FanPost

Analyzing NCAA Tournament Performance By Seed

Hello, college basketball fans!

The NCAA Tournament begins this week and I'm sure most of you reading this are excited, especially since last year there was no NCAA Tournament.

Each of the teams in the NCAA Tournament is assigned a seed from 1 to 16, Several teams are assigned to a "First Four" game and would have to win a game to claim that seed. You can argue some teams should be seeded higher and others should be seeded lower. We always make a big deal about these seeds, probably making the biggest deal about the #1 seeds. USA Today posted an article discussing "Six seeds the NCAA Tournament committee got wrong on Selection Sunday".

The question then becomes how important are these seeds actually? Well we can look at past historical performance of NCAA seeds in the previous tournaments since the field expanded to 64 teams in 1985. There have been 35 NCAA Tournaments and 140 of each seeds (not double counting First Four teams). Here are the number of teams that advanced at least to each round for each seed per round.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Max
2nd Round 139 132 119 111 90 88 85 68 72 55 52 50 29 21 8 1 140
99.29% 94.29% 85.00% 79.29% 64.29% 62.86% 60.71% 48.57% 51.43% 39.29% 37.14% 35.71% 20.71% 15.00% 5.71% 0.71%
Sweet 16 120 89 74 66 47 42 27 13 7 23 22 21 6 2 1 0 140
85.71% 63.57% 52.86% 47.14% 33.57% 30.00% 19.29% 9.29% 5.00% 16.43% 15.71% 15.00% 4.29% 1.43% 0.71% 0.00%
Elite 8 97 64 36 21 9 14 10 8 4 8 8 1 0 0 0 0 140
69.29% 45.71% 25.71% 15.00% 6.43% 10.00% 7.14% 5.71% 2.86% 5.71% 5.71% 0.71% 0.00% 0.00% 0.00% 0.00%
Final 4 57 29 17 13 7 3 3 5 1 1 4 0 0 0 0 0 140
40.71% 20.71% 12.14% 9.29% 5.00% 2.14% 2.14% 3.57% 0.71% 0.71% 2.86% 0.00% 0.00% 0.00% 0.00% 0.00%
Total 413 314 246 211 153 147 125 94 84 84 83 72 35 23 9 1
Avg # Games 2.95 2.24 1.76 1.51 1.09 1.05 0.89 0.67 0.60 0.60 0.59 0.51 0.25 0.16 0.06 0.01
Per Seed

Let's start with the comparison between #1 seeds and #2 seeds. In the first round, only one #1 seed has ever lost (Virginia in 2018) while eight #2 seeds have lost in the first round. That seems like a lot more but over 35 years eight #2 seeds is just 5.71% of them. When you get to the Sweet 16, 85.71% of #1 seeds make the Sweet 16 while 63.57% of #2 seeds do, a difference of 22.14%. 69.29% of #1 seeds get to the Elite 8 vs. 45.71% of #2's, the gap widens between #1's and #2's. The gap narrows for the Final Four, 40.71% vs. 20.71%, exactly 20%, but still pretty big. In total, #1 seeds won 413 games in the first four rounds (games in the Final Four not included) while #2 seeds won 314. This translates to an average of 2.95 games in a year for #1 seeds vs. 2.24 games per year for #2 seeds. The #1 seed averages almost 3 games or an Elite Eight trip while the expectation of a #2 seed is barely above a Sweet 16. So the difference between a #1 seed and a #2 seed statistically historically is huge.

Now let's go to #2 vs. #3. The gap between 2 and 3 is wider than between 1 and 2 when it comes to 1st round games. Only eight #2 seeds lost in the first round but 21 #3 seeds have lost in the 1st round (15%). On the other hand, the gap between 2 and 3 isn't as wide in the Sweet 16. 89 #2 seeds made the Sweet 16 and 74 #3 seeds did (52.86% for #3's vs. 63.57% for #2's). There's a pretty wide gap between #2 and #3 for reaching the Elite 8 (45.71% for #2 vs. 25.71% for #3). The average # of games won for #2 seeds is 2.24 while #3 seeds averaged 1.76 games. The gap is still significant but the gap isn't as wide as it is between #1 and #2.

When we get from #3's to #4's, the gap narrows. #3 seeds won 246 games and #4 seeds 211 games. This translates to an average of 1.76 games for #3's and 1.51 games for #4's. The biggest gap between #3 and #4 is not surprisingly to get to the Elite 8. (25.71% vs. 15%). #4 seeds almost certainly have to beat #1 seeds to make the Elite 8 and 69.29% of #1 seeds get to the Elite 8 (85.71% of #1 seeds make the Sweet 16 meaning the #4's have to beat the best to get the Elite 8).

