"My rule was I wouldn't recruit a kid if he had grass in front of his house.
That's not my world. My world was a cracked sidewalk." —Al McGuire

Friday, February 20, 2009

Predicting the Rest of the Season


I wanted to take a look at the predictions for the rest of the season.

First question - with every team having at least two losses, how many wins will take home the BE championship?

Looking over here on the right, we can see that the most likely scenario (54%) is that 15 wins will be the best record at the end of the season. There's a 32% chance that one team will win out and have the best record, and a slim chance that every team beats each other up for 14 wins.

16 wins for Marquette is a pipe dream (2%). 15 wins is a slim possibility (15%), and 14 wins is about a 50/50 shot. Although Marquette may have had a chance at these types of wins before, it's not very likely now.

Second question - how does Marquette stack up against the other top teams in the BE?

Okay, it's a bit of an eye chart, but here's the information.

  • Pitt - The most likely scenario for both Pitt and UConn is that they finish with fifteen wins, followed closely by fourteen wins. However, Pitt has more chances of getting to sixteen wins.
  • UConn - While Pitt may have the most upside, UConn has the best chance of getting to fourteen wins or more. Their most likely scenario is also fifteen wins, although fourteen wins happens almost as much.
  • Louisville - Louisville's most likely view is that they'll finish with fourteen wins. They have a good chance of getting to fifteen wins as well, but their second most likely view is finishing with thirteen wins.
  • Marquette - As mentioned, Marquette's most likely views are either 13 wins or 14 wins, with the chances of either being about 50/50.
Net predictions - There's a lot of variability here, but if I were to predict the final standings, here's where I see things shaking out.
  1. Pitt - the most favorable schedule
  2. UConn - solidly at number two or better
  3. Louisville - wildcard, but they have some tough games coming up
  4. Marquette - see: schedule, brutal
Third question - If Marquette's chances of finishing better than fourth are slim, how solid are the odds of finishing with the Bye-Bye?

For this, I looked at the predictions for Marquette versus Villanova. Villanova's most likely view is that they finish with 12 wins, followed by 13 wins. In other words, Villanova's upside of 13 wins is less likely than MU's downside of 13 wins.

When we consider Pitt, UConn, and UL in the mix, what are the odds of finishing with the worst record in the group? According to my calculations, Villanova has an 85% of having the worst record (missing the Bye-Bye). Marquette is next, with a 25% chance, followed by UL (18%). (Teams can tie for the worst record). Pitt and UConn are both less than 5%. There's still a one in four shot MU misses out, but that's not nearly as high as for Villanova.

Summary
  • In conclusion, it's looking like 15 wins (or even 16 wins) will be required to win the BE
  • Marquette has only slim odds of getting to that level, or finishing ahead of Pitt, UConn, or UL
  • However, things look solid for getting the Bye-Bye
Obviously, everything starts with playing well this weekend. However, this all leads to one final question... What NCAA seed do you think Marquette would get from finishing fourth in the BE?

4 comments:

James said...

Since the Wade era our NCAA appearances have been depressing at best with last year serving as a full fledged ripping out of the heart. I believe we cannot afford to be higher than a 3 seed. Marquette needs some cupcakes in the first 2 rounds which will give the dynamic trio some periods of rest. Once you get above a 3 into the 4,5,6,7 range there becomes too much parrody and anything can happen. Lets take care of GT, UCONN, and Lville (I'm throwin Pitt in as a loss) and try try try for a 2 seed.

Tony said...

Thanks for the analysis, just what I was looking for.

Hunt said...

I echo those thanks, this is a great post. Thanks CS.

greyCat said...

Have linked to this one from over at Villanova by the Numbers Rob. Very nice work. A good many Villanova fans are counting on Marquette's slightly more difficult schedule to bring the Warriors "back towards" the 'Cats as the season winds up. I frankly felt the same way...until Nova traveled to West Virginia (that was a mess). Did you factor strength of schedule and home/road games into the analysis?

Anyway, speaking from the recent past, it is decidedly more fun to speculate over a bye-bye (and a Dance seed) than whether we make the Dance at all.