Wednesday, April 3, 2019

Upon further review

So it's been two weeks since Selection Sunday and I've been enjoying the tournament too much to let my bad (relative to the Matrix) bracket projection bother me. But that doesn't mean I don't want to learn from it. So I put my seed list next to the Committee's to see just how far off I was.

I was spot-on with the committee on 18 teams. Nine of those were in the top 13 teams, though. Had I not flipped the Spartans and Bulldogs at the last second (literally!), it would've been 10 (I'd have had Sparty ahead of the Vols anyway). I can't decide if 18 out of 68 is actually all that impressive. Maybe a better way to look at it is, I was off by 3 or more s-curve spots on 15 teams (plus the ones I didn't have in the field, so 18). I guess that means that 32 were just off by 1 or 2 places. I don't know how that holds up against other bracketologists... or maybe I do, since 162 of them did better than I did in the scoring! But looking at it in that light makes me feel a little better. I think.

Unsurprisingly, my biggest whiffs (besides the three I included/left out) were Seton Hall, Minnesota (each +9) and VCU (-10). Totally rewarded the Hall too much for their BET finals appearance and punished VCU too much for only making their quarters.

Some closer calls ended up costing me points here and there. I had Virginia Tech as the first 5 and they were the last 4. I had Villanova as the last 5 and they were the first 6. I had Oklahoma as the first 10 and they were the last 9. I only had Purdue and Auburn off by two spots but it was enough to cost me points. So that, plus the Gonzaga/MSU switch, means I wasn't too far off from having an additional... 14 or so points. That would've gotten us into the the top 100.

Furthermore, as I said previously, I don't regret including TCU (or leaving out St. John's, though my son really pushed for them!). I felt Belmont was deserving, but I didn't think the Committee would include them. Indiana was the third team out, but I really should've known better regarding them and Texas. Temple I really waffled on, though. Just including Temple instead of, say, Texas, would've been an additional six points (because I'd have had the Owls as a 11 play-in).

That's an extra 20 points total (potentially), had just a few things broken my way and had I not made a snap last second decision. (I mean, I was closer on VT, Nova, OK, Purdue and Auburn combined than I was on Seton Hall alone!) A 150 would have gotten us into the top 50 or so. I'd have lost on TCU regardless, and probably the Johnnies, but I could've had Temple's points.

I guess what I'm trying to demonstrate (mostly to myself) is that you don't have to be off by much to really mess up your score. Being off just two or fewer s-curve spots for a total of seven teams cost me 14 points because those teams changed seed lines. And at the bubble, I won't beat myself up over TCU or St. John's, but just swapping Texas for Temple cost me six more (that one I AM mad at myself over!).


Wednesday, March 20, 2019

Bracket Preview

Well, my tournament predictions can't be any worse than my bracket projections this year. Visitors who read my article from the Observer will note a few changes from when I wrote that article on Monday. Here goes:

EAST: MICHIGAN STATE
I look all up and down this region and I'm just not seeing anyone that can knock off Michigan State or Duke. I can't blame Tom Izzo for being ticked at being placed in Duke's region. I sort of wonder, if they were the second 2-seed, Sparty wasn't sent to Louisville or Kansas City. Either is closer to East Lansing, and they'd have faced the #2 or #3 1-seed instead of the top 1. Still, who's beating them? LSU's coach is suspended. Virginia Tech isn't going to beat Duke at full strength. This will come down to the two heavyweights. In the early rounds, I'm taking Louisville over Minnesota, Belmont over Maryland and then to knock off LSU early, and Va Tech to beat Mississippi State. In the regional final, though, I'm going to take Michigan State. These teams are virtually identical on KenPom, and I'm going to bet that the freshman-heavy Blue Devils have that one bad-enough game that day against another Final Four quality team.

