Saturday, March 30, 2013

Sweet Sixteen Update, Part 2

The second half of the Sweet Sixteen games have been played, so here’s the update on the Prediction Machines thoughts:
Miami (FL) Marquette 6 3.7 -2.3
Louisville Oregon 10 11.6 1.6
Ohio State Arizona 3.5 5.4 1.9
Indiana Syracuse 5.5 7.7 2.2
Duke Michigan State 2 4.3 2.3
Kansas Michigan 2 4.9 2.9
Wichita State La Salle 4 7.6 3.6
Florida Florida Gulf Coast 12.5 20.2 7.7

The PM went 1-2 on betting predictions.  The Louisville-Oregon prediction was too close for a bet, but the PM would have been on the wrong side there as well.

Michigan-Kansas was the surprise result to me – I really expected Kansas to win that game fairly handily.  They might have still squeaked out the win if not for some bone-headed late-game plays.

I’m traveling again today, so I don’t know if I’ll manage to get out Elite Eight predictions.

Friday, March 29, 2013

Sweet Sixteen Update (Part 1)

The first half of the Sweet Sixteen games have been played, so here’s the update on the Prediction Machines thoughts:
Miami (FL) Marquette 6 3.7 -2.3
Louisville Oregon 10 11.6 1.6

Ohio State Arizona 3.5 5.4 1.9
Indiana Syracuse 5.5 7.7 2.2
Duke Michigan State 2 4.3 2.3
Kansas Michigan 2 4.9 2.9
Wichita State La Salle 4 7.6 3.6
Florida Florida Gulf Coast 12.5 20.2 7.7
The PM went 2-1 on betting predictions.  In its Tournament bracket, it correctly picked Marquette over Miami.  Unfortunately, in the most significant game of the night for the PM, it was very wrong on the Indiana game.  Indiana struggled all night against the Syracuse zone, and (particularly in the first half) Syracuse seemed to make every shot attempt.  Having watched Syracuse’s first two round games at San Jose, I was expecting the zone to cause Indiana more trouble than expected, but I didn’t expect Syracuse to look so good on the offensive end.  At any rate, unless something very unusual happens, that result eliminates the PM from contention in the Machine Madness Contest.  This will be the first year where the PM didn’t go into the final game with a chance to win the contest.
The Arizona-Ohio State prediction was too close to recommend a bet, but the PM ended up on the wrong side of the line there as well.   Ross’s defensive blunder on the next-to-last play of the game probably cost quite a few gamblers a payout.

Wednesday, March 27, 2013

The Prediction Machine’s Bracket

The Prediction Machine is primarily focused on picking the margin of victory for regular season games, but I also use it to create a bracket for the Machine Madness Contest.  The contest has been going on for a few years, and my approach to picking a bracket has evolved.

Initially, the Prediction Machine picked the most likely winner of each game – whichever team it deemed stronger.  But there’s a serious drawback to this approach.  The Committee is already pretty good at determining the relative strength of the teams, so by and large the Prediction Machine’s picks agreed with the seedings.  It only differed where the Committee had “mis-seeded” teams.  That seems to happen every year, but there’s usually only one or two mis-seeds.  So you end up with a bracket that may be the most likely outcome, but which is also going to be very similar to many other brackets.  (In fact, we see that very thing in this year’s Machine Madness competition: “Danny’s Dangerous Picks” and “Predict the Madness” are identical after the second round.)  This makes it very hard to finish high in a pool with a lot of entrants.

In the next iteration, I forced the Prediction Machine to pick about 15% of the games as upsets.  I chose that number because historically, that’s about how many upsets there are each Tournament.  The Prediction Machine did this by ranking the upsets and selecting the top 6 upsets in the first round and 5 more in the rest of the tournament.  The idea was to get away from the consensus picks of the other competitors while picking the most likely upsets.  But this is too risky a strategy.  Depending upon the size of the pool, you probably don’t need to get 11 upsets correct to do very well.  For example, in last year’s Machine Madness pool, it would have been sufficient to get 8 points from upsets – which could be just one correct upset pick in the round of 8.

