[Cracked Sidewalks] Providence: It's all about protecting the ball

[CrackedSidewalksSays]

Providence: It's all about protecting the ball

Written by: noreply@blogger.com (Rob Lowe)

Kind of a crazy game on Saturday.  Very good win for the team, especially when it looked for a long stretch that it just wasn't Marquette's night.




First of all, it's that pesky problem of being bad defensively on the road that pops up again.  This was a problem all last year, and we're continuing to track it.    However, the goofy thing about this game was that Marquette lost on three of the four factors and yet still managed to win the game.  The defensive eFG% was outright, throw-your-remote-through-the-TV bad.  Despite the rash of early foul calls during the 1H, Marquette did not get to the line at a higher rate than Providence.  And, Marquette allowed PC to get a higher offensive rebounding percentage.

Nope, it was all about the turnover rate.  Not only was Marquette able to force Providence into a higher turnover rate than predicted, but they were able to take care of the ball remarkably well.  In the second half, Marquette had only three turnovers to PC's 13.  Just as commentator Bob Valvano said, that was the key to the game right there.

Despite all the giddyness (and deservedly so) from a big road comeback victory, is it time to start expecting to compete legitimately for the BE championship?  Here are the current set of predictions for the rest of the season.




With five games under the belt for most of the teams in the league, the set of data is starting to get better.  However, it's still a reflection of how well the team is playing right now.  Since they're playing well, that helps.

Home wins (high confidence) - DePaul, St. Johns, Seton Hall, Syracuse
Home wins (tougher games) - Georgetown, UConn

Road wins (high confidence) - NONE
Road favorites (tougher games) - DePaul, USF
Road underdogs (but winnable) - Notre Dame, Villanova
Unlikely wins - Georgetown, Louisville, Pittsburgh

If anything, I think the predictions for the home games against Syracuse and Georgetown are a bit high, and the matchups on the road for GU, UL and Pitt are a bit too low.  However, it all passes the sanity check.  Really, the big difference from the last time is that Marquette won two games (WVU and PC) that were considered "tougher" games.

How do things look for the rest of the season?



Start trying to figure out where Marquette will end up playing in the NCAA tournament.  There is less than 1% chance the team ends up worse than 9-9, and 96% chance the team ends up with 10 wins or more.  The most likely view is still that Marquette finishes with twelve wins, with a 70% chance that MU does better than last year's eleven wins.

However, as the team is playing right now, there is basically a 50/50 chance the team has 13 wins or more at the end of the season.  As a reminder, Georgetown won the league with 15 wins last year.  UL and UConn were right behind at 14 and 13 wins, respectively.  Everyone else bunched up at 11 wins after that, including Marquette.  What are those odds?

• 15 wins - 10%
• 14 wins - 24%
• 13 wins - 46%

Is it too early to think that Marquette can compete for the Big East championship?  Maybe, but it's definitely not a crazy notion anymore.  This year's team is not only fun to watch and fun to root for, but they could be special.

http://www.crackedsidewalks.com/2009/01/providence-its-all-about-protecting.html

[🏀]

Great stuff as always.

[Henry Sugar]

Thanks.  By the way, here are the results from the preview recommendations and the impact.  I'm putting these up because ATWizJr asked about them last game.

http://www.crackedsidewalks.com/2009/01/marquette-travels-to-providence-on.html

Specific Statistics Recommendations (In order of importance)
1. Limit Providence to an eFG% of 50% or less, which is right at their season average and gives 61% chance of Marquette win (3 point win)

Result – Providence eFG% was 61.8%.  Advantage Providence.  Changed margin of victory prediction from plus three to minus eight

2. Achieve an eFG% of 51% or higher, which is slightly less than the season average

Result – Marquette eFG% was 57.5%.  Advantage Marquette.   Changed margin of victory prediction from minus eight to minus two points

3. Force Providence to a turnover rate of 24% or higher (~17 turnovers), which is three more than their average

Result – Providence had a turnover rate of 24%.  No Advantage.  No change to margin of victory

4. Hold PC to an offensive rebounding percentage of 31% or less (~12 offensive rebounds), which is two less than their season average

Result – Providence OR% was 33.3%.  Slight advantage Providence.   No change to margin of victory prediction

5. Get to the line with a Free Throw Rate of 38% or more (~22 free throw attempts), which is six less than season average

Result – Marquette FTR was 32.8%.  Advantage Providence. Changed margin of victory prediction from minus two points to minus three points.

Okay, so going by the top five recommendations, we should have lost.  However, when I did the analysis, there was a sixth key recommendation.  Actually, the model I use delivered eight keys, but the last two were tiny and at some point the keys need to be trimmed up.

6.   (Not given) – Commit a turnover rate of 21% or less
Result – Marquette TO% was 12.7%.  Advantage Marquette.  Changed margin of victory prediction from minus three to plus eight.

I was going to load the sensitivity analysis to show that I'm not making up this sixth recommendation.  However, evidently I'm not smart enough to manipulate photobucket images.  You'll just have to trust me.

[CTWarrior]

[With five games under the belt for most of the teams in the league, the set of data is starting to get better.  However, it's still a reflection of how well the team is playing right now.  Since they're playing well, that helps.

Home wins (high confidence) - DePaul, St. Johns, Seton Hall, Syracuse
Home wins (tougher games) - Georgetown, UConn

Road wins (high confidence) - NONE
Road favorites (tougher games) - DePaul, USF
Road underdogs (but winnable) - Notre Dame, Villanova
Unlikely wins - Georgetown, Louisville, Pittsburgh

If anything, I think the predictions for the home games against Syracuse and Georgetown are a bit high, and the matchups on the road for GU, UL and Pitt are a bit too low.  However, it all passes the sanity check.  Really, the big difference from the last time is that Marquette won two games (WVU and PC) that were considered "tougher" games.

How do things look for the rest of the season?

Of the home wins (high confidence), I would remove Syracuse.  That's a 50-50 game at best.
UConn is much less than a 50-50 proposition for us.  We have no one who can match up with Thabeet, Robinson or Adrien. 

On the other hand, I would move DePaul to the road win (high confidence) bucket.  They appear to be the weakest team in the Big East, and maybe by a good margin.  I see 4 highly likely wins (DePaul, at DePaul, St. Johns and Seton Hall), one likely win (at USF), 5 pick 'em/likely losses (Syracuse, Georgetown, at Notre Dame, at Villanova, at Louisville) and 3 highly likely losses (at Georgetown, at Pittsburgh, UConn).

So we win the 5 highly likely/likely wins, find a way in 2 of the pick'em/likely loss bucket, lose the other 3 and finish 12-6.

Best Case 14-4, worst case 9-9.

[Henry Sugar]

I just go according to the numbers, and the numbers don't reflect matchups other than turnovers/offensive rebounding/etc.  Also, I think that the Syracuse and UConn numbers are a little high too.  However, don't forget that teams just play better at home and worse on the road, and the advantage is that both of those games are at home.

Marquette
Offense (116.8 Ortg H ; 108.8 ORtg A)
Defense (93.7 Drtg H ; 102.5 DRtg A)

UConn
Offense (116.7 Ortg H ; 111.0 ORtg A)
Defense (91.2 Drtg H ; 96.9 DRtg A)

Syracuse
Offense (112.7 Ortg H ; 108.2 ORtg A)
Defense (96.3 Drtg H ; 101.8 DRtg A)

[ATWizJr]

Sugar, thanks for the post game comparisons.  I always find it interesting to compare the predictions with the actual outcomes to help explain the result.

[ATWizJr]

Sugar, love this analysis. PLease post it after each game.

Thanks.