Monday, March 9, 2015

Time for Torvik's annual beatdown

It's March, so that means it's time for me to destroy Torvik in the Big Ten tournament bracket challenge. I'll post my picks first and we can laugh at his later.

PSU over Nebraska
Minn over Rutgers

IL over MI
Iowa over PSU
NW over IU
OSU over Minn

UW over IL
Iowa over Purdue
Maryland over NW
OSU over MSU

UW over Iowa
Maryland over OSU

UW over Maryland

Sunday, March 8, 2015

All Big Ten

 I am trying something new with the All Big Ten team. All Big Ten by class. Here's how I think it should look:

Kaminsky, Aaron White, Brandon Dawson, Dez Wells, DJ Newbill
I felt these choices were pretty obvious. There is not a ton of depth beyond the top 5 seniors, which bodes well for a fun Big Ten next year with lots of talent coming back. Rice missed too many games, and Trice was too inconsistent.

AJ Hammons, Dekker, Jake Layman, Yogi, Petteway
Hammons had a comeback player of the year type season if that was an actual award in college. Dekker and Layman were very good on great teams and could have put up huge numbers on a worse team or if they were more selfish. Yogi is Yogi. Petteway just had no help this year.

Hayes, Malcolm Hill, Troy Williams, Koenig, Zach Irvin
Hill and Koenig got huge mid season opportunities when Rice and Jackson got hurt and both took full advantage. The last spot cam down to Nunn and Irvin. I gave it to Irvin because after Walton and Lavert went down he stepped up and helped MI get to 8 wins. This class also doesn't have a ton of depth but I expect to see a bunch of the 2nd tier guys step up next year as juniors.

Melo Trimble, D'Angelo Russell, James Blackmon, Bryant McIntosh, Jae'Sean Tate
The top 3 are easy. The rest were tough as the freshman class is loaded, especially at guard. The Big Ten will have fun players to watch in the coming years. Tate reminds me so much of a young Alando Tucker. I have little doubt he will be great and was super fun to watch this year. He took away a starting spot from a bunch of more experienced players on a good team.

Evaluating the predictions

I feel OK about the predictions this year. It wasn't a great year, but not awful like last year. This was a young Big Ten with a lot of new faces which made it harder to predict than normal, plus there was the addition of 2 new teams (I missed on both). If I got it right or within 1 that's a win in my book. Within 2 is a push. Everything worse that that is obviously bad. Of the misses there weren't any glaring errors like my 8 game miss with Nebraska last year. I was no more than 4 games off with any team. Despite adding 2 teams, I had the same number of misses (5) as last year. I'll only address the misses.

The wins this year are UW, MSU, MI, IL, Indiana, NW
The pushed are PSU, Minnesota, OSU
Misses are Rutgers, Purdue, Nebraska, Iowa, Maryland

Rutgers- One of the two new teams I missed on. I actually had them with 3 wins on my first draft but the total wins and losses didn't add up right. I added a couple wins for them thinking they might pick off some of the mediocre middle at home, but was wrong. I blasted this team in my preseason A-rank and predicted they would be the worst team in the big ten, so I don't feel too bad about the 3 game miss.

Maryland- The other new team I missed on. This team was tough as I didn't know much about them coming in, plus they had 4 freshman in the rotation, plus Wells and Smotryz missed a bunch of games early with injuries. I underrated this team and they kept getting better as the season went along.

Purdue- Speaking of teams with lots of freshman and getting better as the season went along. This is the 2nd year in a row I have missed on Purdue. I feel like I didn't trust my eyes enough on this team. It was obvious this team was vastly improved over last year but I was too chicken to pick more wins. I don't think many saw a 12-6 season, but I should have been better than 4 games off.

Nebraska- This is 2 years in a row missing on this team too. I blasted this team in my A-rank, but again I didn't trust my eyes enough to downgrade them appropriately. They got pretty much the same numbers out of Petteway and Shields but the supporting cast stunk and they averaged 6 fewer points per game as a team. The problem was shooting, as they were a God awful 28.6% from 3 while taking 36% of their FGA from 3.

Iowa- After 3 years of .500 basketball, Iowa finally broke through with a 12-6 season just as I gave up on them. Aaron White was incredible, and Jok got better and better as the year went along to give them something from the perimeter to go with that strong front court.

