Five Tips to Bracket Domination
It’s time to fill out your bracket!
For some, this is an easy task. They pick their alma mater, their spouse’s alma mater, their safety school and the school they always root for. Or maybe they pick teams based on their school colors or the cuteness (or ferociousness) of the team’s mascot. Better yet, they pick the team with the hottest cheerleaders. (“The Arizona Wildcats are gonna go all the way, I tell ya!”) They throw in their $20 and figure the job’s done.
For the rest of us, it can be an arduous task. We’ve watched at least a few games during the season, which leads us to believe that we can predict how the tourney is going to play out. We pour over the seeds, records, rankings, strength of schedule, common opponents, RPI – the list goes on and on. Somehow, we digest all these numbers to come up with a bracket that we feel comfortable with.
Or worse, we fill out different brackets for different pools. Sure, the chances that we’ll still be in the hunt as the tourney wears on go up if we fill out multiple brackets, but that doesn’t make it right. I grow weary when friends and co-workers talk about how they predicted three different upsets, only to find out that they’ve got six different brackets. Big whoop. A monkey could have done what they did. I feel strongly that each person only gets one bracket. Certainly, you can enter that bracket into multiple pools, but you better not change a thing. If you insist on changing your bracket for each pool, you must choose a King Bracket, which is the only one you’re allowed to talk about.
In years past, I combined the information gathered by watching games during the season with Jeff Sagarin’s computer ratings in USA Today to fill out my bracket. I have won my fair share of pools, but I’m by no means prolific. No one is. Even if you’re able to successfully identify the team with the best chance of winning every single matchup, that doesn’t guarantee that you’ll even have picked one Final Four team successfully. Still, I’m usually still alive when the Elite Eight rolls around, and that’s all a guy can ask for.
In my research for this column, I stumbled upon an excellent analysis by Patrick Reilly of BostonSportsHub.com. He examined all the tournaments since 1992 and tested several theories to see if there were any statistical trends to help give him an edge when filling out his bracket. This year, I’m going to use some of his findings as I fill out my bracket.
When examining a matchup, Reilly found that when two teams have a seed differential of 1 to 3, the team with the better seed won 51% of the time. So it’s basically a coin toss. When the seed differential is 4 or greater, the team with the better seed won at an 80% clip. So, generally, if there’s a 4+ seed differential, go with the better seed.
Do teams that play three or four games in their conference tournament feel the effects the following weekend? Reilly examined this question and found that #1 seeds are a solid bet (84%) to reach the Sweet Sixteen no matter how many games they played the previous week. Teams seeded #2 actually fare better (+40%) when they play 3-4 games in their conference tourney, while teams seeded #3-#6 were around 20% less likely to advance to the Sweet Sixteen if they played 3-4 games the previous weekend.
So what does this mean? If you’re going to pick any upsets in the first couple of rounds, look for #3 to #6-seeded teams that played 3-4 games in their conference tournament.
When comparing two teams with a seed differential of three or less, Reilly found five stat categories that had a relatively strong correlation with a particular team winning the game. Since I’ve been using it for years with success, I’m going to use Jeff Sagarin’s ratings as the sixth category:
1. Road/neutral court record (R/N)
When Team A has a road/neutral court winning percentage that is 20% higher than Team B, Team A wins 59% of the time.
2. Last 10 (L10)
When Team A has a “Last 10” (record over the team’s last 10 games) winning percentage that is 20% higher than Team B, Team A wins 56% of the time. (This number is 71% over the last three years.)
When Team A appeared in the previous year’s tournament and Team B did not, Team A wins 57% of the time.
When Team A has a field goal percentage that is 2% higher than Team B, Team A wins 62% of the time. I am going to go one step further and use effective FG%, which gives extra credit for three-point shots, while expanding the margin to 3%.
5. eFG% allowed
When Team A has an opponent’s field goal percentage that is 2% less than Team B, Team A wins 56% of the time. Again, I am going to use effective FG% and expand the margin to 3%.
(Thanks to KenPom.com for these stats.)