This year, the Prediction Machine used an algorithm which took a target number of upset points and tried to select the most likely set of upsets to meet that total.  Initially I planned to use a target number of 8 points – based on last year’s results – but in the end decided to set the target higher, with the goal of ending up in the top 5% of the ESPN contest if the upsets occurred as predicted.  I placed that goal at (a somewhat arbitrary) 50 points.  I then used the Prediction Machine to predict all the chalk matchups in the tournament.  This identified a number of games where the Prediction Machine thought the lower-seeded team would win:

Home Away
Georgetown Florida
UCLA Minnesota
Kansas St. Wisconsin
Colorado St. Missouri
Memphis St. Mary's
New Mexico Arizona

This adds up to 11 points of mis-seeds.  That’s a surprising number and may reflect an unusual basketball season.  When I plugged these upsets in and ran the tournament again, I discovered that the Prediction Machine also favored #3 Florida over #1 Kansas (an 8 point game), so I added that in for 19 total points of mis-seeds.

The PM then identified the most likely upsets in the remaining games.  These were the top results:

Home Away
Gonzaga Ohio St. 17.5
Notre Dame Iowa St. 17.4
Miami (FL) Marquette 17.3
Louisville Indiana 14.1

The PM then added upsets in order of likelihood until it reached 50 (or in this case, 64).  (The next upset on the list was Oklahoma over San Diego State.)

There are a couple of refinements to this approach that I haven’t had time to incorporate.  A simple refinement would be to drop 14 points of upsets to get back to 50 points.  A more complex refinement would be to try different combinations of upsets to get the most likely combination that reaches the target points.  Either refinement in this year would have ended up keeping just the Louisville-Indiana upset in the final game.

It’s just as well that I didn’t have time to implement either refinement.  This year’s Machine Madness field turned out much larger than expected (27 competitors!) and even if Indiana wins everything, I won’t win the competition unless Marquette beats Miami – one of the upsets that would be dropped to get back to 50.

Looking at the Prediction Machine’s performance, in the first rounds it went 2-2 for mis-seeds/upsets, and in the second round 1-1.  50% correct on picking upsets is probably a pretty good performance.  In the ESPN competition, the Prediction Machine’s bracket is at 94.4% out of about 8 million entries, with 7 out of the Round of Eight still alive.

Tuesday, March 26, 2013

Sweet Sixteen Predictions

Here are the Prediction Machines thoughts on the Sweet Sixteen games:






Miami (FL) Marquette 6 3.7 -2.3
Louisville Oregon 10 11.6 1.6
Ohio State Arizona 3.5 5.4 1.9
Indiana Syracuse 5.5 7.7 2.2
Duke Michigan State 2 4.3 2.3
Kansas Michigan 2 4.9 2.9
Wichita State La Salle 4 7.6 3.6
Florida Florida Gulf Coast 12.5 20.2 7.7

The PM likes mostly home teams, although it thinks Marquette +6 is a good bet.  The PM has Marquette picked as a likely upset in its bracket.  It needs Marquette to win this game and Indiana to win out in order to finish first in the Machine Madness Contest.  (I’ll have a blog post shortly about how the PM picked its bracket.)

At the other end of the spectrum, the PM likes Florida to crush FGCU, even with FGCU's recent victories taken into account.  I'm dubious.  I'm also dubious of the Indiana prediction, given how impressive Syracuse was in San Jose.  And the PM has liked Wichita State all along, and is looking for a fairly routine victory over La Salle.

The PM doesn't usually see this many "bettable" games, where the difference between the Vegas line and the PM’s prediction is greater than 2 points.  It’s likely that – since there are many more regular season games to train upon – the PM doesn't do as good a job accounting for the Tournament conditions as Vegas.  Alternatively, it may be doing a better job, or the lines may be more influenced by betting during the Tournament when there’s more action.

Monday, March 25, 2013

Upset Picks Review

I previously posted the Prediction Machine's top ten upset picks for the first round.  How did it do?  In the table below I've added a column for MOV -- a negative number indicating that the underdog (Away Team) won the game.

Notre DameIowa St.17.4-18
San Diego St.Oklahoma7.115
MemphisSt. Mary's2.52
Oklahoma St.Oregon2.4-13
N.C. StateTemple2.2-4
Colorado St.Missouri1.812
PittsburghWichita St.1.6-18
North CarolinaVillanova1.27

According to the PM, the Notre Dame-Iowa St. game had significantly higher upset chances than any other game, and in fact Iowa State won the game handily.  The PM also liked Oklahoma to upset San Diego State, but State won that game handily.

