Showing posts with label acc season sim. Show all posts
Showing posts with label acc season sim. Show all posts

Monday, January 12, 2015

pack-line rising

The week 9 AP rankings didn't change much from the week before.  The top 5 was static; you have to go to 13th before you find anyone who'd moved so much as three spaces, and 17th before you found anyone moving more than that (Iowa State dropped 8 spots for losing to South Carolina.)

That was last week.  This week is week 10.  And because there's nothing quite so humbling as the midwinter conference grind, and because the grind apparently got an early start this year, not a single team in the rankings is where they were last week.  Except Kentucky, and not even them exactly, because as the ESPN headline put it, "Virginia lands two of Kentucky's No.1 votes."  (And because the word "lands" is most often used in recruiting, my heart raced happily for just a quick split second until I finished reading it.)

Well, I'm happy anyway.  Somewhere out there are two writers who think UVA is the best team in the country.  Actually, I'll tell you exactly who: John Feinstein in DC and Zach Osterman of the Indy Star.  (And Marcus Jackson of the News-Gazette in Champaign, IL had the nerve to keep the Hoos in third behind UK and Gonzaga.  Boo this man.)  Osterman explained his vote in a column, which basically sums up as he weights recent results more heavily than most, and UK has been all about the overtime lately.  When the Blogpoll was a thing and I had a vote in it, recency mattered not even a speck (and it still wouldn't with me, resulting in a vote for UK), but it's also at least a little bit compelling to say, look, this is the ranking for this week, and this week, this is the team playing the very best basketball.

Further UVA awesominity: The Hoos are eight thousandths of a point out of first in the KenPom ranks, a gap small enough that it could disappear just by one of Kentucky's opponents having a bad game.  ESPN's Jeff Goodman asked various coaches who they like between UK, UVA, and Duke, and one asshole anonymous coach responded, "Kentucky over Virginia?  Please.  Talent level isn't even close.  Zero Virginia guys would start for Kentucky."  KenPom would disagree there, too; of his top ten players in the country, three are UVA's.  And three are Kentucky's.  Wait, no, recount: it's just two.  Hold on a sec, it's actually just one.  Oops, wrong again: the truth as of right now is that Kentucky has zero of the nation's KenPom-top-ten players.  UVA's so unathletic.

So the ascension continues.  How did this happen?  Exact same way the Hoos earned a pair of 1 seeds last year.  Choke yourself if you said "unbalanced schedule."  Taking care of business when other teams do not.  Top ten teams kept running into bug zappers this weekend.  Duke, Arizona, Wisconsin, Louisville - it was a top-ten bloodbath.  Kentucky was thisclose to being on that list.  UVA - well, it was close, too, but not, like overtime-close, and it was on the road, and the opponent was ranked, and so on and etc.  Give another round of applause to Mike Curtis, I'd say, but save the biggest cheer for Tony Bennett and his "play every game the same" mantra, because that's what you're seeing play out.

***************************************************

-- In a strange way, I think I'm going to miss Karl Hess, exiled from ACC games as a result of saying something kinda dumb to a fan, but also, supposedly, "accumulated incidents."  I'd brush that last one aside as the ACC's attempt to leak something that put them less on the spot for the decision, if it were any ref but King Karl.  Hess made all his assignments a game of referee Russian roulette, but sports are less interesting with fewer villains to kick around.

-- I'm sure a lot of folks saw the study about referee bias that concluded ACC refs - football ones, not basketball - are biased toward home teams and "long-time conference members such as Duke and North Carolina."  I wonder if UVA is a "long-time conference member" in that context, or if you have to reside in the state of North Carolina.  At any rate, they picked the wrong sport.  I'm sure everyone loves to hear that the refs love Duke, but might raise an eyebrow about finding that it's a football study.  I'd like to see one for basketball - not that they haven't happened, as exampled by this one from a few years ago concluding that hoops refs do indeed favor home teams and call make-up fouls - but I don't know that anyone's really studied ACC hoops long-term.

-- Sort of a boilerplate article on UVA's 2016 commitment Kyle Guy, but one thing stuck out: "Interestingly, the 6-foot-2 Guy said he has been deluged with letters from Virginia Tech and Notre Dame since choosing Virginia."  Notre Dame, I kind of understand; Guy is from Indiana.  But I find it adorable that Buzz is of the opinion that Guy can be convinced to make that switch.

-- This week has been something of a microcosm of the football team's struggles.  Since 2015 began, UVA has lost three defensive ends: Max Valles to the NFL (ok, he's technically a linebacker, but still), and former commits Brandon Wilson and Rasool Clemons - the former to Indiana and the latter to the fact that he and academic qualification are a long way from becoming acquainted.  How have they been replaced?  With wide receivers!  Warren Craft is the latest commitment, yet another result of Mike London's incessant digging for unmolded instate athletes, and UVA also announced the incoming transfer of T.J. Thorpe from UNC (like, ten minutes after I hit "publish" last week.)  Actually, Thorpe, if healthy, does have the potential to add a bit of a different dimension to the offense, for one year at any rate.  Regardless, next year UVA will have eight wide receivers with some kind of on-field experience, and three such defensive ends.  Or, put another way, if London succeeds in landing Gary Jennings (who he's pursuing with all expedience and urgency) he'll have as many scholarship WRs on the roster as scholarship D-linemen.

