Key trends that are changing the NCAAs

One year frontcourt dominance foretells tourney success, and the next year team youth might be a more reliable indicator. On the surface, it may seem that things aren't changing. After all, high-scoring teams with experienced coaches are still outperforming their opponents. Below the surface, however, the tournament is gradually mutating. March Madness of 1985 was a very different event than the tourney that will tip off in 2006.

Here are 11 trends that make filling out your bracket more difficult:

1. Reversed trends in 2005: Offense, victory margin and team youth

Last year reversed the course of three key trends that characterized the tourney field. Over the last 16 years, the scoring offense of tourney qualifiers has steadily declined, from 82.7 points per game in 1989 to a low of 73.6 in 2004. In 2005, the average scoring of the tourney field actually rose for only the third time during this period. It was only a slight uptick to 74.1 points a game, but it may signal a renaissance for offensive-minded teams. As we'll see in trend No. 6, offense remains a key indicator of which teams will advance to the Final Four and beyond. Now it might also signal which teams get a ticket to the dance.

Another trend that the 2005 tourney reversed was the diminishing value of victory margin. Considered in relative terms (that is, the percentage difference between how many points a team scores and how many it allows), the margin of victory for tourney qualifiers had been steadily declining since 2000 -- a sign that more battle-tested teams were getting to the dance. Last year, that trend swung back in favor of teams that won by comfier margins. The 13.3-percent victory margin difference was the second highest in the 21-year history of the 64-team era. As we'll see in trend No. 7, relative margin is also a strong indicator of success in the tourney.

Finally, the age of tourney qualifiers and advancers made an abrupt reversal in 2005 as well. By assigning a point value to the four college classes (1 for freshmen, 2 for sophomore, 3 for junior and 4 for senior), we can compare the relative ages of the starting lineups for each year's tourney field. The 2005 field broke a four-year pattern of older teams' making the tourney. On the other hand, as we'll see in trend No. 10, last year emphatically reinforced a three-year trend of older teams' advancing to the Final Four and winning the championship.

2. Despite bracket fortunes, last year's tourney was relatively predictable
After high-profile teams such as No. 3 Kansas and No. 4 Syracuse went down in the first round and No. 2 Wake Forest and top-seeded Duke made early exits, analysts were proclaiming 2005 as the year of the upset. In fact, last year saw fewer than the average tourney's upsets. There were only eight games in 2005 in which a team four or more seed positions below another emerged as the victor. The average tourney has closer to nine -- with the most (13) occurring in 1985, 1986, 1990 and 2002. Last year was also low in the number of near-upsets -- games in which a team three seeds lower than its opponent sprung a surprise. There were only five such games, the most significant of which were No. 5 Michigan State over No. 2 Kentucky and No. 4 Louisville over No. 1 Washington.

If you divide the 21 years of the modern tournament into three seven-year periods, the most recent era has about three more upsets per tourney (9.1) than the calm years between 1992 and 1998 (6.3), but it isn't as crazy as the first seven years of the modern era (10.3 upsets per tourney). One trend in particular mirrors the fluctuations in upsets: Conference affiliation. The average number of mid-majors qualifying for the tourney peaked in the first seven years of the tourney, bottomed out in the second seven-year period and has risen slightly in recent years. Conversely, the number of small conference schools making the tourney has followed the opposite course. Not surprisingly, there have been fewer upsets in tournaments when the number of small conference schools was higher and more when the number was lower.

3. The incredible shrinking mid-major pool has bottomed out
Speaking of conference affiliation, where have all the dangerous mid-major schools gone? (This year, the answer's easy: Big East.) The 2005 tourney involved only 10 mid-major teams, easily the lowest number of the 64-team era ... and down from a high of 23 teams in 1993. Meanwhile, small conferences fielded a record 23 teams, six more than their previous high. We've already touched on how the number of small conference teams has an inverse relationship to the number of upsets. But as of late, even the mid-major schools that get into the tourney aren't making much noise.

The fact is, neither the mid-majors nor the small conferences are performing well enough in the tourney to warrant taking slots from the big six conferences. Of the three seven-year periods of the modern tourney, mid-majors have performed the worst in the most recent period. Not only is their winning percentage the lowest (.438 during 1999-2005 compared to a high of .456 during 1985-1991), but they're performing the worst against seed expectations, winning only 85 games when their average ninth seed says they should win 90. The same goes for the small conferences. In the first seven years of the modern tourney, they won about one in six games (.163 winning percentage). Since 1999, they've won less than one in 10 (.096). And they've also posted the worst performance against seed expectations, winning just 13 games when their average 14 seed dictates that they should have won 19.

There's no reason why the selection committee shouldn't include more teams from the big six conferences. The big six are winning more often than ever (343-210, .620 percentage) and have exceeded seed expectations by more wins (11), despite being saddled with their highest average seed (5.1) of the three seven-year periods in the modern era. It will be interesting to see whether the absorption of old mid-major powerhouses such as Louisville, Cincinnati and Marquette will resolve or aggravate this issue. If the selection committee holds the line of big six tourney qualifiers at the historical average of about 30, the problem likely will worsen. If they accommodate more big six teams, look for a more competitive and unpredictable tourney.

