Tournament Stats Since 1985

Updated through 2003

Who should you pick?  Well, the tables below might be of some assistance.  Feel free to consult them for guidance if you wish.  Or just for enjoyment.  Or look at them and wonder when Walt is going to get a life.  One caveat though:  using the stats below is no guarantee that your pool performance will improve; in fact, it could get worse!  (Another caveat:  these stats are from my personal records; I do not vouch for the absolute veracity of the numbers).

The stats have been compiled since 1985, the year the tournament went to 64 teams - 19 years total.  For each seed matchups there are 76 results for the first four rounds (4 regions/year  x 19 years = 76).  For the final game there are 38 results (2 teams x 19 years); of course, there have been 19 champions.
 

First round results since 1985.  76 games total (1 in each region x 4 regions/year x 19 years).  The final column is % won and record predicted by a simple model explained below.
Seed Pairing
Record of favorite
% won by favorite
(actual)
Predicted favorite win %
(and predicted record)
1 vs. 16
76-0 
100.0 
94.1 (72-4) 
2 vs. 15 
72-4 
94.7 
88.2 (67-9) 
3 vs. 14 
63-13 
82.9 
82.4 (63-13)
4 vs. 13 
60-16 
78.9 
76.5 (58-18)
5 vs. 12 
52-24 
68.4 
70.6 (53-23)
6 vs. 11 
52-24 
68.4 
64.7 (49-27)
7 vs. 10 
45-31 
59.2 
58.8 (45-31)
8 vs. 9 
34-42 
44.7 
52.9 (40-36)

Listed in the first table above is the frequency of wins by favorites (higher seeds) over the 76 first-round games between each pairing over the last 19 years.  The hardest games to pick are the first round upsets.  How hard?  For reference, in the right column is a predicted winning % and record over the 76 games by the higher seed based on a very crude model.  The model works as follows:  Notice that all seed matchups add up to 17 (1 vs. 16, 8 vs. 9).  Assume that the seed is an indication of how many times out of 17 that seed would be expected to lose.  So, a #8 seed would be expected to lose 8 times (and win 9 times) every 17 games.

Remarkably, this simple model works amazingly well, outside of a couple exceptions.  First, a #1 seed has never lost, though the model predicts 4 such losses since 1985; this is actually the number of #2 seed losses (which is predicted to be 8 or 9 losses).  The largest deviation from expected occurs for the #8 vs. #9 matchup, where the underdog #9 seed has actually won more total games.  However, actual results for seeds #3 through #7 are quite close to the statistictally predicted results - within ~2-3% (1 or 2 games over 76 games for each seed pairing).  Thus, the model only errs substantially for the most mismatched teams and the most evenly matched teams!  One question might be, is this a statistical fluke that just comes out this year?  Perhaps, but I did this about five years ago and found similar agreement.

Why does the model break down for the 1-16 and 2-15 matchups?  In terms of the mismatches, the worst 6-8 teams (#15, #16 seeds) are all automatic bids from very weak conferences; these teams are usually not anywhere near the top 57-64 teams (often their rated in the 100s); so there is much more of a mismatch than the seeds indicate.  Why are the 8-9 matchups off?  I have no idea!

One more note, #6 seeds do better than expected and actually match the performance of #5 seeds; so be wary of picking a #6 seed to get upset!  So, now you have the necessary information to decide which seeds are most likely to get upset.  Of course, knowing that it's likely that at least one #5 seed will lose to a #12 seed this year doesn't help one determine which one of the #5 seeds will be the one to get upset!

Now, what about the second round?  See below:
 

Number of seeds to win 2nd round games 
(i.e. to make the Sweet 16).  The avg. frequency indicates, on average, how many teams of each seed (4 total) make it to the Sweet 16 each year.
Seed
# of occurances
(of 76 possible)
%
Avg. Frequency per year
(max. 4 times per year)
1
66
86.8
3.47
2
49
64.5
2.58
3
35
46.1
1.84
4
34
44.7
1.79
5
26
34.2
1.37
6
29
38.2
1.53
7
12
15.8
0.63
8
8
10.5
0.42
9
2
2.6
0.11
10
15
19.7
0.78
11
10
13.2
0.53
12
13
17.1
0.68
13
3
3.9
0.16
14
2
2.6
0.11
15
0
0.0
0.0
16
0
0.0
0.0

Looking at the 2nd round winners (those making the Sweet 16), more intriguing things show up.  While a #9 seed beats a #8 seed more often than not in the first round, don't pick a #9 to make the Sweet 16. #8 seeds are far more successful in the 2nd round (in absolute numbers even though they get to the 2nd round less frequently than #9 seeds).  In fact, getting a #9 seed, while giving a team a chance to win one game, is a virtual guarantee that the team will be gone by the end of the first weekend.  You're better off picking a #13 or even a #14 seed to make the Sweet 16.

How does one explain this?  A #9 seed, having just won a very tough first round game, faces a nearly impossible task of playing a #1 seed with only one day to practice for them.  Lower seeds face an easier 2nd round opponent and have momentum and perhaps a psychological advantage (the Cinderella effect) on their side.  Often, there may be multiple upsets, so that a #12 seed actually becomes a favorite against #13.  In fact, once the #12 seed wins their first round game, more than 50% of the time (13 out of 24) they've won their next game and gone to the Sweet 16!  And now, the third round and beyond:
 

Number of seeds to win 3rd round games 
(i.e. to make the Final 8)
Seed
# of occurances
(of 76 possible)
%
Avg. Frequency per year
(max. 4 times per year)
1
53
69.7
2.79
2
35
46.1
1.84
3
17
22.4
0.89
4
12
15.7
0.63
5
4
5.3
0.21
6
11
14.5
0.58
7
4
5.3
0.21
8
5
6.6
0.26
9
1
1.3
0.05
10
6
7.9
0.32
11
3
3.9
0.16
12
1
1.3
0.05

 
Number of seeds to make it to the Final Four and beyond 
(to date no seed lower than 11th has made it this far)
Seed
Reach Final Four
(76 total)
Reach Final Two
(38 total)
Win Championship
(19 total)
1
33
17
11
2
16
8
3
3
10
6
2
4
7
2
1
5
3
2
0
6
3
2
1
7
0
0
0
8
3
1
1
9
0
0
0
10
0
0
0
11
1
0
0

Back to the main tourney page