-- Per Coaching Search, UVA is bringing back Dave Borbely to coach the offensive line.  Borbely was Al Groh's last O-line coach before his removal, and memory is hazy but I don't recall being unhappy with the job he was doing.  Borbely put a few linemen in the NFL, to be sure.  Maybe, if London pulls off the comeback of the ages and sticks around long term, Borbely can remind him that he used to have a few more players under his purview than he has now.

***************************************************

I told you KenPom loves him some UVA, and the proof is in the season simulations:


That's UVA with a 91.8% chance to snag the top seed.  KenPom gives the Hoos a 95% chance or better in seven of the remaining fifteen games, and the toughest remaining game from this vantage point is the last one: the road match against Louisville, where UVA has a 60% chance of pulling out a win.  UVA also has an awfully significant chance of going undefeated: 12.3%.

That's what the math says, anyway.  Keep in mind the arbitrariness of the tiebreaker: UVA wins all ties against everyone for having the highest KenPom rating, because I don't have a way to make a spreadsheet remember who beat whom.  In the ten thousand different seasons that go into this, I doubt that 9,178 of them actually featured UVA beating Duke and earning that tiebreaker.  It's probably closer to 8,000, since KenPom is giving Duke a probably-too-low 20% chance of winning at the JPJ.  So take this with many grains of salt.  Even so, even if the simulator is more than 40 percentage points off and it's a coinflip as to whether UVA earns the #1 seed or not, that's still what you'd call "heavily favored."

You can find all three of the sims so far piled up at the original post.

Monday, December 15, 2014

twenty-one

National champ-peen-ship, y'all.  I'm long since on the record as admitting I don't get around to watching nearly as much UVA soccer as I'd like.  (I've considered remedying this next year by ignoring football, but that'd be all talk and no action on that front.)  In fact this season I've seen exactly two soccer games: the women's semifinal against Texas A&M and the men's national title against UCLA.  I had to miss the men's semis and women's finals for reasons of State.

It didn't take me long to decide I hated the announcers.  I wasn't even paying attention to them in the A&M game so I have no idea, but it was only about 20 minutes into the UCLA game before I decided the announcing was a zero on a scale of Pam Ward to Keith Jackson.  Their contempt for UVA's game plan was plain as day and got plainer as the first half wore on.  Although I did enjoy their description of the UVA strategy as "cynical."

Guess which strategy UVA went with?

You can't totally fault them, if the idea is that games should always be exciting shootouts.  But UVA came into the game 111th in the country in scoring average.  One-hundred and eleventh, it's not an extra-1 typo.  In only one tournament game did the Hoos score more than once.  UCLA came in with only one tournament game of fewer than three goals.  So it should be no surprise that Gelnovatch decided not to run 'n' gun with the Bruins.  "The beautiful game" it was not, but Gelnovatch isn't paid to entertain the pundits.

And frankly, it worked to almost complete perfection.  Actual perfection would've been a 1-0 win with the one goal probably coming off of some kind of set piece.  You don't counterattack to try and score, because that would expose you; you counterattack with the aim of getting a corner kick.  If it doesn't go in, which it usually doesn't but it's more than worth a try, bunker back down and try again.  I laughed when halftime rolled around and the announcers said UCLA had to be very pleased with the first half and then both coaches said the first 45 went completely UVA's way.  I might've considered it a Bruin domination, if UCLA had generated more than the occasional chance, but UVA's keeper Calle Brown was barely tested.  Owning the possession battle 45 yards out is one thing; it's another thing entirely to dictate the game.  UVA did so without having the ball.

The second half - a little more pressure, the Bruins clearly took their coaching to heart and were less patient, more attacking, but also a great deal more frustrated.  UCLA's Edgar Contreras ought to have been red-carded for a head-butt, but I can't completely fault the refs as the camera was right on it in real time and I still missed it until the replay.  But it was a clear sign that UCLA was used to being able to break down a defense, and UVA's brick wall was getting to them.

The game is likely to attract precisely zero new fans to the game of soccer, but just look at all the bothers I give.  You know I love me some pack-line defense, and so, apparently, does George Gelnovatch.  Tony Bennett wins basketball games 45-26, but he wins basketball games.  At the end of the day, here's the stat that matters most: 21.  And the one that matters second-most is 0.

