Breaking Down Wednesday's UCLA-Wash. Thriller

UCLA has shown a flair in this young season for the dramatic, beginning with an opening-weekend loss to Nebraska (15-13 in the fifth). Wednesday night, the Bruins played another epic match, losing in five at Washington as the Huskies won all three of their games by the minimum two-point margin (22-25, 30-28, 19-25, 28-26, 16-14.)

Despite entering the UCLA match undefeated, Washington was largely untested (a match against Purdue being the exception). Now, however, the Huskies have shown that they belong among the nation's elite.

The UCLA-Washington showdown featured a match-within-a-match aspect, namely a battle between the teams' slugging outside hitters. For the Huskies, it was sophomore Krista Vansant going against Bruin seniors Tabi Love and Rachael Kidder.

Based on the play-by-play sheet (which I accessed by going to the Huskies' schedule page and then clicking on the archived Gametracker for the UCLA match), I created the following chart of what I thought were the key statistical indicators (you may click on the chart to enlarge it). All of my tabulations from the Gametracker play-by-play matched the official box score, with the exception that I counted 30 kills for Vansant and the box score on the UW said she had 31. (Attacked balls kept in play are not listed on the play-by-play sheet, hence my focus on kills and errors.)

UCLA likely would have gone up 2-0 in games but for amazing U-Dub performances on two counts. Not only did Vansant record 10 kills with only 1 hitting error; the Huskies also blocked 6 Bruin spike attempts for immediate points (2 on swings by Kidder, 1 on a swing by Love).

Then, despite fifth games being shorter than the previous games (up to 15, instead of 25), Vansant amassed 7 kills (with 2 errors). She ended up hitting .448 on the night. Love, though hitting nearly error-free for most of the match, struggled a bit in Game 4, with 4 errors (3 balls hit astray and 1 attempt blocked). She ended up hitting .266. Kidder was a little more uneven, finishing up at .196. 

The Huskies return to the court tonight, hosting USC.

Oregon Sweep of USC Highlights Weekend

Two nights after defeating No. 1 UCLA to move to 13-0 on the season, the USC women came crashing back to earth. Oregon came into USC's Galen Center on Friday night and swept the Trojans, 25-19, 25-15, 25-23. The Ducks, now 11-0, must be considered among the nation's elite at this point in the season.

Against the Bruins on Wednesday, the Trojans seemingly sided-out at will, winning points on 71% of UCLA's serves. On Friday, the Ducks took siding-out to a new level against the Trojans, reaching 77% effectiveness. In the first two games especially, Oregon's serve-return performance was spectacular, attaining side-out rates of 84 and 81 percent (box score).

The Ducks outhit the Trojans, .357-.189. Three Oregon players recorded superb hitting nights: outside hitter Liz Brenner, .577 (16 kills, 1 error, 26 attempts); OH Alaina Bergsma, .438 (18-4-32); and middle blocker Ariana Williams, .375 (8-2-16). UO attempted 115 spikes, only 14 of which resulted in errors (6 hit out-of-bounds, 8 blocked by USC for immediate Trojan points). Of Oregon's 101 non-error attacks, USC dug up only 37 of them (37%). In contrast, 'SC dug much more effectively against UCLA, retrieving 46% of the Bruins' non-error attacks [58/(145 - 20)].

***

Across town on Friday, UCLA came back from its USC loss to rout Oregon State, 25-14, 25-18, 25-17. Clearly, the Beavers were not the same team that had beaten then-No. 2 Penn State on September 8.

Against Oregon State, UCLA outside hitters Tabi Love and Rachael Kidder recorded nearly identical hitting lines. Each hit .333 on 27 attempts, with Love amassing 12 kills and 3 errors, and Kidder, 14 kills and 5 errors.

No. 2 USC defeats No. 1 UCLA

No. 2 USC defeated No. 1 UCLA last night, 28-26, 25-20, 24-26, 25-17. The match kicked-off a series of Wednesday-night telecasts on ESPN-U.

