Basketball bracket tips

I got this bit of spam in the email but it’s actually sort of cool, would be an excellent topic for discussion in an intro stat class or a Bayesian class:

MEDIA ALERT: NCAA COLLEGE BASKETBALL TOURNAMENT – MARCH MADNESS
NCAA College Basketball Tournament Bracket-Picking Tips.
RJ Bell of Pregame.com, the top Las Vegas based sports betting authority, provides a simple blueprint to improve anyone’s bracket results.

1st round rules: Pick no more than two seeds worse than #12. #16 seeds are 0 for 96 in the modern era (since 1985). #15 seeds are 4 for 96. #13 and #14 seeds combined win only 18% of first round games.

Don’t be shy about picking upsets with #12 through #9 seeds. Give special consideration to picking #12 seeds: they have won 13 of 32 matchups vs. #5 seeds the last eight years. #9 seeds have won over half their games (54%) vs. #8 seeds.

2nd round rules: Advance #1 seeds almost automatically into the 3rd round. #1 seeds win their first two games 87% of the time.

Keep advancing the #12 and #10 seeds you picked to win in Round One. These teams win over half the time in Round Two.

Seeds lower than #12 DO NOT win in the 2nd round. Only 6 of 368 teams that have advanced past Round 2 were seeded lower than #12.

Sweet 16 round rules: Advance exactly three #1 seeds into the Elite 8. There’s no reason to buck the math: nearly 72% of #1 seeds advance into the 4th round (that’s a higher percentage than #5 seeds who win a single game)!

Advance no team lower than a #11 seed into the Elite 8 (22 have made it to the Sweet 16, but only 1 has ever advanced).

Elite Eight round rules: Advance exactly ONE or TWO #1 seeds to the Final Four. Amazingly, exactly one or two #1 seeds have made the Final Four 19 of the last 24 years. Last year was the first with all four #1 seeds making the Final Four; do not allow a single season fluke to outweigh over two decades of history.

Advance no team lower than a #8 seed to the Final Four. Only 2 of 96 Final Four teams have been seeded lower than #8.

Final Four round rules: Advance NO team below a 6th seed to the Championship game. Not a single one has made it in the last 23 years.
Championship game rules: pick a #4 seed or higher to win it all. For 20 straight years the champion has been a #4 seed or higher!

I know so little about college basketball that I don’t have much in the way of comments. I’m guessing that some of the above suggestions make sense and some are silly. But, as I said above, it could be lots of fun to discuss this all in a stat class.

7 thoughts on “Basketball bracket tips

  1. Stat? Well, probability. And what kinds of inferences you can make from stats.

    It's a single elimination tournement with 64 teams. That's straight forward. The odd thing is that each quarter of the tournment is given it's own 1-16 seeds, as opposed to and overall 1-64.

    Usually, 3 of the 4 top seeds make it to the semi-finals (called the Final Four). What should you pick? Does this mean that YOU should guess which 3 — what is implied above? Should you pick all four of the top seeds to make it that far (the smart way to go, mathematically)? Should you pick more upsets — because part of the fun is getting picks right that no one else did?

    But here's the part that most people don't think of: Where is your competitive advantage? Are you more likely to win by going mainstream, but just trying to do it better, and therefore beat the pack in in tight right? Or, are you more likely to win by taking a lesson view? With this latter option, you've got a clear path to the win if it works out, and not chance if it doesn't.

    I could go on and on on this stuff.

  2. With the exception of picking #9s over #8s, Most people would do better to just fill out their brackets by seed. (if you ignore any penalties for ties)

    Even the #12 vs. #5 thing, so what if the #12s win more than someone thinks they should, they're more likely to lose than win.

  3. I would love to see is some of the relationships behind these odds… e.g. do 12-5 upsets happen more often when the #12 team is strong defensively, or keeps the pace of the game slow (low scoring)?

    Is there a significant home-region court advantage if the #12 or #10 team is closer to home than the opponent?

    …and for the blog's sake — maybe there is a distinct advantage for a team to score an upset if it is from a blue state rather than a red state? :)

  4. The number one way, and easiest too, to figure out who to pick in the first round is look at the betting lines. For example, a #10 seed may be a favorite over a #7 seed, then just go with the the betting favorite and forget the seedings which are irrelevant. Also, if a worse seed is only a slight underdog, then that's useful. For example, a #12 seed this year (Arizona) is only a 1 point underdog to a #5 seed. Might as well pick the #12 seed in that case as you'd expect more people to pick the #5 seed, and you aren't losing much in expectation (being different in a contest with a lot of players is good…if its only 5-10 guys, then positive EV on each game is what you want).

Comments are closed.