*******************************************

The first major bit of football attrition hit last week when Eli Harold declared for the draft.  That makes two; David Watford also decided to transfer, but the effect of that will be almost nil.  Except to eliminate bizarre message board posts wondering why he's on the field in any capacity at all, as if not being a good quarterback is the same as not being a good receiver.

UVA dodged a bullet when Max Valles announced that he'd be returning next year, quashing rumors to the contrary.  Smart - Valles would've been going almost entirely on physical attributes.  A year of opening some scouts' eyes would help him.  Harold, though he could benefit from another year, is probably in good shape anyway.  He showed this year that he can defend the run and isn't just a one-dimensional pass rusher.  Once he gets in front of scouts at the combine, he should make an appearance on draft boards and could easily be a second or third round pick; his ceiling, if the workouts look good, would be the low first.

As for our defense, it's a fairly major hit, but there's a long-run silver lining: next year, Harold and the Moores (Michael and Kwontie) would once again have dominated the playing time.  Great, because they'll do well, but no experience for the boatload of guys behind them.  Trent Corney should start off as the third DE, but there's a trio of redshirt freshmen who will get a chance to make a wave or two as well.

*******************************************

And the second major bit of football attrition is on the coaching staff, as Scott Wachenheim is off to VMI to play head coach.  Good for him - it's his first head-coaching gig, though he did have the title of OC and assistant HC at Liberty for a few years.  There can't be a tougher place in the world to win at football than VMI, except maybe the Citadel.  From his perspective, this is definitely striking while the iron is hot - the chances that most of this staff is out of a job next year are awfully high, and you might as well grab a promotion while it's there.  Even if London was on rock-solid ground, I think he'd go anyway, but still.

Wachenheim leaves with one of the most mixed legacies I've ever seen for an assistant coach.  He was vilified at times for the play of the O-line, and I think at least partially deservedly so.  But he leaves on a positive note, having gained a lot of credit for making the O-line not be a total black hole of suck despite being held together with Scotch tape and having to use 260-pound converted DE (or TE or whatever) Jack English as a left tackle.  And I think also deservedly so.

It leaves UVA with two openings to fill, including the impending (or already-occurred) retirement of Tom O'Brien, whose UVA career was basically a dud.  There's an inexplicable level of support for Ron Mattes, who was here like, a year, and performed no miracles.  He'll be here as soon as Bill Musgrave comes back, I'm sure.  A much more likely name, and these tea leaves sure read awfully clearly, thanks to Streaking the Lawn's Tweety account, is current Edmonton Eskimos O-line coach Jonathan Himebauch.

*******************************************

For lack of anything to do in this basketball wilderness, I put together another season sim, since it had been a couple weeks and stuff happened.  You can find it below and on the original season sim post, for easier comparison to the previous version.


NC State and Notre Dame are on the rise; both won an early-season ACC game against Wake and FSU, respectively, and ND has been handling a lot of business as well.  NC State, not so much, but then, Wofford is actually awfully highly-ranked for a SoCon team.

Being as UVA has also been handling business, the Hoos leapfrogged Louisville, which itself didn't exactly fare badly, just not as well as UVA.  Maryland and VCU are both higher-ranked than all but five ACC teams, and UVA crushed both on the road.  It's becoming clear that there's a top three in this league, and as such, the race for the top seed has obvious huge implications.

Monday, November 24, 2014

acc season sim

I'm going out of town for Thanksgiving, which means a pretty long holiday hiatus, which means if I'm gonna finish a preview of the VT game it's gotta be done tonight, which means if any other content is getting posted tonight it's gotta be short and sweet.  So I figured I'd whip up an old favorite a little early: the ACC season predictor.

This is one of my better spreadsheets, if I do say so myself: I can punch in the KenPom-given probabilities for each game and simulate 10,000 ACC seasons in seconds flat.  The only thing it can't do is the proper tiebreakers; instead, I just tiebreak using the KenPom rankings.  But otherwise it's pretty neato, it spits out each team's chance of landing in a particular seed in the ACC tournament.  Presented without a hell of a lot of comment is this season's first go-round:


I think the main conclusions for now are that there are three main contenders (Duke, Louisville, UVA) and three outside contenders, of which surprisingly Miami is one and Pitt is not.  And that the race for 9th could get very interesting.  9th is a good place to be because you get a bye straight to the 8th team; 10th is not bad either because playing VT is sort of like a bye too.  Drop to 11th, though, and not only does the quality of your opening-round opponent go way up, so does the quality of your next-round opponent.  Naturally, the end result in March won't look much like this, but it's fun to know where things stand anyway.

So, just as a programming note, don't expect anything between now and next week, except for one thing: the VT game preview, which I'll automate to run on Wednesday morning.  Have a very excellent Thanksgiving, and hopefully, an even better Friday night.  If things go well I get to come back and excitedly overanalyze the shit out of East Carolina or Illinois.  If things go even better I'll even get to watch from a Ford Field seat.


Here's the third iteration of the sim: