When Niagra beat Florida A&M Tuesday night, few people in Raleigh seemed to notice, after all, N.C. State was playing its first game in the NIT around that same time.
And besides, a win by the Purple Eagles doesn’t affect many brackets — unless you’re State senior R.J. Menachof.
The history major’s project for Independent Research in Mathematics was off to a good start.
By plugging 22 sets of variables into an extended version of the distance formula, Menachof created a mathematical formula that will try to accurately predict each NCAA Tournament game. Although he and his professor, John Griggs, had agreed that the project would be graded pass/fail, Niagara’s win was still good news for Menachof because it meant the first game was correctly predicted.
“It was a big sigh of relief because I didn’t want to start out 0-for-1 … Niagra was, according to my model, way better than Florida A&M,” Menachof said. “I have faith in my model. If my grade depended on it, I would have faith in it.”
Menachof spent a great deal of time thinking of all sorts of ways to try to maintain accuracy in a game that’s not played on paper.
So besides standard factors like margin of victory and turnovers per game, he added “average points per game without the two leading scorers” and “average rebounds per game without the leading rebounder” to reflect a team’s depth.
And in order to keep things realistic enough so that teams like No. 16-seed Central Connecticut State, who went 16-2 in its conference, didn’t miraculously advance to the Final Four, Menachof multiplied the conference’s RPI with the record in conference to depict a team’s quality of competition.
“Some people would think that maybe it’d be skewed because of the competition you play against,” Menachof said. “But the stats are multiplied by conference strength, so you get credit for playing better competition.”
Menachof tested it on previous conference tournaments in the PAC-10, Big Ten and ACC and was 70 percent right, missing an average of three games per tournament.
In the formula, low number totals are good, because they mean the team placed in the higher-ranking numbers. At the end of its calculation, the team with the lowest total number meant it was higher in more statistical rankings than the rest.
“With my project, the farthest away you can be is to be 65th in everything, which comes up being the number 568.3. So if you were that, you’d be the worst team out there,” Menachof said. “But no one’s that because some teams are just bad in some points in their game but strong in others.”
Kansas’ total of about 135 put it in first. The closest team to that number, Texas A&M, had 10 more points.
“Kansas was the most consistent. They were in the top-20 in just about every category, so which is why I have them as the national champions.” “A lot of other teams are pretty close, but Kansas is a good head-and-shoulders above of everyone else.”
The formula had some interesting findings. For instance, although Xavier was calculated as the 11th best team in the tournament, it is predicted out in the first round because its opponent, BYU, is ranked seventh.
“A lot of it has to do with who you play,” Menachof said.
Likewise, BYU isn’t predicted to go much further because its next opponent, in theory, is No. 6 Ohio State.
The closest first-round game, according to Menachof’s equation, will be Southern Illinois versus Holy Cross.
“That one’s only separated by seven-tenths of a point. If I had to give a game to watch I’d say that one,” Menachof said.
Even though Menachof said he spent more than 35 hours working on the formula over the past few nights compiling the numbers from each team, he doesn’t necessarily believe it wholeheartedly.
He said however, he was entering two brackets in each pool he’s involved in because he didn’t want to miss out in case his figuring was accurate.
“My picks differ somewhat,” Menachof said. “Definitely the Notre Dame [going to the Elite Eight] would be my biggest disagreement.”
“I was very worried it was going to spit out something like Virginia Commonwealth winning the championship, but after looking at this I’m very confident that it will continue to predict well.”
According to R.J. Menachof’s formula:
National Championship — Kansas over Texas A&M
Final Four — Kansas, Texas A&M, Florida, North Carolina
Lowest seed in Sweet 16 — UNLV & Nevada
Lowest seed in Elite Eight — Notre Dame
First 1-seed out — Ohio State (Elite Eight)
First 2-seed out — Wisconsin & Memphis (second round)
First 3-seed out — Oregon (second round)
First 4-seed out –Virginia & Southern Illinois (second round)
First round upsets — (12) Old Dominion, (12) Arkansas, (11) VCU, (11) George Washington
Top 10 teams in the tournament —
1. Kansas
2. Texas A&M
3. North Carolina
4. Florida
5. Notre Dame
6. Ohio State
7. BYU
8. Maryland
9. Texas
10. Nevada
22 factors in the formula
Scoring margin
Field goal percentage
Opponent’s FG percentage
Free throw percentage
Rebounding margin
Turnovers per game
Turnover margin
Assists per game
Turnover-Assist ratio
3-point percentage
Blocks per game
Win-Loss percentage
RPI
Strength of Schedule
Conference Strength X Conf. W-L percentage
Points per game without top 2 scorers
Rebounds per game without the top rebounder
W-L last 5 games
W-L last 10 games
W-L in games decided by five points or less
W-L versus other NCAA Tournament teams
Conference RPI