Anyone who tells you they knew what this team would look like at the start of the year is lying (or youngBUCK).
(Editor's Note: Or Bosshoghazzard)
(Corollary's Note: No, I don't think TC is our best player. It's a headline.)
Coming off an unpredictable season highlighted by 3 wins over top-five squads and a return to the meaningful postseason, yet pockmarked by several atrocious losses and plenty more way-too-close calls, it looked like we were in for more of the same this year – at least as far as the unpredictability went. Mouph was gone, the upperclassmen were underwhelming, everyone else was either in their first or second year in the program, and there was no size behind Ochefu on the front line. A collection of unknown quantities we expected to play hard on defense, shoot a lot of free throws, and….something. What that something would be was anyone’s guess, and fans made plenty. Hell, there was a large group of people expecting the offense to run through Danny Ochefu and his ‘post skills.’ It was nearly impossible to stick this team with an identity, or make concrete predictions about what it would look like, because there was too much that we couldn’t know.
Nine games in, it’s safe to say we know PLENTY more. Villanova’s vaulted back into the national conversation on the back of incredible team defense, a rotation that goes 9-deep, a small-ball/large-wing attack that doesn’t feature (so far – STOP FOULING AND TURNING IT OVER, CHIEF) a replacement-level true big man on either end – and, oh yeah, an undefeated campaign that includes wins over two top 25 ranked teams. Jay Wright has installed inbounds plays, game-planned surprisingly well, worn suits, and recruited towards/fostered excellent team chemistry. It’s a talented, lovable team that’s generated a lot of excitement around here.
James Tahj Bell has been submitting a sublime all-around campaign as a senior, and is almost certainly, in my mind, the MVP of the season’s first third. Tony Chennault has improved from last year to be a do-everything-well-besides-shooting bench guy who’s contributed excellent defense and playmaking whenever he’s on the floor. Darrun Hilliard continues to be a stat-sheet stuffer with a developing/improving outside shot who will never hesitate to drive. New additions to the team have made impacts – the transfer, D(ylan) Ennis "THE MENACE" has brought defensive energy (though not necessarily results), a willingness to drive, and killer 3-point shooting to the 1. Josh Hart and Kris Jenkins have come in as freshmen to provide quality minutes, excellent shooting, and basketball IQ beyond their years at the wing. The rest of the rotation players, while all useful and contributing, haven’t changed much from what we saw last year, so won’t be written up too much in this particular paragraph. Arch, Chief, and JVP continue to be what they are, and we all (sometimes) love them for it. We’re deep, flexible, athletic, and can feel confident in our ability to stay on the floor (while winning more than a fair share) with every team in the country.
In fact, the big questions around the team have generally been answered – they’re good, likeable, hard-working, will be competing for the top Big East spot, and will almost certainly be heading for an NCAA tournament berth even if they don’t land at the pinnacle of the BE. The next 20 games will be about answering the smaller questions – can the team sustain its excellent team rebounding and defense without playing a player over 6’-7"? Can small-ball possibly work for six straight games, come March? Can Ochefu catch a ball without turning it over – and, more importantly, develop into an important rotation cog by lowering his fouls and playing smarter/more confidently on offense? How will minutes shake out among the 9 players currently in the meaningful rotation – and, perhaps more relevant (to our discussions), how should we want them to?
These are the questions I’ll be trying to provide some insight into with this write-up.
Not only have these 9 games given us plenty to love about the team – they’ve also provided us with a semi-meaningful sample size. As a side project, I went through the process of parsing out how the team has statistically played so far with each player both in and out of the game.. I have data on 2-, 3-, 4-, and 5-man lineups, but the samples, especially for the 4- and 5-man lineups, don’t extend much farther than 20-30 minutes– essentially meaningless samples over the course of a season.
We’ll delve into traditional basketball statistics, introduce and revisit some advanced statistical analyses (including the hallowed Four Factors), and generally just talk math. Won’t be exciting, but hoping it’ll be interesting.
It should be noted that I’m not a math major.
All statistics are gleaned from internet-publicly available box scores for Villanova’s games. The Lafayette game wasn’t included in the compiled data, as the box scores I found were generally incomplete when it came to lineup data. At more than one point, only 2 Villanova players were in the game (per the substitutions), and I couldn’t find a way to easily resolve the issues short of rewatching the game and rewriting the box score. Couldn’t bring myself to do it. So, this sample includes the 8 games the team’s played so far AFTER the Lafayette game.
