Can OSU make it to the Final Four

OKC Dave —  February 28, 2013 — 10 Comments
Photo Attribution: USATSI

Photo Attribution: USATSI

In Part 1, I looked at NCAA seeding.

In Part 2, I looked at what it takes to advance in the tournament.

In Part 3, I’ll try to predict how far OSU will advance in the tournament.

I had fun with this part. There are lots of ways to approach this, and I’ll just be upfront with you — none of them really matter.

Trying to predict how far a team will advance in the NCAA tournament is a fool’s errand. The beauty of single elimination March Madness is that you never know what crazy things can happen. In 2003, a solid-but-not-great OSU team had Syracuse down by 17 at one point in a 2nd-round game. It wasn’t Carmelo that led the Orangemen roaring back to a win, but everyone remembers his performances in the rest of Cuse’s games as he played his way to a national championship and a #3 overall draft pick.

Here’s how I attacked this question. Out of the group of 266 NCAA tournament teams over the last four seasons, I found the 25 teams most similar to OSU in offensive efficiency, defensive efficiency, and Pomeroy rating #. These teams have a very similar makeup to OSU’s team: good-to-great defense, average-to-good offense, and a relatively high overall Pomeroy rating.

Here’s the list. The team listed first (after OSU) is the most similar to our current team:

I was kind of thrilled with these results. Look at the top 5 most similar teams to OSU. Sure, Temple lost in the 1st round in 2010. But the other four teams advanced at least to the Sweet 16. Tennessee made the Elite 8 in 2010 (and was a bucket away from the Final Four). Butler was a few inches away from winning the whole thing.

Out of this group of 25 teams similar to OSU, here’s where the teams were eliminated:

Eight out of the 25 teams made it at least to the Sweet 16, so that tells us we’ve basically got a 1-in-3 chance of advancing that far.

As I showed you in Part 2, we don’t quite fit the profile of a Sweet 16 or a Final 4 team — while our defense is outstanding, the offense might just not be good enough to take us the distance. However, this analysis shows that a team with our makeup is very capable of making a run. We certainly have the guard play to do it.

One more thing…I’m happy to play with formulas in Excel all day long, but does anyone really want to wager against Marcus Smart?

  • Andrew P

    Great stuff Dave

  • Doug

    Amazing how fun it is to geek out on college hoops when OSU is a contender….

    I looked at this a couple of different ways.

    1. Look at all #5 seeds over the last 20 years (easy enough to do here: http://www.sports-reference.com/cbb/play-index/tourney.cgi ).

    2. I used pomeroy’s rankings and the bracket matrix ( http://bracketproject.50webs.com/matrix.htm ) to simulate the tourney a bajillion times.

    Here are the results, compared with your similarity method.

    Chances of making round of 32. Historical: 64%, PomSim: 68%, Similarity: 72%
    Chances of making round of 16. Historical: 33%, PomSim: 36%, Similarity: 32%
    Chances of making round of 8. Historical: 8%, PomSim: 15%, Similarity: 12%
    Chances of making round of 4. Historical: 8%, PomSim: 7%, Similarity: 12%
    Chances of making round of 2. Historical: 4%, PomSim: 3%, Similarity: 4%
    Chances of championship. Historical: 0%, PomSim: 1%, Similarity: 0%

  • C.L.

    It’s easy — All they have to do is win-out !!

  • OSUaggie

    I’ve been the company stat guy for a lot of years, systemic problem solving, multi-variant regression analysis, etc, etc….Dave, I feel as a mere boy amongst men in your presence….applying solid statistical work against cowboy athletics is a LOT more fun than fixing problems at work…..I feel the need to back out of your presence……..really nice work dude, really nice….

    • http://gravatar.com/okcdave @okc_dave

      I appreciate that but I assure you it’s not true. This is just the intersection of two of my loves…OSU and working in Excel. I’m sure “real” stat people would find problems with my work but I really try to present things in a way that “normal” people can understand (normal = people that don’t think everything belongs in a spreadsheet).

      Thanks to everyone for the kind words. You can expect more of this type of stuff from me.

      Dave

  • http://gravatar.com/carsonc5 carsonc5

    You have outdone yourself with this 3-part series, Dave. Well done.

  • Newbs

    Really good series. Here’s to hoping for a Sweet 16 and more!

  • Keith

    Maybe a Part 4 showing how OSU compares to the expected field of 68 this year?

    • http://gravatar.com/okcdave okcdave

      I’ll wait until the field is announced to work on that.

  • Justin

    One item that might be worth addressing? Are our offensive or defensive effeciency numbers trending one direction or another? Nash has really picked up his play as of late, Smart has gotten better with experience, and we are finally getting healthy. I imagine that our numbers are improving. It certainly feels like we are much better now than we were at the start of conference play.

    Justin