SOUTH: VIRGINIA
I don't like the South region. I don't think it's very strong, and I think I feel that way because I'm not feeling it from Tennessee. I know they're the top-rated 2-seed, and that they were close to being a 1 (I wonder if they would've been, had they won the SEC tournament...?). I liked MSU's profile better. That's why I took Cincy over Tennessee in the second round (that, and the game's like 90 minutes from Cincinnati). And I'm even more doubtful of Purdue, although fortunately for them Minnesota isn't likely to face them. I'm taking an experienced Villanova team over them in the second round (and would take St. Mary's if they upset 'Nova too). K-State will either lose a key player or have him hobbled somewhat. I'm taking UC Irvine over them. Wisconsin interests me, though. I think they should've been a 4, and I'll take them to make the Sweet 16 and give Virginia a run. I think Cincy goes to the Elite 8, but loses to the Cavaliers.

MIDWEST: KENTUCKY
Like the East, this region is very top-heavy. UNC and Kentucky would be very worthy teams in the title game. Houston could easily play spoiler against a young Wildcats team in the Sweet 16. This isn't Kansas' year (watch for a Northeastern upset!) but Auburn could keep up with the Tar Heels (and KenPom has them above third-seed Houston). So both top seeds will have their hands full in the second weekend. I'm thinking this is a chalky region to me - other than Northeastern, I'm not taking any upsets (I'm taking Washington to beat USU on a hunch, as USU seems to be favored by the ranking metrics). The Tar Heels are fun to watch, but I think Kentucky will beat them in the Elite 8. 

WEST: TEXAS TECH
I had a hard time with the West. Gonzaga ought to be a big favorite out here, but I see land mines for them. Syracuse, even without Frank Howard, will be a tough out in the second round. They've beaten really good teams who aren't prepared for that zone D. And even if the Zags' talent wins out there, Florida State will loom (after beating Ja Morant and Murray State, who will win the Ja Morant vs Markus Howard battle in the first round). That's a really tough game - I think Florida State is the best 4-seed (and I projected them as a 3). If they get that far, most likely Michigan or Texas Tech will be waiting for them, two really tough defenses. I am not a bull-iever in Buffalo and think Bobby Hurley will find a way to beat his old team, then fall to Texas Tech. Personally, I think the Zags get stopped by Florida State, and that Texas Tech will win the region. They're playing well, their D is elite, and if any of these 1-seeds is going to get ambushed early, it'll be Gonzaga. Not because they're not a worthy 1-seed, or an elite team. Just because the bracket broke that way.

FINAL FOUR:
Texas Tech vs Michigan State, I'll have to take the Spartans. They're balanced, strong on both sides of the ball.
Virginia vs Kentucky, I'm taking the Cavaliers.

FINAL GAME:
Virginia vs Michigan State: Finally, Virginia cuts down the nets. Only two teams have beaten them this year. Nobody plays D like them and their offense is more in balance with that this year. Sparty will give them a run but I think this team will be focused after last year's disaster and it's tournament-tested.

FINAL THOUGHT:
Virginia's region gives them the best chance of the 1's to break through. Tennessee, if it survives the first weekend, could give them a game in the Elite 8, but I just have a feeling that they're not escaping Columbus. Gonzaga has too many obstacles, and while I'm taking a big chance on Sparty over Duke, I feel comfortable with this final four.

Now watch Ethan Happ blow up in the second round and spoil everything...


Monday, March 18, 2019

Avert your eyes!

Ugh, please don't look! Nothing to see here.

Yep, the 17th Seed finished in 163rd place out of 195 this year, after finishing in the top 25 last year. We sure did struggle this time around. So what happened?

1. We whiffed on Texas, Indiana and TCU. Everyone whiffed on TCU, so no problem there. We should've known better about UT and IU. That's just too many losses, no matter who you beat or what your other metrics look like.

2. I'm not sure we'd have put Temple, Belmont or St. John's in, either. Dan (my son) argued for St. John's on the basis of their good wins, but I talked him out of it because their NET was so low. Score one for him. I've said all year that the Committee doesn't like to put in mid-major bubble teams at the expense of mediocre majors with better schedules, and here's Belmont with UNCG as the first team out. Maybe something's changed. And I didn't think they'd think Temple did enough, but in hindsight we should have seen that they had - good record, decent numbers. We'll have to look more into it. St. John's would've been worth 6 points to us (they'd have been in the play-in for us) and Temple would've too (they were a late subtraction), so 12 points.