The rebound statistics for each player and lineup will not take into account rebounds credited to the team – also known as dead ball rebounds.
Dead ball rebounds are credited to a team when:
- Possession is retained after a missed free-throw attempt which is followed by another attempt.
- Possession is retained after a missed technical foul shot.
- Time expires prior to a player or team gaining possession of a missed shot.
- A foul occurs on a missed shot and the ball becomes dead before a player or team can gain possession.
Rebounds specifically credited to a player are counted.
Only the 9 players currently seeing meaningful minutes for Villanova – Ryan Arcidiacono, Darrun Hilliard, James Bell, JayVaughn Pinkston, Daniel Ochefu, Josh Hart, Tony Chennault, Dylan Ennis, and Kris Jenkins – have had their stats summarized here. With apologies to Reynolds aficionados and fans of the Bench Mob, there’s really no point in charting their stats – they don’t play minutes while the outcome of the game is in question.
Most are classic counting stats – points, field goals made, attempted, assists, turnovers, and so on. Those of you familiar with Dean Oliver’s work – and, of course, the excellent Ed Donohue – will recognize effective field goal percentage (eFG%), free throw rate (FTR), offensive rebounding percentage (O Reb%), and turnover percentage (TO %).
Effective field goal percentage (eFG%) is a statistic that adjusts for the fact that 3 pointers are worth more than 2 pointers – the value of the 3-pointer isn’t captured by simple FG%. It’s calculated as follows: (FGM + 0.5 * 3PM) / FGA. True shooting percentage (TS%), related to eFG%, takes into account the added value of a 3 pointer, as well as the points generated from free throws. While not one of the four factors (free throws are considered in a separate one), it’s also an excellent indicator of a player’s (or team’s) overall effectiveness from the field and takes into account all methods of scoring. It is calculated thusly: Total Points/(2*(FGA +( 0.475*FTA))).
Free throw rate, as calculated here, is a way to measure how many free throws are generated per field goal attempt, and is calculated simply by divided the free throw attempts (FTA) by field goal attempts (FGA).
Offensive rebounding percentage (OREB %) is the percentage of available offensive rebounds a team grabs, and is calculated as follows: Team Offensive Rebounds / (Team Offensive Rebounds + Opponent Defensive Rebounds).
Turnover percentage (TO%) is simply the percentage of offensive possessions that result in turnovers – divide your turnovers by total offensive possessions, and you’ll get TO%.
These are considered the advanced statistics most closely correlated with winning, other than the obvious (POINTS!) – win the majority of these stats (vs. the opponent’s), and your team will often win the game.
Well-read basketball fans out there will also recognize the tempo-independent stats – offensive rating (O Rtg), and defensive rating (D Rtg) – found on Kenpom and most advanced-statistics-for-the-NBA sites (basketball-reference, etc.). These stats are calculated by dividing total points scored by total possessions - points scored by Villanova / offensive possessions for O Rtg, and points scored by opponents / defensive possessions for D Rtg. These ratings are a method of ranking defense or offense by actual possessions faced, rather than simply per game stats. This is a more accurate indicator of the relative strength of a team’s offense/defense than simply points allowed and scored per game, and helps stack teams that play at different paces up against each other.
The possessions are calculated by summing up field goal attempts and turnovers, while accounting for those possessions that end with offensive rebounds and free throws. While most possessions ending in free throws generate two of them, and-1s and fouls on 3 pointers must also be accounted for. An accepted adjustment factor of 0.475, rather than 0.5, is used to adjust free throw attempts into possessions for the college game (NBA is 0.44). The formula used to calculate possessions is FGA + 0.475*FTA - Offensive Rebounds + Turnovers. The same (using the opponent's corresponding statistics) is used to generate defensive possessions.
Reference will often be made to points/100 possession, which involves simply multiplying these tempo-independent stats (O Rtg, D Rtg) by 100 – the units of Rtg are given as points/possession.
Other stats are listed, but are generally recognizable / labeled advanced statistics. Those relevant to discussion will be referenced, but you guys can do some homework, too.
Remember – these statistics do not indicate how well the individual player (or group of players) does statistically, but how well the team plays around him/them while he’s/they're in the lineup. Obviously, the player’s stats are included if he’s in the game, but I wanted to look at the overall picture of how well the team plays with (or without) a certain player, or group of players.