In reading the game article on the Trojan athletic website, I was impressed with the writer's attention to putting statistical figures in context. Many volleyball articles mention the players who had the most kills. However, merely stating the number of kills does not tell us how many attempts were required or how many errors the same hitter also had. As shown in the following brief excerpt from the article, the writer always made sure to accompany players' kill totals with their numbers of errors and attempts, and hitting percentage on the night.

USC women's volleyball senior opposite Katie Fuller knocked down 21 kills (5e, 36att) and posted a .444 hitting percentage... Fuller's 21 kills matched a career high, but the Trojans got a major contribution from freshman outside hitter Samantha Bricio who also matched her career mark with 19 kills (6e, 50att, .260) to go with three service aces (23.0 points). Sophomore setter Hayley Crone notched a new career best with 53 assists and added four kills on second contact (0e, 5att, .800) and six digs.

HTML Tables

UCLA's performance was hardly terrible. The Bruins sided-out, for example, at respectable levels of 62 and 63% in the first two games/sets, despite losing them both. The problem was that the Trojans did even better (66 and 80%, respectively). In none of the four games was USC's side-out rate lower than 65% (box score).

The Trojans also outhit the Bruins for the match, .311 to .297. For UCLA, Becca Strehlow's quick-sets to Mariana Aquino were working early. Aquino hit .385 on the night, with an 8-3-13 line for kills, errors, and attempts. As seen in the side-out numbers, the UCLA serve didn't challenge 'SC much. The Bruins' 11 service errors (to only 1 ace) suggest UCLA may have adopted a high-risk serving strategy to combat the Trojans' effective serve-receipt.

Also last night (and televised on Fox Sports' new Pac 12 Network), Stanford swept Cal.

Minnesota and Texas Split Matches in Austin

There's a new excitement surrounding University of Minnesota volleyball, with Hugh McCutcheon, who guided the U.S. men and women to, respectively, gold and silver in the last two Olympiad, taking over for retired coach Mike Hebert. The No. 14 Golden Gophers traveled down to Texas this past week for matches Thursday and Friday nights against the No. 4 Longhorns. The visit reciprocated a Texas trip to the Twin Cities last year, when the Gophers swept both matches.

This year, it was a split, with each match going four games (or sets). The Gophers won the opener by scores of 25-22, 25-13, 27-29, 29-27, whereas the Longhorns bounced back the following evening to win 25-20, 20-25, 25-22,  25-21. Here are the box scores for Thursday and Friday nights. I like these two-match series, as the added data provide a more reliable picture of how the teams are doing than the usual single match.

In Thursday night's Gopher win, offensive balance was the key. The following table shows the hitting percentages achieved by five leading Minnesota spikers, with their numbers of swings in parentheses. Two newcomers complement the established Gopher trio of Katherine Harms (senior outside hitter), Tori Dixon (junior middle blocker), and Ashley Wittman (junior OH). Daly Santana is a frosh OH from Puerto Rico, whereas Dana Knudsen (MB) is a senior transfer from Santa Clara. As the table shows, all five hit .276 or better in the series opener at Texas; Knudsen had 14 attempts, but all the others had 25 or more.

	Thursday	Friday
Tori Dixon	.286 (28)	.320 (25)
Katherine Harms	.319 (47)	.357 (28)
Dana Knudsen	.286 (14)	.000 (12)
Daly Santana	.276 (29)	.030 (33)
Ashley Wittman	.400 (25)	.000 (34)

HTML Tables

Friday night was a very different story, however. Dixon and Harms continued to hit well for the Gophers, but the Longhorns shut down Knudsen, Santana, and Wittman. One apparent reason for this is that Texas dominated the blocking Friday, by a 15.5-7 margin (the teams were essentially even on Thursday night, Texas 12, Minnesota 11.5).

For Texas offensively, it was the Haley and Bailey show, as outside hitters Haley Eckerman and Bailey Webster took the largest shares of the team's swings. The problem was that only one of the two could get untracked each night, Webster hitting .368 on Thursday and Eckerman, .289, on Friday.  The following table shows the same statistics as in the earlier one for Minnesota.