The next tier of upset possibilities (> 2) was less likely, but the PM also went 50% in this tier.  (Actually 75%.  Although I marked it as a missed upset here, after the play-in game, the PM had St. Mary's as an outright favorite in the game against Memphis, so the Upset probability was for actually for Memphis to win!  The Upset metric remained almost the same, by the way.)

The next tier is below the cutoff for consideration as an upset, although both the Wichita State and Cal upsets were identified here.  (And Cincinatti, Missouri and Bucknell were all popular upset picks by pundits.)

Combined with last year's results, it appears that the PM's algorithm for detecting likely upsets works fairly well.  (Note that whether the PM should include the upset in its bracket is a different question!)

Saturday, March 23, 2013

Machine Madness: Some Final Four Analysis

Games don't start here in San Jose until 4 pm, so I've started to take a look at the Machine Madness entries.  Here are some observations about the Final Four predictions:

Consensus Champion

To the extent that there is consensus, the predictors seem to think the champion is going to come out of the Midwest bracket -- both Duke and Louisville get five "votes" as champion.  The next most popular picks are Florida (4), Indiana (3) and OSU (2).  The predictors are particularly dismissive of Kansas and Gonzaga as #1 seeds -- they both get only one vote for champion.  Least Likely champion prediction goes to LA's Machine Mad pick:  Notre Dame.  That ain't happening :-).

Final Game

Looking at who the predictors think will be in the final game, Louisville dominates with 10 votes, followed by Florida (7), Kansas (6), Indiana (6), Duke (5), Gonzaga (3) and OSU (2).  Florida gets a lot more love from the predictors than from the Committee; clearly the predictors think Florida is closer to a 1 seed than a 3.  Conversely, the predictors don't think much of #2 seed Miami, which only got one vote to get to the final game.

Final Four

As mentioned, Louisville and Duke dominate the machine picks for the Midwest Region.  In the West, Gonzaga gets 11 votes, followed by tOSU with 6 votes.  Interestingly, New Mexico got 3 votes to win the West -- that obviously won't happen, but I'm curious to see what those predictors have in common that led to that conclusion.  In the South, Florida gets 13 votes and Kansas 7.  In the East, Indiana is the overwhelming favorite with 17 votes -- interesting because only 3 of the predictors see Indiana winning everything.

Craziest Final Four

We have to combine predictions across different entries, but: Missouri, Notre Dame, UNC, Davidson.  Seems unlikely :-)

Friday, March 22, 2013

Machine Madness Brackets

If you're interested in following the Machine Madness Contest, the brackets are available here:
Machine Madness
My predictor is currently in dead last, but I'm hopeful that will change!

Here in San Jose we got a mediocre slate of games.  The best game was Cal-UNLV, which was tight late and featured the correct strategy of fouling when up 3.  Cal benefited from a partisan crowd -- if I were UNLV I'd be pretty unhappy of having to play in Cal's backyard!

Thursday, March 21, 2013

Tournament Picks

I'm currently in San Jose to watch some of the first round live, but our hotel doesn't seem to have working Internet at the moment, so updates will be sporadic at best.

The Prediction Machine went 3-4 on the First Four games, just missing the Boise State win.  That was a matchup between identical seeds, so not too surprising to get one of those wrong.

Here are the Prediction Machine's full picks for the Tournament (click to enbiggify):

Keep in mind that these don't represent the Prediction Machine's most likely outcomes; it picks a number of "upsets" as part of its Tournament strategy.  There 12 upsets in this bracket -- a couple are games where the PM legitimately thinks the lower-seeded team will win, but the rest are intentional underdog picks.  In particular, the PM has Louisville as a slight favor to beat Indiana in the Championship Game, but picked Indiana because that is worth 32 points if it happens.

Here are the most likely first round upsets.  The PM calculates this based upon how close it predicts the game to be, as well as how likely the two teams are to be involved in an upset:

Notre DameIowa St.17.4
San Diego St.Oklahoma7.1
MemphisSt. Mary's2.5
Oklahoma St.Oregon2.4
N.C. StateTemple2.2
Colorado St.Missouri1.8
PittsburghWichita St.1.6
North CarolinaVillanova1.2

The "Upset" number doesn't correspond directly to an upset probability, but you can use it to judge the relatively likelihood of the upset.  As you can see, the PM thinks that Iowa St. over Notre Dame and Oklahoma over San Diego St. are by far the most likely upsets. 