Let’s get into the actual results, starting with lineup analysis of offense/shooting, defense, and the four factors/tempo-independent stats when each of the 9 players examined is in the game.
|Players||MIN||O Poss||Ort||PTS||FGM||FGA||FG%||2P%||3PM||3PA||3P%||FTM||FTA||FT%||AST||TO||TO%||eFG%||TS%||FTR||AST %||A/T|
Anyone who’s followed my comments this season (all 2 of you) knows I’m quite interested in the point guard rotation, and how Arcidiacono, Ennis, and (to a lesser extent) Tony C should be splitting up the minutes. The most interesting takeaway from these stats? Of the 3 point guards currently rotating, the offense functions best with Tony Chennault (!!!!!) in the game – he of the 3 made field goals (all year). The team is scoring 123 points per 100 possessions with Chennault in the game – compared to 115 per for Ennis, and 108 for Arch. How is this possible?
Tony C’s lineups are DRILLING it from 3 – in the 142 minutes he’s played in the sampled games, the team has shot 42.25% from range – nearly 7 percentage points higher than the team’s average (35.20%) through the same games (remember, Lafayette has not been included). The team’s free throw rate also skyrockets when he’s in the game (2nd highest of the 9 rotation players) – and, while some of this is certainly attributable to Chennault closing out a few blowouts and getting the resultant foul-from-behind-free throws, the sample is large enough to suggest these positive indicators could be a legitimate trend.
I’ve long been a fan of Tony’s ability to dribble wherever he wants on the floor, and his passing vision has really never been questioned. What has been a concern is his apparent unwillingness/inability to shoot, which can allow smarter defenses to essentially play 4-on-5 defense, sagging off his drives and pick and rolls to overplay the real shooters and post threats on the team. And, while this certainly will be something to watch against the smarter and better coached teams Villanova plays, it has so far not hurt his team’s offense at all. Tony’s ability to get it into the lane, make solid entry passes, and find open shooters on the perimeter has far outweighed any concerns over his shooting thus far. FREE TONY.
And, while Josh Hart has gotten 99% of the frosh love splashed around on these boards (and I’ll cop to generating more than my fair share of it), Kris Jenkins has the highest offensive rating of any player on the team, beating out Tony C by hundredths of a percent. The freshman big-ish man’s ability to knock it down from 3 (currently shooting almost 35%, 4th highest on the team) while he’s spelling a wing, JVP or a foul-troubled Chief has done wonders for the offense. Lineups that include him are generating the highest free throw rate (55.00%), effective field goal percentage (57.50%), and A/T ratio (1.66). His presence on the perimeter helps pull big men away from the basket, and opens up driving lanes for the rest of the team, allowing for a number of shots at the rim and kick-outs for open 3s. While this team may be all Hart, we need some more Jenkins. He’s been fabulous from an offensive standpoint so far, in terms of helping the team.
- The offensive efficiency of the starters is being far outpaced by the players coming off the bench – a testament to the team’s depth and overall talent level. The team’s ‘bench’ players are simply destroying the 2nd teamers they see, to the tune of about 10 more points/100 possessions (offensive efficiency only) when compared to the starters.
- The effective field goal percentage falls off a cliff when Arch, JVP, or Ochefu is on the floor – lineups including them compose 3 of the lowest 4. The other member (also troubling)? Dylan "My Man Crush" Ennis.
- Hilliard’s minutes have resulted in the highest turnover percentage – certainly related to his willingness to drive – and his low offensive efficiency is also troubling. I love his game, and what he brings to the team. But, it may be time to start keeping closer tabs on his efficiency stats, to make sure what the eyes see are what the stats confirm.
Let’s take a look at the defensive stats – in this chart, all field goal/free throw attempts, makes, percentages, and turnovers refer to the opponent’s stats when the referenced player is on the floor. Steals, blocks, and offensive/defensive rebounding percentage are Villanova defensive stats during the same time frame.
|Players||Mins||Def Poss||D Rtg||% of 2s
|2PM||2PA||2P %||3PA||3PM||3P %||FTA||FTM||Opp eFG%||STL||BLK||STL%||BLK%||OR%||DR%||TO||Opp TO%|
Once again, the story of the chart is Tony Chennault. While he’s on the floor, opponents score an absurdly low 77 points per 100 possessions, with corresponding jumps in steal percentage and opponent turnover percentage. He’s also been part of some of our best 3-point defending lineups – the team, in the 142 minutes on the floor with Tony C, is allowing opponents to hit 27.14% from deep – a great number, and good for third-lowest on the team (behind fellow bench guys Ennis and Jenkins).
