Tuesday, September 25, 2012

weekend review

Funny thing.  We lost on Saturday.  I know that, you know that.  The scoreboard says so, the articles say so, and the fanbase is certainly acting like it.  So why don't I feel like we lost?

For one, it's not like we were supposed to win.  And it never felt like we were out of it.  At least not to me.  Mike London is right when he says there are a lot of positives to take away.  At the risk of sounding like Pete Hughes, a couple plays here and there and suddenly it's a whole new ballgame.  It's a huge week-to-week improvement to be able to point to a few plays instead of, like, all of them.

And that improvement has got me as optimistic as if we were coming off a win.  I wrote on just such a topic for InsideTheACC this week.  This is supposed to be a rebuilding year, right?  Last year was a fun little surprise.  This year is supposed to be a step back for reasons largely already detailed.  And the Hoos are 2-2 anyway, which I said is right where we want to be.

Now take a look at the schedule ahead and tell me that doesn't look like fun.  UVA can come out of the next four games with a 6-2 record.  That would probably make people too happy, but it's not like I'm gonna wish for 5-3 just for the sake of level heads.  The great thing about the potential four-game winning streak staring us in the face is that it would bring that coveted bowl eligibility nice and early.

Now that I've done everything humanly possible to jinx the pants off of the whole season, let's get into the game itself in more detail here.

-- After the Penn State game I took a look at what Penn State fans were saying about the game at their various Internet gathering places.  One of the themes was "argh why does UVA have all these Godzilla defensive linemen knocking everything down??"  (Lots of capital letters and bad language were added to the theme after the blocked extra point.)  This week, in the game recap, TCU's Casey Pachall specifically said he was throwing the ball higher to avoid swatdowns by defensive linemen, something that was probably emphasized in film sessions and hammered home when Jake Snyder batted down an early Pachall throw.  Casey Pachall is six foot five!  That makes it official: this is an asset going forward.  Several of the quarterbacks we face in the remainder of the season, including Louisiana Tech's Colby Cameron, are more like 6'2".  We might have to get a count going of rejected pass attempts.

-- I almost made it this far without saying the word "quarterback," though I wasn't talking about ours.  In a development that's likely to get me throttled by one of my legion of adoring yet frustrated fans, you'll have to wait til tomorrow to get a full rundown of that.  The situation has gotten to the point where it's worth a full post.  I'll leave it at this for a teaser: Phillip Sims throws one of the nicest damn spirals you'll ever see in college, so it's understandable that there's a clamor to see him play more.  It looks purty whether it's headed directly for the intended receiver's hands or the turf ten yards in front of him.

-- I was asstacular once again on my predictions this week, but I do deserve credit for one thing: I said Bill Lazor should attack the edges on the running game and damn if that didn't work.  Every running play that was worth a damn went outside, and every one that went inside got stuffed.

-- The most underrated reason we lost?  Maybe the actual biggest reason?  TCU's secret kickoff weapon.  You know it's not a good day in the field position game when it's the fourth quarter and the announcer says, "UVA starts this drive on their own 22, which is their best field position of the day."  That's awful.  It takes a really, really good, big-play-driven offense to consistently score when it has to go more than 83 yards on average to get to the end zone.

-- Several things were done well on Saturday.  Many things need to get better.  One thing has to get a lot better: Luke Bowanko's shotgun snaps.  We were lucky that one of them didn't end up rolling the wrong way downfield with the quarterback frantically chasing it.  It's obvious Bowanko isn't comfortable with the concept.  I'm no O-line coach but I would guess it means he's rushing the snap and trying to get to his blocking assignment quickly, and not following through correctly as a result.  Most of the snaps were high and required a little hop from the quarterback to snag them, and all of them floated in at half speed.  That throws the timing off, badly.  The even worse part is, the offense looked and felt more effective from the shotgun than from under center, meaning you can see the potential that's there if the snaps improve.

-- More on this and what it means tomorrow, but the most disappointing thing was this: Being three scores from the lead with nine and a half minutes to go is not an insurmountable obstacle.  Obviously TCU didn't think it was, because they put together a quick insurance drive to get the points back that Phillip Sims put on the board.  But our coaches treated it like one.  That was very telling as it regards to the quarterback situation, but, again: tomorrow.

Oh well.  Prediction review:

-- Not counting obvious attempts to mercifully run down the clock, the playcalling split for both teams will be higher than 60-40 in favor of the pass.  A quick look at the stats would seem to prove me wrong, here, but a closer look is required.  Two of TCU's "run plays" were kneeldowns, and five of them were Pachall scrambles.  That pushes them up to at least 64-36.  The same holds true for UVA; when you move sacks and scrambles into the passing category, the balance tips way, way over toward the pass.  More like 65-35.

-- Both quarterbacks will complete better than 65% of their passes.  Well, Pachall held up his end of the bargain.  I can think of three outright drops by UVA receivers that would've gotten Rocco closer, but as the stats will show he fell short by a wide margin.