4. Experienced coaching has never been more important

If you need any proof of how critical coaching is to success in the tourney, consider how the coaching experience grows for schools that advance in the tourney. The average tourney qualifier is led by a coach who's been to the dance an average of 5.6 times. The average Final Four team is led by a coach with 9.5 years of tourney coaching experience. And the average champion has a coach with 11.6 tourneys under his belt. It isn't so surprising that tourney advancers are led by more experienced coaches, but what is surprising is that coaching experience has become increasingly important in recent years.

The average qualifier in the first seven years of the tourney was led by a coach with five years of experience. However, since 1999 the average tourney coach has been to the dance six times. The gulf between these numbers widens the deeper a team advances in the tourney. During 1985-1991, the average Final Four team had a coach with eight years of experience. In the last seven years, the average coach has been to the dance 10.5 times. Champions in the early years of the modern era were guided by coaches with 8.7 years of tourney experience; that number has climbed to 14.4 over the last seven years. The bottom line: Think twice before you pick an upstart coach to knock off a coaching legend (despite the occasional stumbles from the likes of Jim Boeheim, Lute Olson and Mike Krzyzewski), because they're the coaches that are more likely to shepherd their teams to the Final Four and beyond.

5. More experienced teams are making the dance; less experienced teams are advancing
Just as tourney coaches are getting more experienced, so too are their teams. The difference, however, is that team experience is becoming less critical to advance in the tourney. Since 1999, the average squad has come into the tourney with 3.5 straight appearances. That's up from 2.8 consecutive trips in the first 14 years of the modern era. Actually, the number of teams per tourney that hadn't been to the previous year's dance has held steady at about 30; the 34 repeat teams are on longer appearance streaks (5.7 years since 1999 versus 4.5 years from 1985 to 1998). While repeat teams are getting longer in the tourney tooth, those advancing to the Final Four and winning the championship are on decidedly shorter streaks.

In the middle years of the modern era, from 1992 to 1998, Final Four teams had a 7.6 year tourney appearance streak. Since 1999, the number is 6.6. The drop-off is even more dramatic for champions. From 1992 to 1998, NCAA champs had been to the tourney an average of 9.6 straight years. Since 1999, they've only been to the tourney 3.7 years in a row -- about the same number as the overall field (3.5).

The trend toward upstart squads winning the whole ball of wax started with Syracuse, which was the first team in the modern era to take the championship without having gone to the previous year's dance. Then Connecticut and North Carolina won with only three and two straight trips, respectively. There are a couple of implications to the growing inexperience of tourney advancers:
1. College programs seem to be able to retool much faster now after failing to make the tourney.
2. The old rule that a tourney-tested team will conquer the bracket no longer applies.

6. Scoring offense remains low, but offensive-minded teams are more likely to advance
Despite the uptick in the scoring averages of the tourney field (see trend No. 1), offensive production has steadily declined from the high-flying early years of the tourney, when teams such as Oklahoma and Loyola-Marymount were averaging over 100 points a game. During those first seven years of the modern era, the tourney field averaged 78.6 points a game. Since 1999, the average team is scoring four fewer points per contest -- bottoming out in 2004 to 73.6 a game. But just because scoring is down doesn't mean that defense dictates tourney performance.

In fact, offense is still one of the key factors in determining which teams advance deep into the dance. Over the last seven-year period, the Final Four contenders have averaged 4.6 points more than the tourney field (80.2 to 74.6). Meanwhile, the champions averaged more than eight points better than the field (82.7 to 74.6) -- and even higher than champions from the first seven years of the tourney (82.5) when scoring was at a premium. Scoring offense is also a key characteristic of Cinderellas. No. 11 through No. 14-seeded teams that have sprung a first-round upset are 2.5 points more prolific than those that succumbed to their higher-seeded opponents. Bottom line: Yes, scoring is going down, but the teams that succeed in the tourney continue to have relatively high-powered offenses.

7. Teams are winning by wider relative margins, but it's not impacting predictability
A curious ancillary to the decline in scoring offense is the increase in relative margin of victory. You'd think lower scoring teams would translate into tighter contests. Not so. While the absolute margin of victory has held steady throughout the modern era at about 8.5 points a game, the relative margin (the number of points by which a team wins divided by its opponent's average points scored) has grown from about 11 percent in 1985 to 13 percent since 1999. Final Four contenders and champions are also winning by bigger relative margins.

Whereas semifinalists were about 20 percent better than opponents in the first 14 years of the tourney, they're 22 percent better over the last seven years. Champions won by 19 percent from 1985 to 1998, and since then, they've won by 24 percent. Interestingly, there's no connection between increasing relative margins and upsets. During the six years that saw 10 or more upsets in the tourney, the relative margin of victory was 12.4 percent -- exactly the same as it was for all the other years. The point is that a high victory margin is a double-edged sword: It could be a sign of a team that can advance deep into the tourney ... or a sign of a team with a soft schedule that's ripe for a fall.