Thursday, March 5, 2015

How good (or bad) is Wisconsin's defense, really?

If you compare the Kenpom ratings to the T-Rank, one difference that may jump out is that Kenpom ranks Wisconsin's defense 48th in adjusted defensive efficiency, and T-Rank has it ranked 24th. That's a fairly large and significant difference, particularly if you're one of those people who filters out national championship contenders by metrics such as "top 25 in both offensive and defensive efficiency" or some such thing.

So what accounts for this difference? Basically, the difference is caused by a rather extreme feature of Kenpom 2.0: it deeply discounts (sometimes all but ignores) the results of mismatched blowouts. So Wisconsin's opening game demolition of Northern Kentucky -- which Wisconsin held to an incredible .526 points per possession -- likely doesn't much figure into Wisconsin's Kenpom rating.

To see how drastic this Kenpom 2.0 adjustment affects the Badgers, compare their defensive numbers to Villanova. If you look at the raw efficiency numbers, Wisconsin and Villanova have essentially identical numbers:

Team Raw DE Raw DE Rank
Wisconsin 93.5 23
Villanova 93.3 21

This is just "total points allowed divided by total possessions" and by this metric the two defenses are very similar.

But Wisconsin's raw defensive efficiency gets adjusted significantly downward, while Villanova's gets adjusted slightly upwards:

Team Adj DE Adj DE Rank
Wisconsin 94.9 48
Villanova 93.1 20

You might think this adjustment is being made based on strength of schedule -- perhaps Villanova has just played better offensive teams, which would naturally lead to an adjustment in their favor. But according to the Kenpom numbers, that isn't so:

Team Opp. Adj OE Rank
Wisconsin 105.0 51
Villanova 104.9 56

So strength of opponent would favor Wisconsin, if anyone.

Other than strength of opponent and adjustment for mismatches, the other adjustment that could affect the ratings is recency. Kenpom weighs more recent games more heavily than older games (as does T-Rank). But this is a minor factor, and cannot explain the large downward adjustment compared to Villanova. Besides, Wisconsin has been playing better on defense recently.

The T-Rank algorithm discounts the effect of mismatches too, but obviously not as aggressively. I'm quite certain that this accounts for the difference in the adjusted defensive ratings. Many of the Badgers' most impressive defensive performances have come in the kinds of games (such as the Northern Kentucky game) that Kenpom 2.0 discounts or ignores.

There's reason to believe the T-Rank is closer to the truth. One clue to this is that Kenpom had Wisconsin's adjusted DE ranked in the 80s just a few games ago. Meanwhile, T-Rank has consistently had Wisconsin's adjusted DE ranked much higher, in the top 35 all along, and it has therefore more accurately predicted the Badgers' performance in recent games.

One of the fun things about doing the T-Rank this year is that it has given me some insights into the vagaries of the Kenpom ratings. They have amply earned their status as "authoritative" but they are not perfect, and they are by no means the simple application of god-given math. Judgments are involved, and sometimes those judgments -- even if they are correct in the macro sense that they will most often produce better predictions for most teams most of the time -- sometimes produce error.

Thursday, February 19, 2015

Hot Take on the Brandon Knight Trade! AKA Old Fashioneds with Jimmy Boeheim.

Hot takes aren't always the pinnacle of the analysis spectrum, just ask friends and enemies of Archduke Franz Ferdinand, if there are any left. However, theories abound that the gut really does make the best instant decisions.  So, let's fire up that old tankard!
If the rumored trade stays as is (Bucks still need to clear two roster spots((sounds like Kenyon Martin will be waived, and who knows about Sanders)), the Bucks receive Michael Carter-Williams from the Sixers, and Miles Plumlee and Tyler Ennis from the Suns. (It looks like we also sent Kendall Marshall back to the Suns, but he's out for the rest of the year anyhow) Gotta say, the gut says, "I LOVE IT!" Even if this deal was getting Carter- Williams straight up for Knight, I think we come out on top, let alone adding a couple possible rotation guys.  It's actually similar to the deal which brought Knight to Milwaukee.  Instead of paying too much (read: market value for a starting point guard) for Brandon Jennings, who wasn't a perfect fit for whatever the Bucks were trying to be, we shipped him to Detroit for Knight (Middleton, et al.).  Instead of paying Knight too much this offseason or letting him walk, we shipped him out for a couple inexpensive years of Michael Carter-Williams.  If you're a team like the Bucks, and you don't know if a particular guy is part of the path forward, you're no doubt better off not paying him way too much for the next 4 or 5 years (See: Illyasova, Ersan and Sanders, Larry).

