|
|
|
Friday, December 8, 2000
Sorting E-mails & stats
By Joe Lunardi
Special to ESPN.com
|
Welcome to the first edition of "Box Score Banter." We'll keep this
going each week throughout the season. And come February, I fully expect this feature to become "Bracket Banter."
But, more on that later. In the meantime, please remember to include your
full name and location in future E-mails (jlunardi@home.com).
Let's get started:
First, I really like the idea for your new ESPN column. It's great when reporters cut through the rah-rah stuff and actually talk about what happened on the court in an objective way. I have always felt that a good point guard shouldn't take too many shots and that a point guard could very well have a great game and score no points. Most championship teams don't have high-scoring point guards (Wayne Turner, Anthony Epps, Derrick Phelps, Tyus Edney). Thus, I disagreed with all the TV commentators and journalists who declared Omar Cook the second coming of Isiah Thomas after his play in the Coaches vs. Cancer
tournament. Thus, can you write a column that shows a correlation
between shots taken by the starting point guard and wins?
I'll do that column, Andrew. For now, settle for this:
I agree with you philosophically that "less is more" with regard to point-guard play, provided, of course, that a team has other offensive weapons. Wake Forest's Robert O'Kelley comes to mind as a guy who really benefits his
team by playing off the ball. It's no coincidence that Wake has become a more balanced squad -- and therefore harder to defend -- with O'Kelley no longer being both the primary ballhandler and shooter.
Who would you rather have? Georgetown's high-scoring Kevin Braswell, a 36.8 percent
shooter who dominates the ball at the point? Or someone like Pepe
Sanchez of Temple, who never averaged double figures but controlled
games for John Chaney at both ends of the floor?
It might be more amenable to the NBA (that's where I had suggested it), but shouldn't basketball have a "+/-" stat like they do in hockey? That seems really important to me. I realize you can't create this stat
yourself, because you won't have the raw figures, but I did want to
suggest it. I think this is really a key stat to begin correlating
other stats with winning.
I have three thoughts on this excellent point:
1) I believe some teams keep this stat on the bench during games. I'm
checking with coaches and team managers on an informal basis to see if
"+/-" is indeed part of what all those clipboard-carriers do.
2) I'm planning a future column soliciting the manual keeping of this stat by volunteers for a sample of games later in the year. Start
sharpening those pencils ...
3) Finally, I've begun speaking with SIDs at various arenas around the country. With most courtside statistics now handled by computerized packages (which also chart substitutions), it seems to me that "+/-" could
be programmed into the final box score. If that actually happened, I
might be able to retire just as this whole thing is getting started!
Great to see you back on the internet. I forwarded Andy Katz those horrendous foul notes from the UWGB/Idaho game and I'm glad they found their way into your column. As always, I'm eagerly anticipating your
Big Dance projections.
Brian Nicol
Sports Information Director
UW-Green Bay Athletics
Thanks, Brian. For better or worse, the UWGB-Idaho matchup (Nov. 17) is no longer at the "top" of the leader board when it comes to most personal fouls called in a game this season. It was eclipsed late last
Saturday at Washington in a game, which also featured a season-high 87
free throws. Yuk.
Most Fouls Team
64, New Mexico State at Washington (Nov. 25)
62, Wisconsin-Green Bay at Idaho (Nov. 17)
60, Manhattan at Hartford (Nov. 28)
60, Quinnipiac vs. Howard (Nov. 25)
You raise the question in your column, "Doesn't it seem strange that there have been fewer one-point games than 40-point blowouts so far?" Well, I don't think that's at all strange. Consider that a one-point
game will only occur if the final margin is exactly one point; however,
a 40-point blowout will occur if the margin is anything greater than or
equal to 40 points. Teams foul at the end to get possession back if
they're down by one, thus temporarily eliminating those one-points
spreads. Also, at this early point in the season, there are few good
matchups (aside from the tournaments). Major conference teams are
playing their early season "warm-up" games and don't always want a big
challenge. When you have games like Wake Forest against Mount St.
Mary's, or Notre Dame against Sacred Heart, you shouldn't expect close
games. The margins get blown up, too, when weak shooting teams
desperately launch 3s because the game is out of reach anyway. I
don't think that it's so surprising to see 40-points blowouts this early
in the season. As for the one-point games, I don't have all the
numbers, but I'll bet if you include overtime games (which are basically
zero-point games at the end of the second half), you'd come closer to
matching the blowouts. If you include two- and three-point games (since
these are, after all, one-possession games), you'd find far more "close"
games in this age of parity than you would blowouts.
We agree and disagree, David. Even with the arithmetic difficulty of producing a one-point finish, I really believe we'll end up with more one-point outcomes by the end of the season than 40-point blowouts, which I chose as the margin of victory because it represents at least one point per minute. If you throw in OT and other one-possession-type games, the number of "close" outcomes will be far greater. This will be especially true, as you say, when conference play begins. As of Nov.
