Mythbuster #4: “You need a great guard to go deep in the dance.”

Over the next two weeks, you’ll hear a lot of noise about how important having a great guard is to March Madness success. Someone will invoke Trey Burke and show his shot against Kansas. You’ll hear Kemba Walker’s name. They’ll replay his ankle-breaking, step-back jumper against Pitt.

And you’ll almost be convinced. No one, however, backs up the “Great Guard Claim” with any numbers. I looked at the issue from a few different angles:

  1. How has the frontcourt/backcourt scoring balance of the tourney field changed over the 29 years of the 64-team era?
  2. Do backcourt-dominant, frontcourt-dominant or balanced scoring teams fare the best in the dance—and have their fortunes changed over the years?
  3. Which kind of teams have fared better in the last decade—those with All-American guards or those with All-American forwards and centers?

Let’s look at question 1. The evidence is clear that the scoring balance of tourney teams has tilted much more in favor of guard-oriented squads. I divided the 29 years of the dance into three eras (1985-1993, 1994-2003 and 2004-2013) and calculated the percentage of 1-12 seeds that got less than 40% of their points from guards (frontcourt-dominant), 40-60% of their points from guards (balanced) and more than 60% of their points from guards (backcourt-dominant). Here’s what I discovered:

2014_Guard1

The tourney field is comprised of a much smaller percentage of frontcourt-dominant teams than it was in the first nine years of the tourney. Where 38.1% of 1-12 seeds relied on their guards for less than 40% of their points between 1985 and 1993, over the last decade that percentage has plummeted to 17.5%. Meanwhile, almost the exact opposite dynamic has occurred with backcourt-dominant squads. In the early era of the 64-team dance, it was rare to find a team relying on guards for more than 60% of its points (only 15.5% of 1-12 seeds). Since 2004, 38.5% of the tourney field leans on guard scoring that much.

So if you asked me, “Are tourney teams relying more on guards for their scoring?” I would have to say, “yes.” But that’s not what the pundits are claiming. They’re contending that guard-dominant teams (or those with great guards) actually perform better in the tourney. Is this true?

If you look at the overall PASE numbers of the three scoring balance classes, you would have to conclude that guard play isn’t nearly as important as frontcourt dominance. Since 1985, teams relying on guards for less than 40% of their points are solid +.107 PASE performers. Balanced teams getting 40-60% of their points from guards are -.031 PASE underachievers. And teams crutching on guards for more than 60% of their scoring are the biggest underperforming class of all (-.050 PASE).

Of course, we just saw that the composition of the tourney field has changed dramatically over the last three decades. It’s reasonable to ask if we’ve seen a corresponding change in the fortunes of the three scoring balance classes. So I did a PASE analysis of <40%, 40-60% and >60% backcourt scoring teams for each of the three eras. Here’s what I found:

2014_Guard2

You see what I’m seeing? Frontcourt-dominant teams are actually bigger overachievers now than they were in the early years of the dance. And backcourt-dominant teams fall short of expectations at a higher rate. In fact, teams relying the most on guards for scoring have never been overachievers in any of the three eras, nor have balanced teams, though they’ve been playing much closer to expectations of late. Only the teams relying most on their frontcourt have consistently beaten seed-projected win rates.

You might argue that PASE isn’t the be-all and end-all of tourney success; that Final Four runs and championships are the best measure. There again, the numbers favor the frontcourt-dominant squads. Overall, 10.6% of 1-12 seeds getting less than 40% of their points from guards reach the Final Four, and 3.6% cut down the nets. Compare that to balanced scoring teams (7.5% Final Four and 2.1% championship advancement rates) and guard-heavy teams (7.5% Final Four and 0.6% championship rates.

Some might also quarrel with the whole premise that analyzing teams by percentage of scoring from guards doesn’t necessarily equate to “having a great guard.” After all, a team could average 80 points a game, average 40% of its points from guards, and have one hotshot averaging 30 points per game.

Okay…so here’s a different analysis. I looked at the All-American squads for the last decade. Some teams had All-American guards, some had All-American forwards and centers and a few had both. If having a great guard was a key to tourney success, you would expect that the teams with All-American guards would own the best PASE values and Final Four and championship advancement rates.

Not so. The 43 teams since 2004 with All-American guards have reached seven Final Fours (16.2%), won two championships (4.7%) and underachieved at a -.157 PASE rate—worse even than basic guard-dominant teams. The two champs were UConn and Kemba Walker in 2011 and Duke and Jon Scheyer in 2010.