I imagine some pundits will argue the Bucks shouldn't be messing with their chemistry at this point, but that will end up being a Kornheiseresque talking point in an empty argument. This year's team, aiming for a 6 seed in the playoffs, is playing way above their depth with a hodgepodge of journeymen and roleplayers. Don't get me wrong, it's exciting to see good defense and Swiss Army Knife basketball, but they aren't going anywhere this year.  I can't think of another team who could have gotten BETTER after losing (for the year) their: "future superstar", a highly paid defense & rebounding big, and a quality backup point guard, not to mention several other multiweek injuries scattered around the rest of the roster.  Chalk that survival/evolution up to an overabundance of solid, veteran, rotation guys... and good coaching, right?  It's gotta be the coaching.  I mean, just look at OJ Mayo this year vs. OJ Mayo last year.

This is no doubt a move for down the road, and it's not going to hurt anything this year.  Let's hope there's not a hidden first rounder mixed somewhere in the fine print.  Back to the point, just who are these guys we're getting in return for Brandon Knight?

Plumlee- Let's be honest, it's hard to keep the Plumlees straight. It seems like one is good every year, and this year it's not Miles. He was starting for Phoenix last year, and putting up an 8 and 8.  But he's riding the pines this year.  I bet he gets some solid run for the Bucks this year, does Plumleeish things, and lets Henson be a little more flexible in the roles he's used in.  He's on the books for about $2 Million next year, the last year of his rookie deal.

Ennis- A rookie point guard from Syracuse, who played one very solid year in college.  He went 18th overall in a very deep 2014 draft, and was the 5th point guard taken.  He hasn't played much for the Suns this year, but that's not saying too much, as they were apparently following the Kahn blueprint of accumulating as many point guards as possible.  Unfortunately for them, it took until this year's trade deadline for Phoenix to realize there's only one ball to go around.  So, they traded 3 point guards (Dragic, Thomas, and Ennis) at the deadline, and added 2 (Knight and Marshall).  Ennis could be a sneaky good piece of this trade, especially under Jason Kidd's tutelage.  I really enjoyed watching him play at Syracuse last year.  He was very heady, was always under control, was clutch in some big moments, rarely turned the ball over, and played with so much poise for a Freshman.  He's a bit slight for the NBA game, and is not a particularly explosive player, but it's easy to see him as a backup point guard going forward.  His potent shooting stroke has translated to the D-League at least, where he's averaging 18 ppg, while hitting 50% of his shots.  He's under team control through 2018.  Keep an eye on him.

Carter-Williams- On the surface, this looks like the Bucks traded a very athletic, shoot first point guard who didn't create much, and got back a true point guard who can't shoot, but plays great defense, and is a much bigger and more complete player.  In this case, both books may really be their covers.  Granted, MCW has played on an historically awful 76ers team over the past few years, chucking it up there far too often, which has probably stunted his development.  And let's not forget that Brandon was an alternate for the All-Star Game this year, although his numbers truly are inflated on this Starless Bucks team.  In fact, that last sentence is what gives me the most hope about this trade.  If Knight gets paid this offseason like his numbers say he ought to, the Bucks could have been on the hook for at least $12 Million a year for 4 years of Knight, or two more years of MCW at $2.3 and then 3 Million.  For a team that's succeeding in large part because of it's Defense, the money saved over the next two years, and the resulting roster flexibility makes this look like a steal.