	Thursday	Friday
Haley Eckerman	.080 (50)	.289 (45)
Bailey Webster	.368 (38)	.100 (50)

HTML Tables

On Thursday, Eckerman and Webster took 60% of the Horns' hitting attempts (88/147), with MB Khat Bell hitting .292 on 24 swings. Bell was held out on Friday; she is recovering from a torn ACL last season and Coach Jerritt Elliott evidently thought it prudent to rest Bell rather than have her play on back-to-back nights. Eckerman and Webster took on more of the hitting load Friday night, their combined 95 swings constituting 71% of the Longhorns' 134 total attempts. 

Another apparent factor in Minnesota's Thursday win was the Gophers' digging advantage. Looking at the proportion of possible opponent attacks dug up (i.e., digs divided by opponents' non-error attacks), we see the following:

Thursday
Minnesota 67 digs / (Texas 147 spike attempts -  22 hitting errors) = .536
Texas 46 digs / (Minnesota 147 spike attempts - 23 hitting errors) = .371

Friday
Minnesota 51 digs / (Texas 134 spike attempts - 20 hitting errors) = .447
Texas 54 digs / (Minnesota 141 spike attempts - 29 hitting errors) = .482

On Thursday, the Gophers dug up more than half of the Longhorn hitting attempts that were not hit out-of-bounds or blocked for immediate Minnesota points. The following night, Minnesota was just under 50% in this department. Texas performed well in this area Friday night (48%), but not Thursday.

Texas coach Elliott did some shifting of players from their usual positions on Thursday, and then of course went without Bell on Friday. Perhaps the Longhorns will show greater match-to-match consistency as they settle into a regular lineup later in the season (if they do).

Texas Tech Coaches Share Team's Internal Statistics

I am pleased to announce a major new development here at VolleyMetrics, involving a faculty-athletics collaboration at Texas Tech University. Head Coach Don Flora and Assistant Coach Jojit Coronel have agreed to share with me the team's internal statistical data, which are recorded with Data Volley software.

Today, I will present my first statistical analysis of these data. At this point, the Red Raiders have completed roughly half of their non-conference schedule, compiling a perfect 8-0 record thus far. On Tuesday, the Raider coaches sent me a batch of printouts. Each printout covers one match. I see one of my roles, therefore, as testing for trends across multiple matches. Although each printout contains information on Texas Tech and the team's opponent, I am only using data on the Red Raiders today.

For starters, I will examine the side-out rates and points won on serve for each of Texas Tech's six rotations on the court, to see if there are any noteworthy statistical patterns after eight matches. In Rotations 1, 2, and 3, the setter is in the back row (right, center, and left), giving the team three hitters up front. When the setter is in the front row (Rotations 4, 5, and 6), the team only has two hitters. Having three available hitters presumably is more advantageous than having two, in terms of offensive options and greater potential to keep the blockers guessing. Coach Coronel suggested an alternative line of argument to me, however. Three hitters can lead to a "cluttered" front row, with each having less room to maneuver.

Back in 2009, I attempted to compare the effectiveness of Texas Tech's six rotations in the team's home match vs. Texas A&M, compiling statistics myself from the stands. That analysis obviously included a lot less data than what I have available this season, but some readers may be interested in the diagrams I created back then of the players in each rotation (click here for my 2009 analysis).

With these ideas in mind, I used the 2012 data to look first at the average side-out rates of the six rotations. These mean levels are shown in the following graph as red circles.

As can be seen, Rotation 5 achieved the highest side-out rate of the rotations (77%), whereas most of the other rotations' side-out rates were around 60%, give or take a few percentage points. However, as shown via the little grey shapes above and below each red circle, each rotation's success at siding-out showed a great deal of scatter across the team's matches. Due to this scatter and the small number of matches, the differences between the six means were not statistically significant. However, as data from more matches pour in, we'll see if Rotation 5 continues to side-out the best.

The other statistic I examined is the counterpart to siding-out, namely how proficiently the team wins points on its own serve in the different rotations. Of course, if whoever serves in a given rotation is a great server and the team garners a sizable share of points through aces or serves that take the opponent out of its offense, the location of the other players in the rotation doesn't really matter. On all other plays, however, the rotation could well matter (e.g., who is in the front row to try to block the opponent's attack).

