5 tips for winning your bracket pool

1. DON’T GO CRAZY PICKING CINDERELLAS

According to Tiernan, the 2014 tournament was the most unpredictable since the field expanded to 64 teams in 1985 (now 68 including the play-in games). The 2013 tournament ranked second in unpredictability.

Writes Tiernan: “In fact, the last five years have easily been the craziest of the 30-year 64-team era. In the first 25 years of the modern bracket, the average dance deviated from higher-seed perfection by 13.6%. In the last five years, that deviation is 19.2%, more than 40% wilder. Coincidence? I don’t think so.”

Chances are there will be upsets. But history says to go ahead and advance all No. 1 seeds to the fourth round and all No. 2 seeds to the third round. It’s not very exciting to do this, and chances are at least one of those eight teams will lose earlier, but it’s the safe play:  The average No. 1 seed wins about 3.32 games per tourney and the average No. 2 seed wins about 2.41 games per tourney.

Also don't pick many seeds lower than 12 to win a first-round game and don't pick any of them to advance past the second round. In the 64-team era No. 16 seeds have never beaten a top seed, Nos. 13-15 are a combined 59-360 (.164) in the first round and only seven of 120 teams (5.8 percent) seeded 13 or lower have advanced as far as the Sweet 16.

2. PICK A 12 SEED (OR TWO) TO BEAT A FIVE

It's happened enough times now that it's actually wise to pick a 12 seed to win. Only three times since the tournament expanded to 64 teams in 1985 has a five seed failed to lose to No. 12. Three No. 12 seeds beat No. 5 seeds in both the 2013 and 2014 tournaments.

The 5-12 matchups this year are West Virginia-Buffalo, Northern Iowa-Wyoming, Arkansas-Wofford and Utah-Stephen F. Austin.

Pomeroy's rankings show that Northern Iowa is a particularly strong No. 5 seed and Wyoming is the weakest No. 12 so don't pick an upset there (Pomeroy gives Wyoming a 15 percent chance of winning). According to Pomeroy Wofford has a 27 percent chance of upsetting Arkansas. Utah also is a strong No. 5 seed but Stephen F. Austin is the best No. 12 by far in Pomeroy's rankings and has a 26 percent chance of winning. Buffalo is the best bet of all the No. 12 seeds with a 37 percent chance of winning, according to Pomeroy.

So go ahead and pick Buffalo with confidence. If you want to back another No. 12 seed, I'd go with Stephen F. Austin. The Lumberjacks are capable of making Utah, a slow-paced team, play faster because they shoot the ball well and create a lot of extra possessions with steals and forced turnovers (while rarely turning the ball over themselves).

3. DON’T PICK A SEED LOWER THAN EIGHT TO MAKE THE FINAL FOUR

This parallels tip No. 1 above. It just doesn’t happen often that a team seeded lower than No. 8 makes the Final Four, even with the recent trend of upsets.

In the past five years only two of 20 Final Four teams were seeded lower then eighth (No. 11 VCU in2011 and No. 9 Wichita State in 2009). In the 30 years of the expanded tournament only four of 120 (3.3 percent) teams in the Final Four were seeded lower than No. 8.

Only one team seeded lower than sixth has advanced to the championship game: eight-seeded Villanova, the famed Cinderella that won it all under coach Rollie Massimino in 1985.

4. CHECK CHAMPIONSHIP CREDENTIALS

Picking the championship winner usually is worth the most points in bracket pools so get it right. Since the expansion of the tournament, teams seeded No. 1 or 2 have won the championship in 21 of the 30 years. No. 3 seeds won the tournament four times, No. 6 seeds won twice, and seed Nos. 4, 7, and 8 each won it all once.

Picking a one or two seed to win the tournament a good bet but use the numbers to pick from among them.

Since 2002, the first year of Pomeroy's adjusted efficiency rankings, all but one champion ranked in the top 18 in offense and the top 17 in defense. The exception is last year's champion, Connecticut, which ranked No. 39 in offensive efficiency and No. 10 in defense.

Assuming there won’t be another surprise champion, the winner of this year’s tourney likely will rank at least No. 18 in offensive efficiency and No. 17 in defensive efficiency. Teams in the field that qualify: Kentucky, Arizona, Villanova, Utah and Northern Iowa.

Furthermore, Tiernay says 13 of the last 14 champions have shared eight credentials (Connecticut is the lone exception):

The teams that meet all eight of those criteria in this tournament are Kentucky, Villanova, Arizona and Duke.

That leaves just three teams that have championship credentials according to both the Pomeroy and  Tiernay formulas: Kentucky, Arizona and Villanova. Picking one of them to win it all is a sound play.

5. BEWARE BRACKET-BUSTING COACHES

You already know most of the good tournament coaches even if you don’t follow much college basketball. They are the brand names. But certain coaches have earned a reputation for flaming out early in tournaments with higher seeds.

Tiernay developed a formula ,Performance Against Seed Expectation (PASE), that quantifies how much coaches over- and under-perform their seeds. Luckily most of the worst under-performing coaches are not in this tournament but three coaches are waiting to ruin your bracket.

Pick these coaches to lose when their seed says they shouldn’t or, at the very least, don’t advance their teams far in the NCAA bracket: John Thompson III (Georgetown), Steve Alford (UCLA) and Mike Brey (Notre Dame).

Alford is easy to pick against because, not only is he a bad tournament coach, but his 11th-seeded Bruins are up against SMU. The Mustangs are led by Larry Brown, making this game a monumental coaching mismatch.

Also avoid backing Brey or Thompson. The seventh-seeded Irish's 76-58 thumping by 10th-seeded Iowa State in the 2013 tournament was the fourth consecutive season Brey's Notre Dame team lost as the higher seed.

Thompson III made the 2007 Final Four as a No. 2 seed; since then he has five losses as the higher seed in five tournaments, including the memorable loss to No. 15 seed Florida Gulf Coast in the 2013 tournament (he's also lost a first-round game with No. 3 seed).

On the flip side are the coaches that most years you can count on to at least win the games their seeds say they should or overachieve: Sean Miller (Arizona), Tom Izzo (Michigan State), Rick Pitino (Go Cards). Mike Krzyzewski (Duke), John Calipari (Kentucky), Roy Williams (North Carolina) and Thad Matta (Ohio State).

Beware, though, of Coack K. His tourney results have actually been well below seed expectations over the last 10 years. He could be going the way of Bob Knight and Denny Crum and seeing his tourney results fade as he ages (though it's all relative for one of the best tournament coaches in history).