Danny Tarlow over at Number Crunching Life will probably be covering the Machine Madness contest in some detail.  There are a surprising number of competitors -- 17 at last count -- so it should be a fun year!

Monday, March 18, 2013

"First Four" Predictions

Here are the Prediction Machine's predictions for the "First Four" games, in order of certainty:
St. Mary's over Middle Tenn. St.
N.C. A&T over Liberty
Boise St. over La Salle
James Madison over Long Island
The last game is basically a coin flip.

St. Mary's is by far the class of this grouping.  If we ignore their early-season neutral court losses, they've only lost to #1 Gonzaga (three times) and at University of Northern Iowa in a game that was postponed for a day due to weather.  Gonzaga beat them consistently by 10+ points, so I don't think they'll be this year's Cinderella.  Still, if they beat Middle Tennessee State they'll probably give Memphis all they can handle.

Sunday, March 17, 2013

Award for Most Disrespected

What team does the Prediction Machine think was most disrespected by the Selection Committee by being left out of the Tournament?  The answer will surprise most people: Iowa.

Iowa finished the Big Ten season at 9-9, ahead of Illinois and Minnesota, both of whom were invited.  Iowa never managed a single win against a ranked opponent but -- with the exception of a 30 point blowout to Michigan back in January -- were in every game.   They were clearly the equal of teams like Illinois and Minnesota.  In the Big Ten Tournament quarterfinals, they lost by 3 points to a very good MSU team.  But apparently the Committee drew the line at 8 invites for the Big Ten.

The Prediction Machine would have also extended invitations to Baylor and Virginia.

Saturday, March 16, 2013

Saturday Prediction (3/16)

I'll do a final Top Twenty posting after tomorrow's games, but here are predictions for today's match-ups.  The Prediction Machine is a little less accurate on these neutral court games.
#3 Indiana by 7 over #22 Wisconsin
#4 Louisville by 7 over #19 Syracuse
#7 Kansas by 7 over #11 Kansas St
I'm sensing a trend.
#10 Ohio State by 3 over #8 Michigan State
#9 Miami (FL) by 4 over N.C. State
I'm not sure why Ohio St is the home team in their matchup.  (There's actually a "home team advantage" in neutral court and Tournament games, although it is less than in real home courts.)

The close matchup of the day is UCLA vs. Oregon, which the PM expects UCLA to win by less than a point.  Unfortunately for UCLA fans (of which I am one), freshman Jordan Adams broke his foot on the last play (!) of yesterday's game against Arizona.  UCLA is now down to 7 players and is likely to lose today and in the first round of the tournament.

In similar news, UNC power forward PJ Hairston suffered a cut on his hand in yesterday's game against FSU and may not be able to play today.  That might open the door for Maryland to get another victory and squeak into the Tournament field.

Monday, March 11, 2013

Machine Madness?

In recent years, Danny Tarlow and Lee-Ming Zen have run the "Machine Madness" contest for computer predictors.  They've gotten a late start this year and I'd forgotten to remind them, but we can probably still pull something together if there is interest.

If you are interested in competing, check out the Machine March Madness Google Group and indicate your interest. 

I've posted an initial data dump for the season here.  It's in the form of a Google Spreadsheet, but you can download it as Excel or a .csv file.  This should be the team-level roll-up of the box score, e.g.,  Naming of the columns is hopefully obvious -- the H prefix means the home team and the A prefix means the away team.  Let me know if you find any problems.

Sunday, March 10, 2013

Saturday Recap

After writing Saturday's predictions, I found a bug in the Prediction Machine. But as it turned out, Saturday was one of those crazy days where a broken predictor might have done better!
Loyola Marymount @ #1 Gonzaga:  Gonzaga by 10
Gonzaga by 18.  Gonzaga seems to be one of the few teams rolling into the tournament.
#3 Duke @ UNC: UNC by 3
Duke may be another.  They won by 16 in a game where UNC was out of it from the opening tip-off.
#4 Kansas @ Baylor:  Kansas by 1
I warned of a potential upset in this game, but I didn't think Baylor would win by 23 (!?)
#17 Syracuse @ #5 Georgetown: Georgetown by 1
Another potential upset on Georgetown's home floor, but Syracuse doesn't show up -- scoring only 39 (!?!) points and losing by 22.  This victory alone might get Georgetown a #1 seed.
Clemson @ #6 Miami (FL): Miami by 6
Miami wins by 13 to continue their up-and-down season.
#24 Notre Dame @ #8 Louisville:  Louisville by 8
Louisville wins by 16.  Notre Dame is really struggling, with 6 losses in the last two months.
#9 Kansas St @ #13 Oklahoma St: Oklahoma St by 4
Oklahoma State wins by 6.

