Texas vs. Rice Gameflow (Updated)

by GhostofBigRoy on Dec 2, 2009 9:36 AM CST in Texas Longhorns Basketball Comment 9 comments

Gameflow from the Texas/Rice game on Sunday. I haven't worked much with Excel, so I wasn't able to use +/- symbols because I couldn't get the program to stop adding and subtracting. If anyone knows how to get that to turn off, I would appreciate it. Otherwise, hopefully it's readable and provides some insights. Also, if anyone has any thoughts on the best way to determine the best five-man groups for the team, that would also be helpful.

Previous Gameflows:

UC-Irvine

Western Carolina

A few thoughts:

In a game that the Longhorns won by 18 points, any player with a negative +/- in the game didn't add much to the team. Throwing out Matt Hill, who ended up with a -2 after one minute, and Shawn Williams, who played two minutes, the three players who come out looking poorly in this metric are Jordan Hamilton at even, Justin Mason at +1, and Gary Johnson at +2. It's difficult to make direct correlations between the play of an individual to the results of the team, but the first two players listed aren't a huge surprise -- Hamilton shot 3-9 from the field and still failed to make entry passes to the big guy down low, while Mason missed both of his field goals and three of his four free-throw attempts. It's a broken record now -- he just doesn't add anything offensively. For Johnson, it's harder to figure, though he did only have one rebound in his 16 minutes and turned the ball over twice.
James (+21), Bradley (+18), and Balbay (+17) all had the best games by this metric. Most surprising of these three is James, who shot poorly from the field, but rebounded well and didn't turn the ball over. For Bradley, it was probably about playing good defense and playing efficiently, as he shot 3-7 from the field, 2-2 from the line and had four assists and no turnovers. Balbay is interesting as well, as he didn't score much and only had one assist, but rebounded extremely well and probably helped the team by pushing the ball in transition and playing good defense.
Despite what is seemingly a bad combination in Balbay and Mason, the two played five different stretches together and finished +4, with only one of those stretches a net negative for the team at -1. Something to keep an eye on, although the fact that neither can shoot and the amount that tangibly affects the spacing on the court probably trumps this data.
The amount of substituting that Rick Barnes does may also affect the data, as the longest that any group was on the floor together was the nearly three minutes that Brown/James/Pittman/Hamilton/Mason were on the court and registered a -5, by far the worst stretch of basketball during the game, when Rice climbed back into it, cutting the lead 12 to seven and then to five when Bradley replaced Mason. Does substituting so frequently hurt the rhythm of the team because the players on the court don't have time to establish a good dynamic and decide where they want to go with the ball? There's certainly accountability, but there may be side effects as well.
It's not surprising in some ways that the group of Brown/Bradley/James/Pittman/Hamilton registered a +4 in the first half, as that's probably the best combination on paper right now for the Longhorns, although that group also registered -2 in their appearance together in the second half, when Hamilton earned himself a spot on the bench for the rest of the game by jacking a step-back three pointer after making a three the previous trip down the court.
Context plays a factor here, as the needless fouling by Rice at the end of the game resulted in a +5 and +4 for the players on the court at the end of the game, despite not having to do anything offensively other than hold onto the ball long enough to get fouled and make free throws, which James and Brown did at a high rate -- 7-8 in the last minute. However, those two groups did also help hold Rice scoreless over the last 4:17 of the game.

[Update]: Here are some other charts from StatSheet for the Rice game. I'll try to post them after every game.

NCAA Basketball Stats

As a note, the last chart has J'Covan Brown's numbers listed as Varez Ward, hopefully they will get that changed.

0 recs | Comment 9 comments |

Story-email Email Printer Print

Comments

Excel answer

I don’t know how to send you a personal message… but to get the plus/minus to work in excel you can do several things, the easiest is to type a single quote mark (‘+ 2). That forces excel to treat the following cell content as text, and you shouldn’t have a problem.

by scottmcc on Dec 2, 2009 9:55 AM CST reply actions

Ah, thanks.