3. We blew it on seeding four teams: Iowa, Seton Hall and Minnesota were all overseeded, and VCU was underseeded. That cost us 8 points. I thought Seton Hall's late-season run and the others' overall profiles were worth more, and we made a stupid mistake on VCU, penalizing them too much for the early loss to Rhode Island.

4. Our last-minute change from Gonzaga to Sparty for the last 1-seed cost us four points total. I love the Zags and they're a worthy 1, but I don't feel bad about this change. Michigan State won the regular season and tournament titles for a conference that got 8 teams in the tournament, they had the Quad 1 wins too. They were worthy of a 1. Still, we probably should have stuck with our earlier gut feeling.

5. I probably overreacted to the tournaments. I was sure the Seminoles would move up a line after beating Virginia. I thought less of Purdue after their early exit. I already mentioned VCU.

6. I should have spent more time on the bottom quarter (the AQ's). I mostly sorted them by NET and checked against SOS, but I should have dug deeper.

7. Most of all, we did terrible with the bubble, as evidenced by the teams we put in/left out. Partly that was due to time... we were rushed at the end. But that's also not great planning by us. But on the third hand, the committee threw us a curve with Belmont and reached (NET-wise) on St. John's over TCU.

Well, we'll take a look at it and do better next year. Right?

Sunday, March 17, 2019

FINAL BRACKET PROJECTION



Just under the Matrix deadline. We debated Gonzaga vs Michigan State and ultimately picked Sparty after their win, and we barely made it work in the bracket. I hope. We'll see how we did.

------------

Well, that was frustrating.

We took a risk going with Indiana and Texas, and lost on both. We chose them because of their profiles - Texas had the great SOS and some big wins, and Indiana had the Q1 wins. Of course, they also had tons of losses.

I'm not upset about missing TCU. That was a coin flip.

I am eager to read the justification for taking Belmont. Not because I'm upset that they made it - quite the opposite. I'm thrilled that a great mid-major season was recognized (even with a play-in game). But I'm wondering why they got in ahead of, say, TCU or Indiana.

I feel good about switching to Sparty. That wasn't at all a terrible decision by us. I can accept the Zags as a #1, that loss to St. Mary's shouldn't have been a game changer for them. But Sparty was the Big 10 regular season and tournament champ, a conference that got more teams in than any other, and they had the Q-1 wins to back it up and fought through injuries. I feel justified.


Friday, March 15, 2019

A response to Michael Beller's Bubble Showdown Series

Before going further, be sure you've read Michael Beller's article on SI.com. Here it is: https://www.si.com/college-basketball/2019/03/13/march-madness-ncaa-tournament-bubble-bracket-format

You're back? Good.

First of all, I like Michael's stuff and I learn a lot from his articles. I'm not pretending that I have anywhere near his level of knowledge or experience in this area, I'm just an amateur bracketologist thinking through what he said. So all due respect to the idea and its creativity.

Since Michael asked us to ignore the elephant in the room (the gobs of money that the conference tourneys reap and would likely lose through his BSS), I won't address that concern either. I just think there's a simpler solution: tweak the play-in games themselves. Here are the tweaks:

1. No AQs can play in the play-in games. This will mean that the 16's will not be involved in the play-ins. Instead, they will go straight to the round of 64, which is what they should get as an AQ. We'll finally have true play-in games that involve teams that haven't earned anything as of yet, and they will play for the last four spots in the bracket.

2. With eight teams involved, at least four MUST be from conferences that otherwise only have their AQ team in the tournament. This will mostly help the mid-major conferences, though occasionally a league like the Atlantic 10 or even the Pac 12 might benefit.

That's it.

Let's look at it in practice, using SI's latest bracket as an example. Their "last four in" are St. John's, Florida, Arizona State, and Belmont. Their next four out are Indiana, Clemson, Alabama and UNC Greensboro. In my scenario above, you would include Indiana and UNC-G and find two more teams from this year's one-bid leagues. Say, Davidson and Toledo (two "almosts" from the Bracket Matrix). Seed them any way you want.

St. John's vs UNC-Greensboro
Florida vs Davidson
Arizona State vs Toledo
Belmont vs Indiana

Clemson and Alabama might squawk (we'll leave alone the funny fact that those two schools just played for the national title in football...again) but really, they aren't among the best teams in their own conferences. They had their chance to prove something.

