Analyzing the February Fade

What do the numbers tell us about Stanford's struggles in February, and what has happened since?

A 3-3 February set Stanford up for a brutal 2-4 March that detonated any hopes of participation in the NCAA Tournament. Much was made (by me, among many others) of the role fatigue played in the declines of Chasson Randle and Anthony Brown. Last week we looked at their monthly splits and here is what we found:

Jan eFG
Jan 2P%
Jan 3p%
Jan TO/G
Feb eFG
Feb 2P%
Feb 3p%
Feb TO/G
Chasson Randle
Anthony Brown

Check the January performances against those same numbers in February. There is no question at this point that they suffered a massive drop-off in performance, and it's hard not to attribute the decline to the extraordinarily high volume of minutes played by both. Chasson ended the season having played 88.7% of the Cardinal's minutes, while Brown wasn't far behind at 87.5%. Anecdotally, we all remember last year, when back to back 40 minute games in St. Louis led to Chasson's collapse against Dayton in Memphis. Oh, to have that "problem" this year!!

Anyway, it got me thinking about this team as compared to last year. Was the workload too much? Is that really what undid The Platinum Backcourt? I compiled the numbers from the top for minutes leaders from last year and this year. Here's what they show:

Minutes %
Jan eFG% (Min/Game)
Feb eFG% (Min/Game)
Mar eFG% (Min/Game)
Chasson Randle
47% (38)
57% (34)
54% (37)
Josh Huestis
45% (35)
46% (35)
49% (37)
Anthony Brown
51% (34)
61% (35)
45% (32)
Dwight Powell
50% (34)
42% (34)
46% (30)

Now here's the same table for this year:

Minutes %
Jan eFG% (Min/Game)
Feb eFG% (Min/Game)
Mar eFG% (Min/Game)
Chasson Randle
54% (37.4)
33% (36)
45% (36)
Anthony Brown
56% (38)
48% (37)
49% (35)
Stefan Nastic
50% (29)
41% (29)
47% (29)
Rosco Allen
55% (31.6)
69% (29)*
33% (24)

* two games played

First off, let's look at the most important player on this list. Chasson shouldered a heavy workload this year, but it really wasn't that much more than the time he got last year. In fact, he actually improved his numbers in February 2014. We know about the Memphis Meltdown, but overall in March he shot it better than he did in January. Also, his usage was basically the same the last two years, so the idea that he was carrying a heavier load this year is specious. Looking at these numbers in isolation, it's hard not to speculate about the health of #5 this year. That 33% was such a deep crater, when you consider the other five months studied in these two tables. It's hard for me to accept any of the alternate solutions. I don't buy that it was "just a slump" or that teams in February found some magical defensive concoction, but something happened.

Anthony's performance presents a more compelling case, because he did have more court time and took more shots this year than last, and we can't factor in that he typically had far more defensive responsibilities than Chasson. We know that what killed Chasson in February was a massive slump from behind the three-point line, while Anthony' s 2P% dropped significantly. We don't have shot charts to identify the location of Anthony's two-point shooting, but the drop-off in FTA/G strongly suggests that he stopped attacking the rim and getting shots in the paint (aka high percentage looks), which is what tired players normally do. Like Chasson, he experienced a surge in February of 2014, and in this case the minutes seem to clearly have mattered.

Furthermore, the Nasty Man's numbers fell off in February this year, as did the Cardinal's primary interior offensive weapon last year. Powell fell off in February, then picked it up in March, but look at the minutes played. Whether by foul trouble or by coaching decision, Dwight's numbers were limited in March, and his productivity enjoyed a bump. That was the same with Chasson, if you look at February 2014. Again, I didn't go back and contextualize the minutes he played in relation to the demands of any individual games, but the averages are still instructive. We can't say that the coaches intentionally limited his Feb. '14 minutes, but they went down and his production went up.

The first thing we'd want to do when comparing Feb. results is look at the competition. We'll use the Pomeroy ratings for convenience. In Feb. '14, Stanford played seven games and went 5-2 against the #'s 45, 64, 95, 183, 163, 15, and 45th ranked teams. The 45's were both Arizona State, against whom they split and the 15 was the home victory over UCLA. This year, Stanford went 3-3 against the #'s 37, 149, 8, 94, 113, and 115th rated teams. I'd consider these comparable schedules, so I don't buy that it was the uptick in competition this February that did the Cardinal in, especially since they lost to #94 on the list this season. They lost to the three best teams they played that month, and all were tournament teams, but the Feb. '14 slate featured three games against NCAA tournament competition (ASU 2X and UCLA) and Stanford went 2-1. Also, last year's UCLA team was far better than this year's team that won in Maples.