The Excel help feature was decidedly unhelpful.

by GhostofBigRoy on Dec 2, 2009 10:14 AM CST up reply actions

Another way

Highlight the cells you are wanting to use as only text and right click. Select Format Cells and under the Number tab, click Text. That will set all of the selected cells as text and will read exactly as they are entered.

by afaeguy on Dec 2, 2009 10:27 AM CST up reply actions

Hamilton mystery

There are moments when Jordan Hamilton looks more poised and smooth than anyone on the court, when his talent and potential are apparent. But mostly, he looks uncomfortable, kind of hopping around the 3 point line, waving for the ball, making token attempts to move. Still, he hits a good number of shots.

The team as a whole has had moments when everyone looks uncomfortable, reminiscent of many games last year, when the Horns would experience stretches of indecision and strings of what appeared to be bad choices on offense. James, and to a greater extent, Johnson, seem to be forcing shots, trying to create in a way that only seasoned scorers can. Brown takes a good share of shots (with more success than James and Johnson, it seems) and Pittman has been resoundingly good.

So I wonder, with so much offense going through Pittman (as it should), and with the older forwards taking lots of shots without fear of getting yanked immediately, and with J’Covan feeling free to shoot — where does this leave Hamilton in the set offense? By “set” I just mean, when there isn’t a fastbreak. The second chance points will go to Pittman and James, and a little to Johnson.

With these, and other elements still taking shape, what will Hamilton look like by, say, the time Big 12 play gets going?

by Kool Hand on Dec 2, 2009 10:27 AM CST reply actions

Also, Thanks for the Gameflow GoBR

I’ve never taken the time to learn about +/- any more than knowing that it’s one metric, and that a player with good numbers is generally doing good things.

Having watched most of the basketball games this season, these graphics help me digest a little bit more than I’m accustomed to as a basketball watcher. Gives me, somewhat of a bball novice having never played or attempted to learn the intracacies, some good things to chew on during and after the games.

by Kool Hand on Dec 2, 2009 10:35 AM CST reply actions

Excel

My suggestion:

Create a new column where you keep track of the lineups, giving each one a different number. Using “if” statements (under the functions), you can create a scenario that counts the score variance for that player (will need a column that calculates the variance). As an example, the score variance was +4 for lineup 4. Damion James plays for lineups 1-2, 4-5, and 7 (you’ll probably want the lineup numbers to remain constant for each game). Under James, the formula would be “if(or(”cell the includes the lineup"=1,cell=2, cell=3,cell=5,cell=7)),cell that is the variance,0).

To calculate the best lineup, you would create a second area where you sum the value of the various lineups for the game. Using a max function you’ll be able to easily identify the best lineup.

Hopefully this isn’t too confusing. Keep an eye out for an email and you can let me know if you need more help.

by Bullard's Boy on Dec 2, 2009 7:31 PM CST reply actions

My bad

I’ve been either too busy or lazy to make the game flows. I had already made the Iowa and Pitt gameflows. I planned on making the Rice one and posting them all in one post tonight.

I’ll go ahead and finish the Rice one out of guilt. Do you plan on making these the rest of the season? Just out of curiosity, how much time did it take you to make the whole thing?

For the bigger upcoming games (UNC, UConn, MSU, etc.), I planned on making a gameflow for both teams. We all know that the +/- of our guys is just half the story. If we have one unit that performed terribly, maybe it has more to do with the other team having a unit that was flat out dominating.

by goingforthecorner on Dec 2, 2009 9:45 PM CST reply actions

No problem.

I had been thinking during the off-season about whys to improve the statistical take on basketball here and this was a great idea.

If you can post the Iowa one, that would save me the trouble. I also had a couple people email me about doing the five-man unit analysis. I can forward them to you or we can talk about that.

The first one took me a couple of hours because I didn’t really know what I was doing and the formatting took me a while. The second one went fairly quickly — about an hour and a half or so.

The best thing in the future might be to just send me an email and let me know if you aren’t available to do it and if not I can probably pick up the slack.

It would definitely be interesting to see a Gameflow from the other team, so feel free to do that. I also included the three charts from Statsheets my embedding the html they provide at the site. Or not, it doesn’t matter, but I find them interesting, particularly the Four Factors and Player Impact.

by GhostofBigRoy on Dec 2, 2009 11:04 PM CST up reply actions