8. Tourney qualifiers are deeper than ever ... while champs continue to rely on starters
The NCAA Tournament has never been about who has the better bench. Over the course of the 64-team era, the average tourney qualifier gets a little over 20 percent of its points from nonstarters. In the first seven years of the modern dance, the field averaged just 19.8 percent of its points from the bench. That number has steadily grown over the last 21 years. Since 1999, the average tourney team relies on its bench for 21.2 percent of its scoring.

Final Four contenders are also deeper, averaging 21.4 percent bench scoring against 19.1 percent. But champions remain just as reliant on their starters as they did in the first seven years of the tourney, getting a scant 17.6 percent of scoring off the bench for both periods. Curiously, during the middle period of the modern tourney era, from 1992 to 1998, champions were much deeper, relying on their benches for 26.6 percent of their points. This was also the period that saw the fewest upsets in the tourney. Go figure.

Still, the main message is clear: Don't be swayed that the depth of a tourney contender will translate into a deeper run.

9. Backcourts have grown more important for making the tourney, but strong frontcourts determine champions
Last year's tourney was a perfect case study in the paradoxical value of guard play. Teams need a strong backcourt to get into the tourney and advance to the Final Four, but they can't rely on guard play to win the championship. Illinois found that out against North Carolina ... and the numbers underscore the point. With each succeeding seven-year period in the modern era, the tourney field has relied increasingly on backcourt scoring. From 1985 to 1991, tourney qualifiers got 44.6 percent of their points from guards. That rose to 46.6 percent during 1992 to 1998 and has been an even 50 percent for the period since 1999. Over the first 14 years of the modern era, Final Four contenders were more reliant on forwards and centers for scoring, getting just 45.8 percent of their points from guards.

Since 1999, however, semifinalists are actually more guard-oriented than the rest of the field, with their frontcourts shouldering 52.6 percent of the scoring load. Just when you're ready to buy into all the pundits' jabber about how critical good guards are to tourney success, the champions disprove the myth. Tourney champions are far more frontcourt-dominant than the rest of the field, consistently averaging about 5 percent more points from forwards and centers in each of the three seven-year periods of the 64-team era. From 1999 to 2005, the champs relied on guards for just 44.9 percent of their scoring. The last two tourney winners, Connecticut (33 percent scoring from guards) and North Carolina (29 percent) are prototypical champions of recent years -- big, dominant frontcourts with competitive enough guard play to neutralize their more backcourt-dependent opponents.

10. Tourney teams are getting younger, especially champions
While it's true that 2005 saw older teams advancing deeper into the dance, the seven-year trends still suggest that progressively younger squads are achieving tourney success. Given how many players are cutting short their college careers to join the NBA, this shouldn't come as a surprise. And it goes a long way toward explaining why schools can miss the tourney one year and quickly reload for success (trend No. 5). By assigning a point value to the four college classes (1 for freshmen, 2 for sophomore, 3 for junior and 4 for senior), we can compare the relative ages of the starting lineups for each year's tourney field. Over the first 14 years of the tourney, the average team fielded a starting lineup a shade more than all juniors (3.02). Since 1999, the field is made up of teams with younger starters (2.93). And the starting lineups get younger the deeper teams advance in the tourney, with Final Four contenders (2.87) and champions (2.83) relying more on sophomores and freshmen. The last true veteran squad to win the tourney was Maryland in 2002 (3.4). Since then, Syracuse, UConn and North Carolina have all relied on youth for their success -- and promptly lost their stars to the NBA. So as you fill out your bracket, don't be swayed by a veteran team. The fact that they have a lot of juniors and seniors in their starting lineup might just mean that those players weren't strong enough to make the leap to the pros.

11. RPI might determine seeding, but it doesn't predict success
The controversial RPI, which has been used since 1994 to help determine seeding, provides little guidance in predicting tourney performance. If the RPI was a perfect system, all the top-seeded teams would have an RPI value of four or less. Any top seed with an RPI greater than four would presumably be "over-seeded" and more likely to underperform against seed expectations. In fact, top seeds with an RPI greater than a four actually have a better winning percentage (49-11, .816) than those with an RPI of four or less (109-29, .790). For an example, look no further than last year's Tar Heels squad, which had an RPI of seven. In fact, of the top six seeds, the teams with an RPI higher than their seeding would dictate (more than eight for two seeds, more than 12 for three seeds, etc.) outperform their lower-RPI counterparts at the one, three and six seeds. The two, four and five seeds show a slight advantage for the higher-RPI squads, but not significant enough to warrant using RPI as a performance indicator.

Freelance writer Pete Tiernan has been studying the NCAA Tournament for 16 years. E-mail him here.