So it wasn't the competition. Maybe it was cohesion. The biggest discrepancy in these tables is in the minutes column. Look beyond Chasson and Anthony and you have two players in 2014 who received comparable minutes. This year, neither Nastic (due to fouls) nor Allen (due to injury) logged nearly as much court time, and maybe that's the big takeaway. In February, Stanford had a core of four players spending time on the court, while this year, the Cardinal struggled to find even three players it could keep on the floor consistently. We've talked about how Nastic's fouls meant more minutes for Brown and Randle. Remember, usage stats are based on the number of possessions used while the player is on the floor, not a percentage of the team's total possessions. In other words, the usage numbers in 2014 and 2015 for Brown and Randle are similar, but they don't necessarily equate to the total number of shots being equal.

For example, Anthony took more shots in fewer games this past February than he did last year. Chasson's shot volume essentially remained the same, and that's where consideration of the defense's tactics come into play. With Nastic off the floor, Stanford became an offense that was very easy to target, because it was clear that there were really only two players the Cardinal really wanted taking shots. We all know in the endgame that that number essentially shrunk to one. It's much harder to succeed when there is only one plausible threat on defense. Now, Stanford from last year wasn't markedly better overall, but the difference from last year and this year was ultimately only a couple games, and perhaps the most relevant are the home games against UCLA. Last year's Stanford team would not have made the tournament without that win, and this year's Stanford team didn't make it due in large part to failing against the Bruins.

Last year, Stanford got 75 points from its Core Four. This year, it only got 55 points from the four players in the table above. Both groups generated basically the same amount of shots (43 this year, 40 last), but there was a discrepancy in free throws. Last year's quartet generated 19 free throws, while this year's managed only 14. That's a relevant difference when you consider that this year was a two-point loss. Chasson got more free throws this year, but Anthony only managed four this year after eight in the victory. Huestis also got to the line six times. The biggest discrepancy though? It's one very obvious category.

In Stanford's 83-74 victory over UCLA last season, Randle was a spectacular 7-10 on three point shots. That game was played in late February, and we've already taken note that he played basically the same amount of minutes last year as compared to this year. This year, he shot 1-8 from the three point line. That's +18 in a victory compared to this year's loss. Now, was he just on one day and off the other? As a proponent of analytics, we have to acknowledge the small sample size here, but this game is being used as an illustrative example of a systematic flaw. UCLA's defense was rated fourth in the Pac-12 both last year and this, by the way. So I'm going to make this argument. Last year's Stanford team had more true threats on the floor. Maybe not much more, but on that day the Bruins had to account for Huestis, Powell, and Brown, who was essentially the same player in both games (18 pts in '14 vs. 19 in '15). This year, Randle was an easier cover because options three and four (Nastic/Rosco) were not as effective as Brown and Powell.

So where does that leave the fatigue theory? In my mind, it still plays a prominent role, but not the only one. The March numbers argue that if it were just an issue of a certain tipping point in mileage, than the numbers should have been worse in that month than in February. Things are just a bit more complicated than I'd like them to be. Certainly the overall issue of depth was a relevant one, but there's nuance even in that. Last year, Stanford played its Core Four almost exclusively. They played 150 of 160 minutes. This year's top four played 135 of 160 minutes. So if it were about fatigue, then the more rested group would have played more effectively. Except that's not what happened. 2015 Stanford was ultimately easier to guard in this game, and they were far too easy to guard in February and March. Last year's team wasn't night and day better, but night and day wasn't the margin by which Stanford failed to qualify. Fatigue, possible injury, less effective third (in that one game) and fourth (all season) options, all conspired to derail Stanford's offense. The rankings will tell you that last year's offense wasn't better than this year's, but ironically at a school where diversity is embraced as a core value, it was lack thereof that ultimately doomed the Cardinal's offense.

That and the defense, but that's a whole other story.....

The Bootleg Top Stories