Meanwhile, the 59 teams with All-Star frontcourt players have reached 12 Final Fours (20.4%), won five championships (8.5%) and overachieved at a +.167 PPASE rate—exactly the PASE that basic frontcourt-dominant teams have posted since 2004.

The Verdict: By any measure, you don’t need a great guard to go deep in the dance. If anything, you’re better off with a great frontcourt—and that’s good news to the likes of Arizona, Kansas and Kentucky. I’m not saying that you don’t need good guard play. Heck, I’d take a low-scoring, lock-down defensive guard in a heartbeat. But when the pundits talk about great guards, they mean someone who can put a team on its back and shoulder the scoring load. By that standard, the numbers don’t support the claim.

This entry was posted in Mythbuster, Tourney Trends. Bookmark the permalink.

8 Responses to Mythbuster #4: “You need a great guard to go deep in the dance.”

  1. Gary Diny says:

    Pete,

    Couple of questions/thoughts, not sure if you have the information but I will throw them out anyway:
    1. Are more players noted as Guards rather than Forwards? Seems that more team play a 3 guard lineup and that might skew the numbers of more points from guards now than in the past.
    2. How much of an effect does the increase in teams using the 3-point shot increase the points from guards, along with more 3 guard line-ups? I realize the 3-point shot came into use in the mid 80′s, so it has been around almost as long as the 64 team tournament, but the first 5-10 years it was not used nearly as much as it is now.
    3. How much does the early entry to the NBA affect the number of front court players in college BB? I might be dreaming this, but it seems to me that more front court players will leave early for the NBA than back court players. Generally there just seems to be less quality front court players (true post players) now than 10-20 years ago. The elite teams that generally make a deeper run, seem to have more of those quality bigs than the teams that do not.

    Thanks again!!

    Gary

    • ptiernan says:

      Gary,

      You may be right that more players are listed as guards or even guard/forwards (in which case I split the points). I also agree that the three-point shot has changed the dynamics of college hoops. On the third point, you may be right (call it the Kosta-Koufas Factor). Despite all, the fact remains that those fewer teams overperform–at a higher rate even than in the past.

    • Tom says:

      You kind of have a point re: #1. Rodney Hood is listed as a forward on Duke’s roster; two years ago, he was listed as a guard on Mississippi State’s roster. As I point out below, Kansas’s percentage of points from guards would look very different if Andrew Wiggins were listed as a forward (he’s essentially a classic three, as is Duke’s Hood.)

  2. Tom says:

    Current percentage of points from guards for top 10 teams (courtesy sports-reference.com):

    Florida: 41.1%
    Wichita St.: 59.7%
    Arizona: 44.9%*
    Syracuse: 42.5%
    Kansas: 59.5%
    Duke: 45.2%
    Louisville: 51.1%
    Villanova: 76.4%
    Creighton: 37.0%
    St. Louis: 56.9%

    Eight of ten get pretty balanced scoring; the exceptions are Villanova (whose only forward among its top five scorers is JayVaughn Pinkston) and Creighton (who, of course, has Doug McDermott on the team.)

    For Arizona, I’m not counting the currently-injured Brandon Ashley; if you counted him they would be at 29.3%. Kansas illustrates a different sort of problem: they have Andrew Wiggins listed as a guard on the roster when probably everybody except for Bill Self would consider a forward. Were he listed as a forward, KU would only be getting 32.1% of its points from guards.

    • BH says:

      Wiggins is a guard because he plays a guard position in self’s offense. He and wayne selden are both ‘wings’ in a system that has a lead guard (pg), dual wings, and dual forwards. In a more traditional offense you usually have a center, power and small forwards then a SG and PG. That is not how Self runs things. Each wing has the same role as does each forward. the PG or lead guard as Self call it is the only player with a unique role but once the play changes sides from strong to weak then on the way back it turns into a 3 guard rotation offense where they all have the same duties. to clarify if wiggins is on the strong side then he is a SG and on the weak side is what some might call the SF…but its really semantics at some point.

      • ptiernan says:

        No disagreement that the definition of Guard/Forwards–as well as the nuances of offenses–might influence my “guard scoring” numbers. But I’ve been measuring them the same way for the 29 years of the dance. I committed to staying consistent. It’s the only way to have a basis of comparison.

  3. Eric Timm says:

    Pete, you mentioned positions “listed” as the criteria you use to define a player…..listed by whom? The school’s official roster?

Leave a Reply

Your email address will not be published. Required fields are marked *

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>