NCAA Tournament Historical Seeding Analysis

By: Dylan Elder – Editor-in-Chief.

Hopefully by now you’ve filled out at least one bracket, or else you probably wouldn’t be reading this article.

To help you out with your picks, I went back through the past 10 NCAA tournaments and made a chart for each one, showing how far each seed advanced. No school names attached, just numbers.

For our purposes, I will only be using the past 10 years of data, which is more than enough to attribute to this years’ version of March Madness. Let’s take a look at the breakdown, round by round.

Picking Round of 64:

Screen Shot 2014-03-17 at 10.30.51 PM

What it means for Seeds 1-4 and 13-16:

The one fact that everyone knows about March Madness is that a 16 seed has never beaten a 1 seed, which gives them, relatively speaking, a 0% chance to do so in the future (that’s not completely true, but we are basing our picks off of history here). However, since three 15 seeds have knocked off 2 seeds, that gives them a 7.5% chance. Combine that with the chances of a 14 or 13 seed pulling off an upset and the total percent chance of one of the top four seeds falling in round 1 is a whopping 44% chance.

In fact, all top 4 seeds haven’t advanced to the round of 32 since 2007, when Bush was still President. Another way of saying that: a 13 seed or lower has won a game in each of the last 6 tournaments. It’s up to you to find this years’ diamond in the rough.

My advice? Take a flier on an upset from one of the 13 or 14 seeds this year, history is on your side… 44% of the time. Your best bets this year are UL Lafayette over Creighton and New Mexico St over San Diego St.

What it means for 5 vs 12:

While the best stories come from the lower seeds, these mid-tier groupings have quite the fascinating trends themselves

The ever coveted 5. vs 12. match ups are the ones that prognosticators salivate over. And with good reason, the difference in seeding may be 7 places, but the twelves upend the fives 42.5% of the time. Not only that, but in 9 out of 10 years (’07 being the exception) at least one 12 seed has made it to the round of 32. If you don’t have a 12 seed advancing in your bracket then you’re doing it wrong. Your best bets this year are NDSU over Oklahoma and Harvard over Cincinnati.

What it means for 6 vs 11:

Another interesting matchup is the 6. vs. 11. battle. While it doesn’t invite as many upsets as the 12. vs 5. behemoth, there’s still plenty of excitement to go around. The favorites win this game 5/8 of the time. However, if you split that fraction in 2.5/4 (because there are only 4 games with these two seeds) then it shows that every year, 1.5 teams underdogs win.

All that is just a long and fancy way of saying: pick at least one 11 seed to advance, they are bound to win at least one game. Furthermore, the last time an 11 seed failed to make it to the round of 64 was in 2004. I’m not betting against those odds. Your best bets this year are Providence over UNC and Iowa/Tenn play in winner over UMASS.

What it means for 7 vs 10:

Next up is the 7. vs 10. game, perhaps the most intriguing of them all. As you can see by the graph above, these seeds have evenly split their 40 games over the past ten tournaments. It goes without saying that despite the discrepancy in seeding, both teams are usually very evenly matched. Pick wisely when choosing you 10s. over 7s. but just know that there will most definitely be 10s. over 7s. Your best bets this year are St Joseph’s over UCONN and Arizona St. over Texas.

What it means for 8 vs 9:

Rounding out the round of 64 is the 8. vs 9. joust. History shows us that 8 seeds do win (albeit by a slim margin) most of the time in the first round of the tournament. Keep that in mind while making your picks, even if a 9 over an 8 doesn’t make for much of an upset. Your best bets this year are Pittsburgh over Colorado and Oklahoma St. over Gonzaga.

Final Key Question: How many upsets should I pick in the first round of my bracket?

Answer: Give or take 6-7 upsets, not including the 9. vs 8. games.

Picking Round of 32:

Picking the games in the round of 32 is exponentially harder than picking games in the first round because you don’t know what lasting effects from round 1 will drift over. Take a look at this graph:

Screen Shot 2014-03-17 at 11.55.14 PM

87.5% of all 1 seeds advance to the Sweet 16, so for those of you thinking that Kentucky will knock out Wichita State in the round of 32, think again. Similarly, the 67.5% mark for all 2 seeds to get to the sweet 16 is actually quite advantageous, and picking chalk through the first 2 rounds (up to the first two seeds) is a smart play.

I’m not telling you anything you don’t already know. Basically the 1 and 2 seeds are good at basketball. But get ready for something a little more interesting:

When many people fill out their brackets for the first time, they pair up the best 4 teams in a single region quite often. An example of that this year would be putting Florida, Kansas, Syracuse, and UCLA all into the sweet 16 from the South Region. However, in 40 instances over the past 10 years, that 1,2,3,4 seeding has only happened 7 times.