The mean percentages of points won on Texas Tech's own serve in Rotations 1 through 6, respectively, across the eight matches, are as follows: 44%, 38%, 45%, 53%, 38%, and 50%. Again, there are no statistically significant differences between these values. On a purely descriptive basis, however, Rotation 5 was the team's least effective when serving, in contrast to its performance on serve-receipt.

Over the next few weeks, I'll be presenting additional new analyses from Texas Tech's first eight matches. Then, as I receive new sets of printouts, I'll analyze those. I hope you'll enjoy this exciting new feature on VolleyMetrics.

Women's College Round-Up Aug. 31-Sept. 1, 2012

No. 4 Penn State hosted two other top-10 teams in the AVCA poll, No. 2 Texas and No. 9 Stanford, along with No. 18 Florida, this weekend in the Big Four Classic.

The Nittany Lions came away with the championship, outlasting Stanford in a five-set match Friday and then sweeping Texas on Saturday.

In the Penn State-Stanford match, four of the five games were decided by the minimum two points (the Lions' score is listed first for each game; 25-13, 25-27, 25-23, 23-25, 15-13). It was not a big-hitting match, on the whole, as the Nittany Lions bested the Cardinal .231 to .173 (box score). There were some notable individual hitting percentages, however.

For Penn State, sophomore outside-hitter Nia Grant (9-0-13, .692) and junior middle-blocker Katie Slay (10-0-21, .476) each turned in error-free hitting performances against the Cardinal. Teammate Ariel Scott (junior right-side hitter) took a whopping 85 swings (39% of the team’s 216 attack attempts), registering a .165 percentage. Soph OH Morgan Boukather paced Stanford (11-2-21, .429).

In the Penn State-Texas match, the Longhorns' side-out percentages in the three games -- 63%, 54%, and 75% -- really don't look like those of a team that was swept. The only problem is that the Lions' side-out rates were 72%, 63%, and 86% (box score). Frosh OH Megan Courtney aided Penn State with an (8-0-15, .533) hitting line.

Florida, which lost both of its matches (to Texas and Stanford), has lost a lot of top seniors from last year, including setter-hitter Kelly Murphy. However, one player who may lead the Gators back into elite company is frosh OH Ziva Recek from Slovenia. She hit .475 against the Horns and .278 against the Cardinal.

***

Yesterday, I attended Texas Tech's match against Northwestern State (Louisiana), as the Red Raiders held off a surprisingly tough Demon squad in four games to go 7-0 on the season.

Making the headlines for Texas Tech at this early point in the season is libero Rachel Brummitt, a sophomore transfer from Radford. She is averaging 4.92 digs per game, second in the Big 12 to Kansas's Brianne Riley, whose average is 5.57.

A dig is defined as "when a player receives an attacked ball and keeps the ball in play" (see  the AVCA's "Making Volleyball Statistics Simple" in the links section to the right). After a ball is dug, a few different things can happen, such as the digging team running its offense "in system" culminating in a spike attempt or the digging team having to bump the ball back over to the other team, which gets a "free ball" attempt.

During the recent Olympics, I was impressed by how the Italian men were able to dig a lot of spike attempts by the U.S. and transition quickly into running their offense. Knowing ahead of time about Brummitt's digging prowess, I decided to keep an eye out for digs leading to an in-system attack, a new statistic I'm tentatively calling DLISA. The way I'm conceiving DLISA in my mind, the opponent's attack must be hard-hit and/or well-placed in order for the digger to be eligible for a DLISA.

I know there already are grading systems in place, in which raters can study videotapes and assign a quality score on each dig, for example, from 0 (passing error) to 5 (pass enabling multiple attack options). However, I'm simply looking for a way to augment the "dig" statistic reported in box scores with an extra designation for really good digs, hence DLISA.

Brummitt was credited in the box score with 21 digs vs. Northwestern State. I (unofficially) credited her with 4 DLISA. I'll have to watch a lot more matches to get an idea of what a high number of DLISA digs is for one player, but yesterday's match is a start!

A scientific ideal is that two judges, watching the same play, will have a high level of agreement in awarding credit for a particular kind of play (such as a DLISA). Scoring decisions are sometimes made unilaterally, however. In baseball, for example, there is one "official scorer" who decides between hit and error on a play.