Now, these results are clearly backed up by the eye test. Chennault has been incredibly disruptive on defense, most notably in Bloody(ing) Kansas (in the Bahamas), where he generated two steals and numerous other interferences while the team was pressing Kansas’s ballhandlers. His senior presence, defensive instincts, and actually positive impact on the offensive end haven’t really been captured by the traditional box score just yet. But, don’t know about you, I’m going to be rooting for more Tony C minutes in the future.
The strangest note of the table – or perhaps just a reality check – is how poorly lineups including Josh Hart have defended. Lineups including the frosh have, by far, the highest points allowed per possession (94 points/100 possessions). There’s certainly some evidence on tape to back this up. Despite excellent one-on-one defensive intensity and great lateral quickness, Hart has a tendency to overhelp and gamble for steals in some unfortunate situations. Good passes out of these traps/rotations often lead to excellent looks for the opponent, especially from outside – evidenced by his lineups allowing the 3rd highest percentage of 3s (behind JVP and that paragon of perimeter athleticism, Ryan Arcidiacono), and the highest eFG% of the entire lineup. In the midst of the praise for his willingness to take it to the rim, offensive rebounding abilities, and generally energetic play early this year (all completely justified), we may have lost sight of the fact that he’s just a freshman. Grasping the more complex defensive schemes at the college level, and finding a comfort level within them, certainly has an adjustment period for freshmen – and that may be what we’re seeing here. Here’s to hoping what we see on-court translates to what stat analysis shows later in the year.
- Hart and Jenkins are currently the worst defensive players on the team - per the lineups they’ve played in. Sorry, guys, but I don’t have time to chart how they do as individual defenders, or how well their main scores – so not sure if this is a sample size anomaly or a trend indicator. Certainly another thing to keep an eye on.
- This an area where the starters are somewhat stronger – over generally double the sample size – than the bench players (in terms of D Rtg). Despite the occasional offensive stagnation, the general starting lineup of Arch-Hilliard-Bell-JVP-Chief has done a great job defensively.
What follows is a summary of the lineups (including each highlighted player), ranked according to their contributions in the 4 factors (eFG%, TO%, OR%, and FTR). Each player is assigned a rank relative to the others (1-9) based on the highest eFG%, OR%, FTR, and lowest TO%. A weight is then applied to each rank, and divided by 100 for the final determination. The weights associated with each factor are, respectively (eFG%, TO%, OR%, and FTR), 40%, 25%, 20%, and 15%.
|Players||eFG %||TO %||OR %||FTR||O Rtg||D Rtg||Differential/100
While this chart backs up much of the analysis we’ve already been through, a few notes. First – Tony C’s lineups are outscoring the opposition by 45 points per 100 possessions. That’s an incredible rate – and it’s not borne of a tiny sample size, as Tony C has logged over 140 minutes in the 8 games sampled. JAY, PLEASE READ THIS. Arcidiacono, my least favorite PG, currently has the lowest plus/minus (as that’s what it basically boils down to) per possession, at +22.45 points/100, and ranks 4th lowest on the ranking scale in terms of contribution to the Four Factors. He needs to improve his play, or make way for PGs who can.
Ochefu’s lineups rank 3rd in offensive rebounding percentage, behind wings Josh Hart and James Bell. While Ochefu may grab the highest percentage of offensive rebounds on the team while in the game, his presence may not necessarily be best for the overall offensive rebounding percentage of the team. As we saw in the Iowa and Kansas games, the team can have a ton of success attacking the boards with the oversize wing rotation – which is something they wouldn’t necessarily do with such abandon while Ochefu is in. Yet another stat to keep monitoring as we move forward.
The starters, outside of Bell (MVP! MVP!), generally sit at the bottom of the rankings and differential/100 possessions. While the fact they often play against the best of the other team is certainly a factor, as well as the noticeably larger sample size of minutes, such a large disparity between the starters and bench players suggests that Jay should consider giving some of the bench guys more minutes while he’s fine-tuning the rotation for tournament time.