It's time for some fully Subjective Hot Take Fandom- OR, continuing to attempt to translate Collegiate success to the NBA level (See: Tyler Ennis paragraph above).  But really, let's be honest, who's watched much, if any of the Sixers three year shitshow?  Gotta speak to what you know- and I know I thoroughly enjoyed MCW's game at the college level, and that guy is closer to the guy we'll see in Milwaukee than the guy who's chucking it in Philly.  Even though he couldn't shoot a lick (still can't), he was so long (Bilas?  Bilas?), so solid defensively, such a good rebounder, and so darned competitive that the poor shooting could be overlooked.  He's the type of player who really did seem to make other guys better.  In fact, at the time of the wretched 2013 draft, I thought he should have been the #1 overall pick.  He didn't make me look like a complete idiot when he ran away with the Rookie of the Year Award.   He has been pretty good this year too, even if his numbers are a bit inflated.  Here's a stat that can't be all air- Carter-Williams and Russell Westbrook are the only NBA players averaging at least 15 points, 6 rebounds, and 7 assists.

Numbers aside, it's pretty exciting that we'll get to see MCW in a Bucks uniform for the next few years.  The ungodly positional wingspan between Giannis and Carter-Williams alone should be fascinating to watch, especially on defense.  And mark my words... kinda... I think Ennis can be a quality scoring point on the second unit.  Of course, I thought Jordan Taylor could do that in the NBA too, and he had an NBA body.  That's why we write on the internets instead of becoming Assistants to the Assistant General Manager of the Sacramento Kings.  Maybe I just have a thing for Syracuse Point Guards.  Maybe it's because Jimmy Boeheim is quite the drinking buddy, and has been deluding my thinking with his constant stream of libations and his fervent narration of days of yore.
Maybe Boeheim has been pulling a Madden and "driving" to away games, only to sip Old Fashioneds with the Bucks brass, whispering sweet nothings about how good his former Orangemen would look in a Bucks uniform.  For once, it seems Boeheim, the Bucks front office, and yours truly seem to be on the same page.  Maybe we can agree a tall, (6'6) athletic, defensive minded point guard is still an advantage in the modern NBA.  Maybe we all remember Jason Kidd was a point guard who used his size to his advantage, and specialized in making his teammates better.  Ask yourself this- did it ever seem like Kidd ever really was sold on Brandon Knight being "his guy"?  Maybe it's just an irritable bowel type of gut reaction, but I really do think Carter-Williams (and quite possibly Ennis) is going to flourish in Milwaukee.  Maybe Boeheim will put down the beer and the gun and realize his true calling is as a Bucks' Assistant Coach.  Maybe, just maybe this trade is just what the Bucks needed- to add another star, and become relevant in the next few years.  Maybe Scott Walker wasn't just full of hot shit when he told Wisconsin just how much the Bucks mean to this great state of ours.  What?  A bridge too far?  Stage Left?  Oh... Stage Right then.


Saturday, February 14, 2015

The Eye Test

A sportswriter for the Chicago Sun-Times recently insulted the Badgers on Twitter:
This is a strange thing to say about the team with most efficient offense in the country this year —in fact, the most efficient offense since at least 2002. Some inquired whether he'd actually watched the Badgers play:
This is quite strange. The Badgers played Duke this year. They played Georgetown. They played Oklahoma. They played Michigan, Iowa, and Indiana. (By the way, they scored in bunches in every one of those games.)

But this guy somehow managed to catch the Rutgers game and one other game? Suspicious.

And, of course, the Badgers played that game without the frontrunner for player of the fucking year, Frank Kaminsky. But Mr. Potash probably didn't know that, because he didn't know better than to tune in for Wisconsin versus Rutgers in the first place.
This a half-hearted recantation. He admits he doesn't watch college basketball, but somehow knows it ain't what it used to be. This is really doubling down on his original sin: first he ignorantly impugned the Badgers; when called on it, he put down the whole sport.

To his credit, he at least claimed later to have recanted the entire episode:
This is incorrect. He actually doubled down on the tweet by insinuating that the entire sport fails the eye test. But, to his credit, he did at least admit that he at no point had any idea what he was talking about.

Ultimately, this is the problem with the "eye test." No one watches enough college hoops to have an informed subjective opinion about every team. It's really the "I test." Usually it's a random assortment of stupid opinions based on five minutes of a game that a guy happened to watch at a bar. If you've ever been to a bar, you know that guys at bars have very strong opinions about whatever it is they happen to be observing at that moment. And there's no way that "facts" -- like points scored per possession minus points allowed per possession, adjusted for opponent and venue -- could possibly affect those ignorant opinions.

