As reported by DCBruins ("Putting a Dagger Into a Yahoo Hit Piece on UCLA Alumni and Kareem"), college basketball writer Jeff Eisenberg launched a defense of Steve Alford's coaching ability following critical comments by UCLA legend Kareem Abdul-Jabbar about the state of UCLA basketball.
Eisenberg writes that criticism of Alford's teaching ability is unwarranted since Alford's teams have improved throughout the course of the season. That sounds like a reasonable argument, but is it true that "UCLA has improved steadily from November to March in both Alford's two seasons?"
Although Eisenberg insists that the Bruins have "improved steadily" in both of Alford's seasons at UCLA, simple plots of the Bruins' winning percentage over the course of each season shows exactly what you'd expect to see: winning streaks and losing streaks, not a steady progression. However, since Eisenberg's defense of Steve Alford's teaching ability doesn't depend on whether or not the Bruins' progress is steady, I'll disregard that element of his argument and focus instead on the crux of his claim; i.e., that UCLA improved from the start of the season to the end of the season.
Before looking at the evidence, it's essential that we define "improvement" as something measurable. Eisenberg's claim is absolutely meaningless if "improvement" refers to something which relies on subjective judgments and doesn't take into account baseline improvement that any competent coach should develop in his team. For example, when a new group of players develop better teamwork over the course of the season, that should be regarded as the minimum standard for any collegiate basketball coach, not as evidence of an elite coach.
The traditional method for gauging improvement is to examine changes in winning percentage over time. However, since we are interested in looking at macro changes from the beginning of the season to the end of the season rather than micro changes from game to game, the simplest way of testing Eisenberg's claim is to compare UCLA's winning percentage from the first half of the season to the second half of the season. If, as Eisenberg asserts, UCLA improves from November to March, then the winning percentage from the second half of the season should be higher.
As you can see from the chart above, the facts don't support Eisenberg's claim. In Alford's first season at UCLA (2013-14), the team's record was worse in the second half of the season. In Alford's second season (2014-15), UCLA's winning percentage was the same in the second half of the season as it was in the first half. In no way does that qualify as improvement from November to March.
In fact, if we break the season in to sixths in order to get a bit more detail about how the Bruins' perform in various segments of the season, what becomes apparent immediately is that aside from the cupcake part of the schedule (the first sixth of the season), Alford's UCLA teams pretty reliably win about 2 out of every 3 games throughout the remainder of the season. Of course, that's precisely what Fox71 has been telling us for most of the past two seasons.
Simply put, an examination of UCLA's record does not support Eisenberg's claim that "UCLA has improved steadily from November to March in both Alford's two seasons..."
I imagine that those who think that Alford is a swell coach might object to the fact that the early part of the Bruins' schedule tends to feature lightweight opponents and thereby creates the appearance of a team playing well. Therefore, they might argue that it's unfair to measure "improvement" from the start of the season. There are a couple of problems with this argument.
The first is that this article a fact check on Jeff Eisenberg's claim that "UCLA has improved steadily from November to March in both Alford's two seasons." If Eisenberg didn't want to include UCLA's early season performances, he should have chosen his words more carefully. In any case, Eisenberg wrote the following in response to a letter from Fox71:
"Alford's UCLA teams have both improved as the season went along..."
Since he chose to repeat his original claim, I'm unsympathetic to arguments that he meant something other than what he wrote.
The second problem is that as the plot above shows, the Bruins' don't show improvement in their winning percentage even if the first sixth of the season is ignored. Or even if the first half of the season is ignored.
To find improvement from a specific starting point other than the first game, you have to resort to cherry-picking. To avoid that error, let's look at how the Bruins have fared in just the Pac-12 portion of the season. If Eisenberg's claim is correct, then we should find improvement in UCLA's winning percentage from the first half of the Pac-12 season to the second half.
As the chart above shows, the 2013-14 Bruins showed no improvement from the first half to the second half of the conference season. The 2014-15 Bruins did improve--by one game, the minimum possible--from the first half to the second half of the Pac-12 season. Even so, a one-game improvement over two seasons is not at all impressive. Moreover, the facts still contradict Eisenberg's claim that the Bruins improved in "both Alford's two seasons."
Finally, for what it's worth I've included two charts showing the game-to-game changes in the Bruins' winning percentage for both of Alford's seasons at UCLA.
There is simply no way to look at the chart for Alford's first season (2013-14) and conclude that the team improved during the course of the year.
It's tempting to conclude that UCLA did improve during the latter half of 2014-15. That's because our eyes are drawn to the low point of the plot (game 15). However, using that point as the starting point for gauging improvement is a classic example of cherry-picking. After all, if you choose that point, you're essentially saying "Look! UCLA's winning percentage improved from the point at which it was the lowest!"
It is fair, however, to ask if UCLA's winning percentage improved after certain important events. For example, UCLA's worst loss of the season came against Kentucky (game 12). Did UCLA's winning percentage improve after that? The answer is NO. Did UCLA's winning percentage improve from the beginning of the Pac-12 season to the end? The answer is NO. And in any case, if we go back to the wording of Eisenberg's claim, it's impossible to conclude that UCLA "improved...from November to March."
Jeff Eisenberg's claim is wrong. The evidence does not support his assertion that "UCLA has improved steadily from November to March in both Alford's two seasons."