WITHOUT 1-MAN LINEUPS
Moving on. Next we’ll look at the (without) 1-man lineups, that analyze how the team does WITHOUT the particular player referenced on the left of the table. Starting off with the shooting percentages/offensive statistics..
|Players||MIN||O Poss||O Rtg||PTS||FGM||FGA||FG%||2P%||3PM||3PA||3P%||FTM||FTA||FT%||AST||TO||TO %||eFG%||TS%||FTR||AST %||A/T|
As far as the point guard battle goes, these stats are laughably damning of Arch’s play so far. In the 78 minutes (admittedly, a much smaller sample size than what we’ve been analyzing so far) the team has played without Arch, they’re shooting better than 50% from 3 point range, generating free throws at a 60% rate (FTA/FGA), shooting with an eFG% of almost 59%, and scoring 130 points per 100 possessions. The splits between on/off are easily the highest for Arch (130 points/100 off vs 108 points/100 on); very simply, the offense is miles better without him in the game. The extended shooting slump – though, again, I will note that poor shooting is a trend in his college career, not an anomaly – this year, combined with a lack of drives, free throw generation, and generally everything else you want to see from a PG (outside of assists – he’s mildly decent providing those), has seriously hurt the team’s offense. He needs to improve, or get less minutes, when the squad plays better defenses. Tough to debate that’s not true.
JVP clocking in as the second highest split between on/off offensive efficiency (the team scores 19 points more per 100 possessions more when he’s not in) is just an indication of the offensive game the team plays while he’s in there. JVP is a black-hole on offense, a usage % machine who’s generating a layup, free throws, or a turnover whenever he gets the ball. He’s a microwave post scorer – a guy the team can dump it to inside when they need to generate offense, by hook or by crook – but not someone to be trusted to generate high-level offensive play throughout a game.
- Hart-less lineups currently have the 3rd worst points scored per possession. THIS COMPLETELY JUSTIFIES ALL OF MY HART LOVE – DISCUSS. Worst two? Missing Chennault and Bell.
- Can we just go back to that stat that shows the team shooting 50% from 3 when Arch isn’t in the game? Think I’m in love.
On to all that Defensive Efficiency.
|Players||Def Poss||D Rtg||% of 2s
|2PM||2PA||2P %||3PA||3PM||3P %||FTA||FTM||Opp eFG%||STL||BLK||STL%||BLK%||OR%||DR%||TO||Opp TO%|
Most interesting stat for me here? The defensive efficiency with Ochefu on/off the floor is exactly the same (85 points / 100 possessions). He literally has not made a difference on the defensive end – and if he’s not doing that, what’s he good for? Personally, I believe he’ll be useful in terms of sustaining the team’s energy level during conference play – it’s easier to defend when you’ve got a rim protector behind you – but when it gets down to crunch/tournament time, honestly don’t think I’d want Ochefu in the lineup.
James Bell has been one of the most important defensive players on the team – yet another case for his MVP-of-the-first-3rd-status. Teams are scoring 91 points/100 possessions when he’s not in – 2nd highest on the team. The offensive rebounding rate also literally craters when Bell is not in – 31.17% is the lowest by an appreciable amount. He’s been a do-everything monster who provides serious size and effort from the 3. Love that guy.
- Opponents shoot less than 15% from 3 when Arch isn't in. It’s unclear if this has anything to do with his defensive ability. I think it does.
- Lineups without Josh Hart are incredibly stingy - we’re talking Chennault-ian levels, here. Another indicator of his ‘secret’ struggles on the defensive end – and a trend I think we all hope he can reverse. Love watching that kid play.
And, here’s the Four Factor contributions of each lineup that doesn't include the referenced player.
|Players||eFG %||TO %||OR %||FTR||O Rtg||D Rtg||Differential
The conclusions we can draw from this graphic are much the same as I already have – the team is better, in a lot of ways, without Arch (remember, all these stats show the rates WITHOUT the referenced player). Look at that offensive rebounding percentage! The team rebounds almost 47% of its misses when he doesn’t play. That’s insane.
JayVaughn hurts more than he helps (a lot) on offense – except with free throw generation. He’s actually amazing at that – which makes him an integral, though not necessarily crunch-time, part of the offense.
Strange/non-already reached conclusions to draw – Darrun Hilliard has hurt the offense almost as much as Arch (though he seems to help more on defense). Again, his efficiency stats are something to monitor going forward. Not the most encouraging thing.
This is the end. Was going to go through 2-man lineup data, but after 4,000 words, I figure I’d hear some feedback before I put more out there.
Arch's crunch-time lineup stats (not just his own). Only 20 minutes to sample.
|Players||MIN||O Poss||O Rtg||PTS||FGM||FGA||FG%||2P%||3PM||3PA||3P%||FTM||FTA||FT%||AST||TO||TO %||eFG%||TS%||FTR||AST %||A/T|
More from VU Hoops Follow @VUhoops