6. Sagarin ratings
If Team A has a Sagarin rating two points or higher than Team B, Team A is more likely to win the game. I’m not sure how much more likely, but the bookies use Sagarin ratings to help set their lines, so that’s good enough for me. Keep in mind that I use Sagarin’s “Predictor” rating, not the more politically correct “ELO Chess” rating, which doesn’t take margin of victory into account.
Using these six categories, we can come up with a series of tests to determine which team is most likely to win in a particular matchup.
All these numbers paint a nice picture, but it’s not complete without a little real world information. Be sure to take reality into account - injuries, travel, home court, physical matchups, etc.- when there isn’t a clear favorite.
If, after these four steps, you don’t have a favorite, pick the team you like more (or hate less). If you are a Duke hater, you don’t want to put yourself in a position to be rooting for them unless there’s good reason to believe they’ll win the game. This isn’t really something that will help you win your pool, but it should make watching the games much more enjoyable. After all, that’s what it’s all about, right?
Using these five steps, I’ll go game by game, from the first round to the title game, listing my pick along with the results of the Six Cat analysis in parenthesis. For example, in Game #1 I picked the Gators, who have an advantage in four of the six categories. Keep in mind that there will be a few picks that don't have a Six Cat advantage, like in Game #4 where Davidson has a 3-2 advantage over Maryland, but I still choose the Terps. Clearly, I’m not going to hit on all of these picks, so check back next Wednesday for an updated analysis of the remaining games.
Game #1: #1 Florida vs. #16 Jackson St. (Florida, 4-0)
Game #2: #8 Arizona vs. #9 Purdue (Arizona, 2-1)
The Wildcats’ R/N record and appearance last season gives them the edge.
Game #3: #5 Butler vs. #12 Old Dominion (Butler, 1-1)
Though the Six Cat test is tied, Butler holds the advantage in seed, Sagarin rating, and both shooting percentages.
Game #4: #4 Maryland vs. #13 Davidson (Maryland, 2-3)
Even though Davidson holds the advantage in R/N and L10 record, I have to go with the battle-tested Terps due to their higher seed and Sagarin rating.
Game #5: #6 Notre Dame vs. #11 Winthrop (Notre Dame, 1-3)
Even though Six Cat says that Winthrop is a live dog, the Irish still hold a five seed advantage and just gave Georgetown all they could handle in the Big East tournament.
Game #6: #3 Oregon vs. #14 Miami (OH) (Oregon, 2-1)
Game #7: #7 UNLV vs. #10 Georgia Tech (GT, 2-2)
This is a tight matchup, so I’m going to go with the Yellowjackets, who have a higher Sagarin rating and are better rested. When a matchup is this close, it’s better to go with the lower seeded team because most people will go the other way.
Game #8: #2 Wisconsin vs. Texas A&M CC (Wisconsin, 3-2)
This game might be closer than expected.
Game #9: #1 Kansas vs. #16 Play-In Winner (Kansas)
Game #10: #8 Kentucky vs. #9 Villanova (Villanova, 2-1)
This is a tight matchup, but Villanova holds the advantage in R/N and L10 percentages.
Game #11: #5 Virginia Tech vs. #12 Illinois (VT, 0-2)
This matchup is tight across the board, but VT is better rested and has the better Sagarin rating.
Game #12: #4 Southern Illinois vs. #13 Holy Cross (3-0)
Game #13: #6 Duke vs. #11 VCU (Duke, 3-2)
The Blue Devils are shaky, but played a brutal schedule and will welcome a new life in the Big Dance.
Game #14: #3 Pittsburgh vs. #14 Wright State (Pitt, 3-1)
The Panthers fit the profile of a team that may not get out of the first two rounds (#3-#6 seed, 3-4 conference tourney games), but they should get past Wright State.
Game #15: #7 Indiana vs. #10 Gonzaga (Gonzaga, 2-0)
I like the ‘Zags chances since they play in Sacramento and have a higher R/N record.
Game #16: #2 UCLA vs. #15 Weber St. (UCLA, 3-0)
Game #17: #1 North Carolina vs. #16 Eastern Kentucky (UNC, 3-0)
Game #18: #8 Marquette vs. #9 Michigan St. (MSU, 0-1)
This is a pretty tight matchup, but I’m going to go with MSU, who has the higher Sagarin rating and better shooting percentages.
