature of the Red Storm’s defensive tendencies), these are the 6 players that should receive the most minutes.
Love the Ruben prey respect
If pitino stays stupid I wouldn’t be surprised to see an analytically sound college basketball team pick him up next year