Here's a little bit of a surprise to me. You'd think #4 vs. #5 wouldn't be that big since they should match up in the 2nd round. But the gap between #4 and #5 is larger than the gap between #3 and #4! #4 seeds average 1.51 wins. #5's average just 1.09 wins. #4's by performance are closer to #3 seeds than they are #5 seeds! 29 #4 seeds get dumped in the 1st round, not that many more than #3 seeds (21). But 50 #5's have lost in the first round, an average of well over one a year! We talk a lot about the "5-12" upsets because on average at least one happens every year (it's actually approx. 1.42 upsets a year!). How about the Sweet 16? You would think they would be close to even because 4's and 5's should be close to even footing in the 2nd round. But 47.14% (slightly less than half) of #4 seeds made the Sweet 16 but only 33.57% of #5 seeds (barely more than a third!) made the Sweet 16. The gap between #4 seeds and #5 seeds in the Sweet 16 is 13.57%. The gap between #4 seeds and #2 seeds is 16.43%, that's not that much of a difference. How about Elite 8? 15% (21) #4 seeds made the Elite 8. 6.43% (9) #5 seeds made the Elite 8. And almost twice as many #4 seeds (13 to 7) made the Final Four than #5 seeds. The gap between #4 seeds and #5 seeds is .42 games/year. The gap between #5 seeds and #8 seeds (1.09 games to .67 games) is also .42 games/year!

Once you get below #4, the gaps shrink. The gap between #5 and #6? 1.09 games vs. 1.05 games. 90 #5's made the 2nd round, 88 #6's did. 47 #5's made the Sweet 16, 42 #6's made the Sweet 16. #6's are actually more likely to make the Elite 8 (14 #6's vs. only 9 #5's). On the other hand, 7 #5's made the Final Four vs. only three #6's.

There are slightly more significant gaps between #6 and #7 (1.05 games vs. 0.89 games) and #7 and #8 (0.89 games vs. 0.60 games). Between #6 and #7, there is hardly any difference in winning your first game (88 vs. 85), making the Elite 8 (14 vs. 10) or making the Final Four (3 each) but a huge difference in getting to the Sweet 16 (42 vs. 27). #6's seeds are most likely getting #3's in the 2nd round vs. #2's for the #7's and #3's are a lot more likely to be upset in the 1st round. Now between #7 and #8, there is a big difference in the first round (a statistical weirdness is that #9's actually have outplayed the #8's, 72-68). #7's are also twice as likely to make the Sweet 16 (27 vs. 13) because of course you're comparing a potential #2 to a potential #1. After that, #7 vs. #8 is negligible.

Even though #9 seeds have a winning record vs. #8 seeds, #8 seeds have outperformed #9's the rest of the tournament. Neither are likely to make the Sweet 16 because both likely have to beat #1 seeds but 13 #8 seeds have done it and only 7 #9's have. In the Elite 8, 8 #8's have gotten to the Elite 8 while only 4 #9's have won three games. Finally 5 #8 seeds have actually gotten to the Final Four (more than #6's and #7's) while 1 (ONE) #9 seed has.

I've heard people think #10 and possibly #11 seeds are better than #8 and #9 seeds. Statistically, #10 seeds have 84 games, exactly the same number as #9 seeds. #11 seeds have won 83 games, one less than #9's. #8 seeds have won 94 games. So you definitely want to be #8 over #9, #10, and #11 and there isn't any statistical difference between #9, #10, and #11. Now there is a significant advantage to being a #10 and #11. Not surprisingly, it's the Sweet 16 because 8's and 9's have to play the #1 seed in the 2nd round. 23 #10 seeds and 22 #11 seeds have made the Sweet 16 compared to just 7 #9 seeds and 13 #8 seeds (in fact 21 #12 seeds have made the Sweet 16!) But while #8's and #9's are practically 50-50 to win their first game, #10 seeds are 39.29, #11 seeds are 37.14%, and 12 seeds are 35.71%. So if your goal is to win one game, you would rather be an #8 or #9. But if you want to make the Sweet 16, yes you'd rather be a #10, #11, or even a #12.

#9 seeds are the statistical exception as #10's and #11's are more likely to make the Elite 8 (8 each) than #9's (although 8 #8's also made the Elite 8). In terms of the Final 4, #11 is the seed you want to be as four #11's made the Final Four vs. just 1 #9 and 1 #10. Once you get past the Sweet 16, if you are below #11 you are done (one #12 made the Elite 8, no one below #11 made the Final Four).

Usually the top "one bid" conferences get #12 seeds. That is relatively huge for the conferences that can get it. 50 #12's won a game vs. 29 #13's. 21 #12's made the Sweet 16 vs. 6 #13 seeds. Overall, #12 seeds won 72 games while #13 seeds won 35 games, less than 1/2. #13 vs. #14 doesn't look like that big a deal (29 1st round wins vs. 21 although #13's have 6 Sweet 16's vs. just 2 for #14 seeds). #14 seeds have a lot more of a chance to win a game than #15 seeds (21 vs. 8) and #15 seeds have a lot more of a chance than #16 seeds (8 vs. 1) although is a 15% chance of winning a game in the NCAA Tournament that much different than a 5.71%?

So this tells you the comparison between various seeds. If you're a #6 and you feel you should be a #5, it's probably not a big deal. But if you're a #5 and you feel you should be a #4, that is probably more of a big deal. We make a big deal about #1 seeds and rightfully so. This isn't counting the champions. The last three national champions (North Carolina, Villanova, and Virginia) were all #1 seeds and 22 of 35 (way more than half) of the national champions have been #1 seeds.

#8 (1): Villanova (1985)

#7 (1): Connecticut (2014)

#6 (1): Kansas (1988)

#4 (1): Arizona (1997)

#3 (4): Michigan (1989), Syracuse (2003), Florida (2006), Connecticut (2011)

#2 (5): Louisville (1986), Duke (1991), Kentucky (1998), Connecticut (2004), Villanova (2016)

#1 (22): Everyone else.

User-created!