Instead of chalk over the past ten years, 34 teams seeded 6 or lower made the sweet 16. That’s an average of 3.4 per year. Of those 34 teams, ten were 6 seeds, so be sure to pounce on those 6. vs 3. upsets in the round of 32.

But what does all that data mean?

In your sweet 16 field, you should have about 3-4 teams that are seeded 6 or lower. Use the red graph above to determine which seeds have been considered “sweet” over the past 10 years.

-25% of 6 seeds advance to the sweet 16 (Easiest paths: Ohio State, Baylor)

-17.5% of 12 seeds advance to the sweet 16 (Easiest path: North Dakota State)

-15% of 7 seeds advance to the sweet 16 (Easiest path: New Mexico)

-15% of 10 seeds advance to the sweet 16 (Easiest path: St Joseph’s)

I recommend picking at least 3 of the 5 schools listed above to advance to the sweet 16.

Picking Sweet 16:

Once a team gets into the Sweet 16, they have undoubtedly knocked off some impressive competition and are deserving of the “sweet” title. However, picking (for us) and advancing into the Elite 8 (for the players) is a whole different animal. This is where the top talent starts to separate itself from the pack and the cinderella stories try to hold on for the ride.

Screen Shot 2014-03-18 at 1.06.07 AM

Even though over the past 10 years, 65% of one seeds have advanced to the elite 8, it’s still a semi-trap. Only 3 years over this period have all four number 1 seeds advanced to the elite eight, and never since 2009. At this point, it’s safe to eliminate at least one of your one seeds now. Your best bet to eliminate is Wichita State.

Almost half of the 2 seeds (19/40) advance from the sweet 16 to the elite 8. However, not once in the entire 10 year period did all four 2 seeds win and move on. Likewise, there was never a time when zero 2 seeds advanced to the next round. Anywhere from 1-3 advancing is a good number. Your best bet to win here is Michigan.

If you combine the 3 and 4 seeds, 19/40 of them also progress to greener pastures, so using the same rule of taking 1-3 of these guys applies here as well. Your best bets are Duke, Louisville, and Michigan State.

So far, let’s say you took 7 teams from the 1-4 seeds (per my suggestion). That leaves you with one more open spot in the elite 8. Well, if you look again at the orange graph, you’ll notice that 15 teams seeded 5 or lower have made the elite 8 since 2004, including each of the last 4 tournaments and 8 of 10 overall.

Picking this cinderella team is often the hardest part of filling out your entire bracket. If you don’t have the guts to do it, then it is completely understood. However, if you choose to have a 5 or lower seed enter you elite 8, you just might be rewarded handsomely. Your best bets here are Oregon and New Mexico.

Picking the Elite 8

Reaching the Final Four is every college basketball players’ dream. Picking the final four correctly is every prognosticators dream.

Screen Shot 2014-03-18 at 1.19.52 AM

As you can see, 37.5% (15/40) of 1 seeds have made the Final Four over the past 4 years yet there was only one instance where no 1 seeds made it (2006) or all 4 top dogs made it (2008). Normally it’s best to stay away from the outliers and stick with the high probability. Choose 1-2 number 1 seeds this year. Your best bets are Florida and Arizona.

2 seeds advance to the Final Four at a respectable 25% clip. However, there have never been 3 or more 2 seeds in the Final Four over the past 10 years. Your best bet is Michigan this year, but I don’t like any of the options. Stay away from 2 seeds in the final four if you can.

3 and 4 seeds grouped together make the Final Four at a 20% clip yet I love Louisville and Michigan State (both 4 seeds) to get that far. Your best bets are the Cardinals and Spartans.

Again our attention turn to the bracket busters. If you were brave enough to have New Mexico or Oregon in the elite 8, you might as well bring them along to the final four too. But I have my worries here.

Picking the Final Four

Enough with the silly graphs.

All you need to know to pick your two championship teams is that half of them (10/20) have been 1 seeds over the past 10 years. Your best bet here is Florida.

As for the other team, 8 seeded Butler was the lowest team to make the NCAA Championship in the last 10 years. That is more of an aberration than anything else, so stick with 2,3, or 4 seeds. Your best bet here is Louisville.

Picking the Championship Game

Easier said than done getting this far, but the winner of the NCAA Championship has been a 1 seed for 7 out of the 10 past years. Your best bet to win it all is Florida.