28, there have been 22 "blowout" (40-plus) games and 26 decided either by
one point or in overtime. Close enough to track later in the year, don't you think?
In light of your column, more statistics I enjoy reading about are: assists-to-turnovers, shots attempted-to-possessions minus offensive rebounds (rebounds are a different category), rebound margin and loose
ball recovery. I want to know who values the ball and who does not.
So do coaches (at least most of them). While I don't know how many benches track shots per possession, I know almost all track the percentage of possessions on which teams score, including sub-categories of how many times they score on "Play A" or "Play B." Many teams also chart charges taken, loose balls collected and other such "hustle"
stats. These figures just aren't public ... yet (that's why we're here!).
The other two items you mention -- assists-to-turnovers and rebound
margin -- are now published by most conferences.
In a recent column, you mention "points-per-shot" as a useful statistic to gauge a hoop team's efficiency. I find that "points-per-possession" is a slightly more accurate measure. Evaluating the number of points
divided by the number of possessions (or trips into the halfcourt) factors turnovers and steals into the equation, whereas points-per-shot does not. I suppose the appeal of "points-per-shot" is that it can be easily calculated from the box score, while points-per-possession must be tracked. The first time I heard of this stat was back in the late
'60's; Ray Mears, coach of Tennessee was an advocate of it. As a
diehard Vandy fan, I always root against the Vols, but I don't mind
picking up useful ideas from anybody as long as they work.
I'd heard of points-per-possession as a team measurement, and it is something that maybe we can track when "volunteers" are organized for a sample of games later in the year. My reference to points-per-shot was
as an individual statistic. It factors in shooting percentage, two-point vs. three-point frequency, etc. What I like most about it is
that it weeds out players who are high scorers only because they take so many shots, arguably hurting their team in the process. You'll be reading a lot of points-per-shot stuff in this space.
Just ran across a number that I wanted to pass along to you about my favorite college coach. After leading his Cougars back from five points down with 1:01 to play against Chattanooga, John Kresse of the College of Charleston is now 20-6 in one-point games as a head coach (the game, an early-season conference road game, ended 57-56). While it may not be
a record, 77 percent isn't bad.
Chopper Johnson
Charleston, S.C.
If you flipped a coin 26 times and it came up heads 20 times, that would be pretty amazing. Plus, no one would be trying to block your shot! Thanks for the great note.
As a college hoops fan, I'm, of course, thrilled to be at the dawn of a
new season. And doubly thrilled, being an Illini fan with all sorts of
fun games coming up. And, well, 2.45 times as thrilled, being an
epidemiologist/stat geek, at the prospect of your new column. Nice job
using just a small bit of data for some fun stats.
A couple of ideas for future stats:
Bruise-O-Matic: Ratio of Team A's fouls to Team B's shots taken. Even
though all fouls aren't shooting ones, this might give a rough measure
of what teams rely on the hack as a primary defensive weapon. This
would, of course, be better if you could get some count of shooting
fouls.
You can't score if you don't shoot: Percent of times a team wins when it takes more FG (or 3PT) shots than the opposition. Or the percent of time Team A wins when it take 1-2 more shots, 3-4 more shots (and of
course 1-2 less, 3-4 less, etc.). Is there a point at which the
differential becomes crucial?
Which glass should I clean first: How do offensive rebounds (or offensive rebound differential between two teams) affect winning as compared to defensive rebounding? Are there teams that dominate
O-rebounds but not D-rebounds, or vice versa?
When to hoist it: Is there an optimum ratio of two-point to three-point shots taken that delivers the win? What are the average
and extreme ratios?
This is kind of fun. Wish I could help out with your analyses. Let me
know if you need another number cruncher!
Dave Kaufman,
Washington, D.C.
Dave, you are clearly our kind of guy. I'll be in touch!
Just wanted you to know that someone out there is appreciating this statistical stuff. Now, I doubt this column will become as popular as Rob Neyer's column simply because college basketball isn't as popular as baseball, but hopefully you'll continue to write one column a week throughout the season and it starts to catch on. Good luck! This basketball nut appreciates it.
Here's one suggestion for something to investigate. Which home court gives the best home-court advantage in college basketball? Maybe you can compare point differentials at home and on the road?
The first one is easy. We'll run a list at some point. I'll look into the second, at least for a few major conferences. To me, the best
sample would be a round-robin within a single conference in which every team is playing at every other team's home court once (it doesn't happen as much anymore as you think). That would be a closed sample in which
conditions are theoretically equal for each conference member. Good
idea.
Joe Lunardi is a regular in-season contributor for ESPN.com. He is also
contributing editor of the Blue Ribbon College Basketball Yearbook,
www.collegebaskets.com. Write to Joe at jlunardi@home.com.
|
|
ALSO SEE
Chart wrap: Joe Lunardi
Box scores and more
|
|
