Who doesn't love hearing this?
Ahh, March Madness is finally here, and it's almost time for everyone to start filling out their brackets! Everyone has their own way of doing it, and most of the time it doesn't work too well, but usually picking the Final Four correctly, and definitely picking the champion correctly will get you close to the money in a lot of bracket pools.
So, what gives? The NCAA tournament has been going on for 25 years in the 64-team (or 64+1 team) format. Some stat geek had to have crunched the numbers and given us something to work with!
It turns out, that is the case. One such man is Pete Tiernan who cranks out tons of articles every year around this time jam packed with information. He does offer one huge caveat though:
Of course, every March Madness forecasting strategy should come with a gigantic disclaimer: Past performance does not guarantee future results. Each tournament manages to violate rules that were once considered ironclad.
But enough of that, let's jump into the stats!
Pete Tiernan's "Crown Your Tourney Champion"
The last 21 champions in a row, and 23/25 champions have possessed the following attributes:
- Seeded 1, 2, 3, or 4 (obviously we won't know this for a few days)
- Score at least 76 ppg
- Win on average by at least 10 points
The teams who currently possess such attributes? Well, ignoring the seeding requirements, those teams are (through games of 3/11): New Mexico, Sam Houston State, Gonzaga, Kansas State, Texas, Xavier, Missouri, Marshall, Maryland, Baylor, St. Mary's (CA), Kentucky, Syracuse, Murray State, Duke, Kansas, and BYU. I've also crossed out the teams that basically have no chance of scoring a top seed.
This leaves us with 13 teams. Not exactly narrowing it down, is it? Well, keep that group in mind, because on average 2.83 of those teams per tournament will be making it to the Final Four. Obviously, take out any one of them that don't get the top seeds, but that's a pretty solid group to start with.
Additionally, the vast majority of the champions had gone to the tournament the previous year. This qualification eliminates New Mexico, Kansas State, Baylor, and Kentucky. That whittles us down to 10 teams.
You can whittle it down even more by upping the scoring and scoring margin requirements and adding a few more. 17 of the past 19, and 18 of 25 champions overall have had the following requirements:
- Score at least 77 ppg
- Outscore opponents on average by at least 11 points
- Come from the ACC, Big East, Big 12, Big 10, SEC, or Pac 10.
- Have a coach who has been to the tourney at least 5 times.
This will eliminate Gonzaga, Xavier, BYU, and Missouri and leave you with 6 legit contenders this season that based on overall tournament history have about a 72% chance of winning the tournament (and based on recent trends about a 89% chance). Those teams are: Texas, Maryland, Syracuse, Kentucky, Duke and Kansas. However, I will make a caveat that based on Texas' recent collapse (and thus likely not getting a top 4 seed) you could probably eliminate them. Maryland will be close to a 4 seed and it will depend on their ACC tournament performance probably.
Pete Tiernan adds a few more juicy stats in another article, saying that all but one tourney champ has never lost 2 or more games in a row, and all but two have gone into the tournament winning at least 7 of their last 10 games.
This eliminates Syracuse and Maryland, and leaves us with a trio of solid picks - Kentucky, Duke and Kansas. Of course this doesn't mean that the champion will definitively be one of those three teams, but there is a high statistical likelihood that it will be, at least if the past is any indication.
Other Encouraging Signs
Another quote from Pete Tiernan before we jump into something else:
The 111 teams with a pre-tourney scoring margin above 15 points per game won nearly 48 games more than seed projections. That works out to a weighty +.429 PASE-about half a game per tourney better than expectations. Perhaps even more impressively, 41 of these 111 squads (37 percent) have reached the Final Four and 12 (11 percent) have won the tourney. Both of these percentages are the highest for any single attribute we examined.
That only begins to explain the importance of victory margin in foretelling tourney overachievement. Consider this: of the top eight attributes I studied, four of them related to scoring margin. Teams that beat opponents by more than 14, 13 and 12 points all had impressive PASE values that would have landed them in the chart above. However, I elected not to repeat similar attributes.
The only two teams of the "contending five" that I isolated that have a scoring margin greater than 15 ppg were Kansas and Duke, for what it's worth. Also, considering that performance above seed expectation (PASE) is somewhat important in forecasting other games correctly, consider that the only other two teams with a >15 ppg scoring margin are BYU and Murray State. Take note of them as possible spoilers.
Jumping to Adjusted Scoring Margin
Here is an excerpt for something I wrote for the blog I write for (Crimson and Cream Machine) last year:
Another interesting metric for tournament performance, and for performance overall is adjusted scoring margin (ASM) that ESPN's Joe Lunardi uses in his bracketology segments. Adjusted scoring margin is a part of their insider content, so I won't give away the rankings, but I'm sure some of you are wondering what ASM is. It is the sum of two components:
- The number of ppg a team scores above/below the average ppg allowed by all of it's opponents combined. Basically, if a team doesn't score many points per game, but they play in a conference with great defense, it will take that into account.
- Basically the opposite of the above concept to take into account a team's defensive prowess.
Therefore this tends to be a little bit better than simply looking at scoring margin. I'm not sure how it's been in recent years, but it has a history of success. The Florida Gators of 2000 had the 2nd highest ASM in the nation, but were a 5-seed. They got to the Final Four.
The top 5 in ASM right now, in order, are: Kansas, Duke, Syracuse, BYU, and Kentucky. Do the teams sound familiar? The fact that both of these proven tournament metrics are turning up the same teams consistently is definitely something to take note of!
Looking for cinderellas? The same philosophy can apply. BYU keeps turning up and they are projected to be a 5 seed in Chris' latest bracket. Murray State is an interesting team as they are in the top 4 nationally in scoring margin and they are projected to be a 13 seed, so there's a potential lead on a big upset. Maryland also has a lot of good attributes and they are projected to be a 5 seed like BYU. They are #7 nationally in ASM, so nothing to sneeze at.
Does ASM identify any other potential sleepers? Here they are with ASM rank followed by seed projection by Chris in parentheses: Wisconsin (8, 4-seed), Utah State (10, 10-seed), Baylor (11, 5-seed), Missouri (11, 10-seed).
So, in summary, if you're looking for a tournament champion, I'd zero in on Kansas or Duke. At least that's what the statistics are telling us.