Game #19: #5 USC vs. #12 Arkansas (Arkansas, 1-0)
The Razorbacks’ experience in last year’s tourney gives them the slight edge.
Game #20: #4 Texas vs. #13 New Mexico St. (Texas, 2-0)
Game #21: #6 Vanderbilt vs. #11 George Washington (GW, 4-0)
GW’s dominance in the Six Cat leads me to believe that they’ve got a good shot at upsetting the Commodores.
Game #22: #3 Washington St. vs. #14 Oral Roberts (WSU, 1-1)
Game #23: #7 Boston College, #10 Texas Tech (BC, 2-0)
BC’s experience and Sagarin rating should give them the edge.
Game #24: #2 Georgetown vs. #15 Belmont (Georgetown, 1-0)
Game #25: #1 Ohio St. vs. #16 Central Connecticut St. (OSU, 5-0)
Game #26: #8 BYU vs. #9 Xavier (Xavier, 2-0)
Xavier holds an edge in Sagarin rating and tourney experience.
Game #27: #5 Tennessee vs. #12 Long Beach St. (Tennessee, 2-2)
The Vols stink on the road, but they should have enough to get past LBS, since they play relatively close to home.
Game #28: #4 Virginia vs. #13 Albany (Virginia, 1-3)
Even though Albany holds the Six Cat advantage, I can’t ignore the Cavaliers’ edge in Sagarin rating and strength of schedule.
Game #29: #6 Louisville vs. #11 Stanford (Louisville, 2-0)
The game is in Lexington, KY and Louisville is playing better ball.
Game #30: #3 Texas A&M vs. #14 Pennsylvania (Texas A&M, 2-1)
Game #31: #7 Nevada vs. #10 Creighton (Nevada, 3-0)
Though Creighton has a slight advantage in Sagarin rating, the Wolf Pack has tourney experience and a better R/N record.
Game #32: #2 Memphis vs. #15 North Texas (Memphis, 5-0)
Game #33: #1 Florida vs. #8 Arizona (Florida, 3-0)
Arizona might give them a run, but the Gators should prevail.
Game #34: #4 Maryland vs. #5 Butler (Maryland, 1-0)
The Terps are playing better ball right now and should win this tight matchup.
Game #35: #6 Notre Dame vs. #3 Oregon (Notre Dame, 1-1)
ND has the better Sagarin rating and Oregon fits the profile of a #3-#6 seed that won’t get out of the first two rounds.
Game #36: #2 Wisconsin vs. #10 Georgia Tech (Wisconsin, 3-0)
This is a tight matchup, but the Badgers have a better R/N record and previous tourney experience.
Game #37: #1 Kansas vs. #9 Villanova (Kansas, 5-0)
The Jayhawks shouldn’t have a problem with ‘Nova.
Game #38: #4 Southern Illinois vs. #5 Virginia Tech (S. Illinois, 3-0)
This should be a good game, but the Salukis hold the advantage in three key categories, and I’m not even sure that VT will get out of the first round.
Game #39: #3 Pittsburgh vs. #6 Duke (Duke, 0-1)
Pitt’s fatigue from the Big East Tournament might be the difference in this extremely tight matchup.
Game #40: #2 UCLA vs. #10 Gonzaga (UCLA, 1-0)
This is a rematch of the great game last season, but the ‘Zags don’t have Adam Morrison this time.
Game #41: #1 North Carolina vs. #9 Michigan St. (UNC, 3-0)
The Spartans’ brutal R/N record contributes to the Tar Heels’ statistical advantage.
Game #42: #4 Texas vs. #12 Arkansas (Texas, 1-0)
This is a tight matchup on paper, but Kevin Durant and D.J. Augustin should put the Longhorns into the Sweet Sixteen.
Game #43: #3 Washington State vs. #11 George Washington (WSU, 1-1)
I don’t have much confidence in the Cougars, but they should have enough to get by GW.
Game #44: #2 Georgetown vs. #7 Boston College (Georgetown, 5-0)
The Hoyas shouldn’t have a problem with BC.
Game #45: #1 Ohio St. vs #9 Xavier (OSU, 2-0)
This should be a fun game if both teams’ fans make the short trip. This will be a game for a while, but the Buckeyes should prevail.
