BPI projects how the 2020 NCAA tournament could have played out

Hier informiere ich über Neuigkeiten, die die Entwicklung meiner Kinder betreffen.

Moderatoren: Angela, merin

Beiträge: 5
Registriert: Freitag 20. März 2020, 08:46

BPI projects how the 2020 NCAA tournament could have played out

Beitrag von Bellinger12 » Freitag 20. März 2020, 09:05

It will be remembered as the NCAA tournament that never happened.

But what if it had?

That's what we're here to settle.

Armed with Joe Lunardi's final bracket and our own Basketball Power Index (BPI), we decided to simulate the 2020 tournament.

Normally when we do this, we're providing probabilities. We tell you that Kansas was the pre-tournament favorite, with an 18% chance to win. That Baylor had a 7% (!) chance to lose in the first round to Winthrop. Or that Texas Tech was the double-digit seed with the best chance at reaching the Sweet 16 (18%).

That's good stuff, and if you want to see who the pre-tournament favorites to win it all were, just look at this list:

Kansas 18%

Duke 17%

Gonzaga 15%

Michigan State 10%

Dayton 7%

Baylor 6%

Ohio State 3%

San Diego State 3%

Louisville 3%

Maryland 2%

But here's the thing: Life is a single sim. The 2020 tournament was only going to be played once. And so that's what this article is actually dedicated to: a solitary simulation of this year's tournament. One-and-done, just like March Madness.

A single simulation means that upsets will happen. The unexpected will rear its head. Because any given Thursday-Sunday in March, there's some weird in college basketball. And that's reflected here.

In practice, here's roughly how that works. Let's take that Baylor-Winthrop matchup we mentioned earlier. BPI has determined that given the relative strength of the two teams and the location of the game (Omaha), Baylor has a 93% chance to win. So it effectively rolls a 100-sided die, and 93 of the sides show a Baylor win, and the other seven have a Winthrop win. The winner advances, and we go through the same process for every game.

And we made a pact, too. We decided to use simulation No. 2020 and use it no matter what. If it's the one simulation where Vermont wins it all, we're doing it. No takebacks. Because sometimes the wildly improbable does happen. You're just gonna have to trust us on this point.

Oh, and BPI doesn't provide any information about score or stats, so we're taking some liberties with the details along the way. OK -- the rules are set. Now, it's time for tipoff.