Not a great week for the Prediction Machine forecasting the Top Ten games, but then, it wasn't the only predictor being driven crazy by this season's unpredictability.

Saturday, March 9, 2013

Top Twenty (3/9)

10-2Ohio St.31.6
12Michigan St.31
14NRMiami (FL)30.7
16-5Oklahoma St.30.7

You'll note the rating scale and other significant changes.  I was getting bored with the old Top Twenty, so I've adjusted the Prediction Machine to use a different rating algorithm, to give a bit of a different look at the teams.

Regardless of the rating algorithm, Florida and Indiana remain the class of the field, although Louisville closes the gap with the new rating.   Louisville has fallen in and out of favor with the pundits, but hasn't had a bad loss since 1/22 against Villanova.  Duke and Pittsburgh are other teams moving up, and the big losers here are Wisconsin (losers of two straight, including Purdue), Oklahoma State (lost to Iowa State), Minnesota (lost to Nebraska!?) and Georgetown (lost to Villanova).  Georgetown's hopes for a #1 seed rest on winning today at Syracuse and then winning the Big East tournament.  The lesson here, if there is one, is don't lose late-season games to weak opponents.

Several surprise teams appear in the bottom of the list, including Missouri, Creighton and Arizona.  Despite 6 losses, Missouri is a surprise #2 in the SEC and has looked strong recently, including a win over Florida.  Barring an unlikely loss to Indiana State, Creighton looks set to win the MVC.  Arizona, meanwhile, lost both Los Angeles games and is an increasingly distant 4th in the Pac-12, so this rating is a bit of a mystery.  This may be a case where the Prediction Machine is being influenced by strong halftime results.  It will be interesting to see how Arizona performs in the Pac-12 tournament.

Saturday Top Ten Predictions

Loyola Marymount @ #1 Gonzaga:  Gonzaga by 10
#3 Duke @ UNC: UNC by 3
#4 Kansas @ Baylor:  Kansas by 1
#17 Syracuse @ #5 Georgetown: Georgetown by 1
Clemson @ #6 Miami (FL): Miami by 6
#24 Notre Dame @ #8 Louisville:  Louisville by 8
#9 Kansas St @ #13 Oklahoma St: Oklahoma St by 4

Today's games feature a couple of mismatches between the AP's rankings and the PM's rankings, I'll be interested to see how they turn out!

Wednesday, March 6, 2013

The Bottom Ten

I'm home for a snow day courtesy of the "Snowquester" so for fun I thought I'd take a look at the ten worst teams in NCAA Div I:

10S. Carolina St.-36
9Fair. Dickinson-36
6Miss. Valley St.-41
5New Orleans-45
4Alabama A&M-46
3MD Eastern Shore-50
1Grambling St.-134

What's remarkable here is: (1) How terrible Grambling State is, and (2) How terrible the SWAC is.

Grambling State is 0-27 and has failed to break 40 points twice this season.  Perhaps more amazingly, they've lost to #7 MVSU by 22 and to #5 Alabama A&M by 25.  They lost a pre-season game to Auburn -- who is dead last in the SEC -- by 50 points.  No other team in the nation has won less than three games.  So this is an awesome level of ineptitude.

If Grambling State played at Florida, the Gators would be favored to win by almost 70 points. 

And when your conference has 3 of the ten worst teams in the nation, your conference ain't very good either.  The conference as a whole is 1-45 against teams in the top 100.   The best team in the conference is Texas Southern, and they would be dead last in most of the power conferences. 

It makes you wonder whether the NCAA needs a relegation system.