But these ignorant opinions rule the day, it seems. The rules of the sport must be changed to appease NBA beat writers who didn't like what they saw when watching the best offense in recent memory play without the best player in college basketball.

Or maybe not. Mr. Potash's Twitter bio proudly proclaims that he's a "Native Chicagoan." So maybe, just maybe, this is a FIB thing. One can only hope.

Thursday, February 12, 2015

A disastrous evening for the Foul-Up-3 strategy

We are in agreement that there is no obvious answer to the #FoulOrDefend question (that is, when up by 3 points with less than 10 second left, should a team play defense or foul intentionally?).

And I think we are in agreement that it is annoying that there are some people who insist that defending in that situation is just stupid. The events in college basketball last night are why.

As far as I can tell, there were six "Foul or Defend" scenarios, and here's how they went down:

Defend: 3 for 3

La Salle v. VCU, second OT. La Salle, up 3 with less than 10 seconds left, allowed VCU to take a 3-point jumper, which missed. They collected the defensive rebound, hit two free throws, and won.

Southern Ill. v. Loyola Chicago. Loyola, up 3 with less than 10 seconds left, allowed Southern Illinois to dribble up the court without fouling. Then the Saluki player fell down while trying to turn a corner, and turned the ball over. Loyola hit one free throw, and the game was over.

Tennessee v. Vanderbilt,  OT. Tennessee hit two free throws with 8 seconds left to go up by 3. They allowed Vandy's Riley LaChance to take a three-pointer with one second left, which missed. Game over.

Foul: 1 for 3, with an Instaloss

Tennessee v. Vanderbilt. We just learned that Tennessee won this game by defending in OT. But here's how the game got to OT: LaChance hit 1-2 free throws with :08 left to give Vandy a three-point lead. Vandy fouled intentionally with :06 left, and Tennessee hit both FTs to cut the lead to one. Tennessee then fouled immediately, and Vandy hit just one of the two free throws, so the lead was just two. Tennessee's Robert Hubbs III nailed a jumper to tie the game and send it to overtime.

Winthrop v. High Point. This was the nightmare scenario: the foul strategy leading to a loss in regulation. Winthrop was up 3 and fouled intentionally with just :05 left. High Point made both free throws to cut the lead to one, and then fouled Winthrop with still :05 left. Winthrop missed the front end of the one-and-one and then committed a foul going for the rebound with :03 left. High Point's John Brown nailed both free throws to win the game in regulation. INSTA-LOSS

Richmond v. Fordham. This was the foul strategy's only "success" of the evening, and even it led to some serious heartburn. Richmond took a three-point lead with :11 left, and fouled intentionally with :05 left. Fordham hit both free throws to cut the lead to one and fouled immediately, with :04 left. Richmond hit both free throws to go back up by 3, and then immediately fouled again with :03 left. Fordham's Antoine Anderson hit the first FT, then missed the second intentionally allowing Fordham's Mandell Thomas to grab the offensive rebound and attempt a two-point jumper as time expired. It missed.

So the defend strategy was perfect, and foul was 1-3 with a double asterisk: one of the fails was an insta-loss and the "success" allowed a higher-percentage shot at a tie than likely with the defend strategy.

Friday, January 30, 2015

Technical T-Rank Note

Close observer(s) of the T-Rank may have noticed that all the top teams' Barthags dropped last Friday. Very astute observer(s) would have noticed that the worst teams' Barthags went up.

This happened because I lowered the "exponent" used to calculate the Barthag. This is a constant in the formula used to calculate the Pythagorean Expectancy. Changing it doesn't change the rank order of teams, and different Barthags don't change how the projected scores of games are calculated -- but it does change the projected certainty of the result. In other words, that percentage figure in the T-Rank predictions. That figure is actually important, because it is what I use to project future records. (For example, a team that is projected to have a 50% of winning game 1, an 80% chance of winning game 2, and a 70% chance of winning game 3 will have a projected record of 2-1 because .5 + .8 + .7 = 2.)

I lowered the T-Rank exponent from 10.25 to 9.0 because favorites were not winning as often as projected. Also, the projected records for the top teams just seemed overly optimistic to me.

So that's the explanation for that.