Bid thieves would cut into this the same way they do now, but would only affect the multiple-bid league teams, not the others. So if West Virginia steals a bid in the Big 12, Indiana would drop out, and the at-large above St. John's would drop into the play-in.

This change would not be as drastic as Beller's, nor would it cost anyone any money (except a couple of power conferences already rolling in it). In fact, it would distribute a little more money out to the middle class of conferences, and that's not a bad thing. Feel free to comment below.

March 15 Bracket Update


This will probably be my final update before Sunday's "final exam." Some decisions I've made since last time:

I've evicted St. John's from my bracket. Yes, 3-1 vs Villanova and Marquette is impressive. And the VCU win looks better too. But if I had a committee spot, I'd be seeing the 2-5 finish, the lousy performance against Marquette in their own backyard last night, and an unimpressive NET compared to some other bubblers and thinking this may have once been a tourney-worthy team, but not anymore.

They aren't quite equivalent, but some might say "Why Texas but not Indiana?". Well, Texas lost to a better team in more of a "road" environment yesterday. And Texas has four wins against top 4 seeds in the bracket (albeit, Indiana has three). Personally, I don't think either will make it in the end, and Texas is my last team in right now. But Texas is also a dozen points higher in NET and has Indiana whipped in SOS, plus they're noticeably higher than them in the other ranking systems. Texas has the better argument, at least between these two teams. Might not keep them in though. I can't imagine that the Committee will let in a 16-16 team as an at-large, despite the profile.

I have NC State in as an 11 right now, despite them having the worst non-con SOS in all of D-I. Doesn't putting them in but kicking Texas out reward weak scheduling? Does the Committee think about that?

And I've added Belmont. I won't lie, I'm glad to do this. I've harped a lot about how mid-majors usually get the shaft from the Committee. But if they're going to pay attention to their new NET rankings, Belmont should get an at-large. Furman's a little higher in NET and has the better "best" win, but Belmont has the superior non-con SOS and is slightly better in Q1+Q2 games. Had Villanova finished among the top four seeds, I might've given Furman this spot. But I like Belmont's profile a little better overall.

I need to research why some folks are putting Arizona State in as a rising 10 seed. A 67 NET, a 71 SOS, and two Q-4 losses? Must be the wins over Kansas, Mississippi State and Washington. I've heard that they value big wins more than they ding you for bad losses, but still.

I'm not penalizing Texas Tech for losing to the Mountaineers. They're still a 3 as far as I'm concerned. It would take some shuffling for TTU to lose enough ground to fall to a 4, especially since I thought they'd be worthy of a 2 if they won the Big 12 tournament.

Syracuse holds their 9-seed, I think. Buffalo's ceiling is probably a 6 if they win out and their floor is probably a 7 at this point.

Tuesday, March 12, 2019

For Game of Thrones fans!

Not sure how much of a Venn diagram we have showing the intersection of fans of fantasy like Lord of the Rings and Game of Thrones with fans of college hoops and bracketology (not to mention my obsession with Peanuts), but I'm in that zone myself. Game of Thrones is easily my favorite work of fiction and favorite television show, and not just because of the hype. If you haven't read or watched it yet, my recommendation probably won't get you to but I'll give it anyway!

So the point here is that, as a big fan, I am a frequent visitor to Tower of the Hand, an excellent fan site with amazing chapter and episode summaries and many deep-dive articles about the characters, story, and fan theories. They are having readers do summaries of each episode leading up to the start of the final season on April 14, and I was given two episodes in the middle of Season Four: Mockingbird and The Mountain and the Viper. I didn't know if I'd get selected at all, since while I visit the site a lot I haven't participated actively in it for many years. Not only was I selected but I also got two of the best episodes in what for me was the very best season of the show (taking place at the climactic end of the third book in the series). I thoroughly enjoyed reviewing and writing about these two episodes.

So if you're a fan of the books and/or show, please check out my review here: https://towerofthehand.com/blog/2019/03/11-mocking-mountain-s4e7/index.html. I highly recommend the site, www.towerofthehand.com. And don't let my poor graphic design abilities deter you.