Saturday, March 2, 2013

Saturday Review

Iowa at #1 Indiana:  Indiana by 14
Portland at #2 Gonzaga: Gonzaga by 26
Rutgers at #7 Georgetown: Georgetown by 12
Alabama at #8 Florida: Florida by 15
Indiana by 13,  Gonzaga by 29, Georgetown by 13, Florida by 12.  Four out of seven predictions within one basket.  The PM is often eerily accurate.
#5 Miami at #3 Duke: Duke by 9
Duke by 3, thanks to a massive performance by newly-returned Brian Kelley.   Given that this was at Cameron, it's clear that on a neutral court Miami is better than Duke.  The question is why Miami dropped a game to Wake and struggled against Clemson, and how they'll respond in the Tournament.
WVU at #6 Kansas:  Kansas by 15
Kansas wins by 26 in a game that turned into a rout.
More interesting will be #10 Louisville @ #12 Syracuse, where Louisville will have a good chance (~40%) of pulling the upset. 
And so they do, winning by 5.
Other good games outside of the Top Ten include Arizona at UCLA and Clemson at Virginia Tech, both of which are coin flips.
UCLA had a comfortable lead for much of the second half against an Arizona team that wasn't playing very smart basketball but still had some chances at the end.  Clemson outplayed Tech in the second half and got it close but couldn't get the win.

Friday, March 1, 2013

Top Twenty (3/1)

1 +1Indiana131
2 -1Florida127
3 +1Gonzaga111
4 +1Kansas110
5 +2Wisconsin110
6 -3Michigan110
7 -1Louisville109
8 +1Ohio St.105
9 -1Syracuse103
11 +1Oklahoma St.96.6
12 -1Michigan St.96.2
15 +3Minnesota91.4
16 -1Saint Louis89.8
18 -2Kansas St.88.9
19 NRVirginia87.6
20 NRIowa St.87.6


It was a week of "upsets" in college basketball -- or was it?  The PM correctly predicted likely upsets of Indiana at Minnesota and Duke at Virginia.  The PM also predicted that Michigan at Penn State was likely to be a lot closer than most people expected -- although even the PM didn't think PSU could win!

People focus too much on outcomes and have a hard time objectively assessing team strength.  Going into the Indiana game, Minnesota had lost 8 out of 11 and was being described as a "collapse."  But take a look at those losses (showing the PM rankings of the opponents): 

1/12: @ #2 Indiana (-7)
1/17: vs. #3 Michigan (-8)
1/23: @ NR Northwestern (-7)
1/26: @ #7 Wisconsin (-1)
2/6: @ #11 MSU (-11)
2/10: vs. NR Illinois (-4)
2/17: @ #14 Iowa (-21)
2/20: @ #9 tOSU (-26)

The truth is, when you play teams that are better than you on the road, you're going to have a bad time :-)  To put this stretch in perspective, Duke/Kansas/Syracuse would all be expected to lose 6 of those games.

The only "bad" losses here are to Northwestern (but that was on the road) and to Illinois at home (but that was only by 4 points).  They did lose by 20+ to both Iowa and tOSU, but those teams are both better than Minnesota and were playing at home.  And in the same period they won at home against #14 Iowa and #7 Wisconsin.

Minnesota lost to Indiana by 7 at Indiana.  If we take a rough figure of 3.5 points for the Home Court Advantage, then we'd expect Indiana at Minnesota to be about a toss-up.  (The PM actually had Indiana as about a 2.5 point favorite based on other factors.)  So it really shouldn't be a surprise that Minnesota beat Indiana, and that probably shouldn't affect your assessment of either team that much.

The win over Duke (another not entirely unexpected result) was enough to pull Virginia up into the Top Twenty.  They should be a lock for the Tournament at this point, although they might need a few more solid wins to convince the Committee.

Saturday Picks

Iowa at #1 Indiana:  Indiana by 14
Portland at #2 Gonzaga: Gonzaga by 26
#5 Miami at #3 Duke: Duke by 9
WVU at #6 Kansas:  Kansas by 15
Rutgers at #7 Georgetown: Georgetown by 12
Alabama at #8 Florida: Florida by 15
#10 Louisville at #12 Syracuse:  Syracuse by 2

Most of the Top Ten teams are playing comfortable home games against mediocre opposition.  #5 Miami @ #3 Duke is being touted as the marquee matchup of the day, but it's likely to be a fairly routine Duke victory.  More interesting will be #10 Louisville @ #12 Syracuse, where Louisville will have a good chance (~40%) of pulling the upset.  Other good games outside of the Top Ten include Arizona at UCLA and Clemson at Virginia Tech, both of which are coin flips.