Game #46: #4 Virginia vs. #5 Tennessee (Tennessee, 1-1)
This matchup is a tossup, but the Vols’ tourney experience should help.
Game #47: #3 Texas A&M vs. #6 Louisville (Texas A&M, 3-0)
Despite the game being in Kentucky, I think A&M has the grit to get past the Cardinals.
Game #48: #2 Memphis vs. #7 Nevada (Memphis, 2-0)
The Tigers haven’t fared very well against good competition, and though I like Nevada’s chances, the Tigers are still a good bet for the Sweet Sixteen.
Game #49: #1 Florida vs. #4 Maryland (Florida, 3-0)
The Terps certainly have the juice to get past the Gators, but I think Florida is the much safer bet.
Game #50: #2 Wisconsin vs. #6 Notre Dame (Wisconsin, 2-0)
I don’t feel good about putting either team through to the Elite Eight, so I’m going with the Badgers, who have the advantage in Sagarin rating, R/N record and tourney experience.
Game #51: #1 Kansas vs. #4 Southern Illinois (Kansas, 2-0)
The Jayhawks hold the advantage in every Six Cat stat save for tourney experience.
Game #52: #2 UCLA vs. #6 Duke (UCLA, 1-0)
Considering Duke’s spotty play this season, this is a much tighter matchup than you’d think. It’s a safe bet that the Blue Devils won’t play three good games in a row, so I can’t put them through to the Elite Eight. UCLA’s consistent defense makes them a safe bet.
Game #53: #1 North Carolina vs. #4 Texas (UNC, 2-0)
The Tar Heels hold a 10-point advantage in Sagarin rating and have a much better R/N record. I like the Longhorns, but they fit the profile of a #3-#6 seed that may not make it out of the first two rounds.
Game #54: #2 Georgetown vs. #3 Washington St. (Georgetown, 4-0)
The Hoyas hold the advantage in every Six Cat stat and are a good bet to make the Elite Eight.
Game #55: #1 Ohio St. vs. #5 Tennessee (Ohio St., 4-0)
I have little confidence in the Vols making it this far, and if they do, they’ll be running into a brick wall.
Game #56: #2 Memphis vs. #3 Texas A&M (Texas A&M, 1-2)
Memphis had a cakewalk for a schedule, so their R/N record is misleading. A&M has been in tight games all year and holds the advantage in Sagarin rating and both shooting percentages.
Game #57: #1 Florida vs. #2 Wisconsin (Florida, 1-0)
I don’t feel good about the Badgers making it this far, and with Brian Butch out, they won’t have enough to get past the Gators.
Game #58: #1 Kansas vs. #2 UCLA (Kansas, 4-0)
The Bruins won’t be able to score with the Jayhawks, who are on a serious roll.
Game #59: #1 North Carolina vs. #2 Georgetown (Georgetown, 1-1)
The Tar Heels are still the favorite, but the matchup is tight enough to give the Hoyas the nod. They have a better R/N record and hold the advantage in both shooting percentages.
Game #60: #1 Ohio State vs. #3 Texas A&M (Texas A&M, 1-2)
The Aggies hold the advantage in guard play, Sagarin Rating and both shooting percentages. This is a close matchup, so I feel compelled to go with the dog. Of all the Elite eight games, I feel least comfortable with this pick.
Game #61: #1 Florida vs. #1 Kansas (Florida, 1-1)
This is another tight matchup, but I have to go with the Gators, who have proven they have the cojones to win a NCAA title. Florida has hit some bumps along the way, but when they’re focused, they are a very tough out.
Game #62: #2 Georgetown vs. #3 Texas A&M (Georgetown, 2-0)
Will the GU guards be able to play with Acie Law? The Hoyas are tough defensively and hold the advantage in R/N and L10 records.
Game #63: #1 Florida vs. #2 Georgetown (Florida, 0-1)
The Hoyas hold the advantage in R/N and L10 records, but the Gators have a better Sagarin rating and eFG%. When the game is tied with five minutes to play, which team will be relaxed and which team will be shitting their pants? Florida wins back-to-back titles for the first time since Duke did it in ’91 and ’92.
Strap ‘em up! We’re in for a wild ride!