-- Casey Pachall will top 350 yards.  I really thought our secondary did at least a decent job, and maybe holding Pachall to 305 yards is proof?  Especially since over 100 of them came on two plays.  Like I said: much better to be able to point to a couple plays instead of the whole game when you're looking for ways to improve.

-- UVA's running game again fails to generate 4 yards a carry.  No, Mike Rocco scrambling for 27 yards does not count as "UVA's running game."  However, Parks's big play does.  Take out that and I get it right.....but it's not fair to do that.

-- More of Rocco's passes are caught by TEs and RBs than WRs.  Should've said "Sims and Rocco" but either way, I am thoroughly right here.

A 2-for-5 showing moves me to 7-for-21.  Also damn you TCU for scoring that last touchdown; if I'd put money on this game the way I did on the Michigan game I'd be even more pissed, but as it is the only consequence is me dropping to 0-2-2 against the spread, but moving to 4-0 overall on game predictions.


You get this week's Blogpoll ballot and that'll be it for today.  I have a lot to go over re: the ballot, so once again, Senior Seasons gets moved to another day.  It'll probably be at the bottom of tomorrow's quarterback manifesto.  Das balloten:

SB Nation BlogPoll Top 25 College Football Rankings

From Old Virginia Ballot - Week 4

Rank Team Delta
1 Alabama Crimson Tide --
2 Notre Dame Fighting Irish Arrow_up 1
3 Florida St. Seminoles Arrow_up 2
4 LSU Tigers Arrow_up 2
5 Georgia Bulldogs Arrow_up 4
6 South Carolina Gamecocks Arrow_up 7
7 Florida Gators Arrow_down -5
8 Oregon Ducks Arrow_up 2
9 Kansas St. Wildcats Arrow_up 8
10 TCU Horned Frogs Arrow_up 5
11 UCLA Bruins Arrow_up 9
12 Texas Longhorns --
13 Oregon St. Beavers --
14 Ohio St. Buckeyes Arrow_down -7
15 Arizona St. Sun Devils --
16 Clemson Tigers Arrow_down -8
17 Northwestern Wildcats Arrow_up 6
18 USC Trojans Arrow_up 4
19 Louisville Cardinals Arrow_down -1
20 Stanford Cardinal Arrow_down -9
21 Arizona Wildcats Arrow_down -2
22 Ohio Bobcats Arrow_up 3
23 Miami Hurricanes --
24 Iowa St. Cyclones Arrow_down -3
25 Virginia Tech Hokies --
Dropouts: Oklahoma Sooners, Michigan St. Spartans, West Virginia Mountaineers, Georgia Tech Yellow Jackets
SB Nation BlogPoll College Football Top 25 Rankings »
So this week I move to the system that'll determine the ballot from here on out.  (Hence the large number of weird-looking changes.  If you want to suggest a change, I'm all ears, but it has to be based on somebody having a better or worse resume than I gave them credit for, not "why did so-and-so drop so far?")  I'm a long-winded ol' bastard who enjoys hearing himself talk (type) and am also fully onboard with the spirit of transparency that the Blogpoll operates on, so this is how it works:

-- Step 1: Determine an eligible pool of teams.  This is done by selecting a record that teams under consideration must have, or be better than.  This week, that's 2-1.  You had to be 2-1 or better (3-1, 3-0, 4-0, whatever) in order to be under consideration.  To further narrow things down, unless a team was 4-0, they had to have a win over a Big Five team (the Big East is no longer a power conference) in order to be considered.  (And if a team was 4-0, they had to not be named "UTSA" and have two of their wins be over Division II teams.)  This put 37 teams into contention (and removed a couple stalwarts like Oklahoma and Michigan State.  They'll be back though).  That's a little bit on the large side, but well within a normal range regardless.

-- Step 2: Each team's slate of games is ranked from 1 to whatever.  Step 2 guidelines:
  • Wins are always ranked above losses, no matter how competitive the loss or ugly the win.  This minimizes later juxtapositions of Team A's wins to Team B's losses and allows for an apples to apples comparison.
  • If a team has had a bye week, the bye is placed exactly in the middle, or if we're in an even-numbered week, one below the middle.  This week, byes are #3 of 4.  Again allowing an apples to apples comparison, and giving all teams with fewer games than weeks the same amount of "credit" for their byes.
  • Generally, of course, beating a Big Five team is better than beating a mid-major is better than beating a I-AA team, but I also look at it in terms of what you should and should not do.  For example, let's say you're Mississippi State (who came in 30th this week.)  You should beat Jackson State by 47; you should not beat Troy by only 6.  (Not if you want to be ranked, that is.  Therefore the Jackson State game is Miss State's third-best, and the Troy game is their worst.  These get looked at fresh each week, so if Troy goes on a Sun Belt rampage and ends up 10-2, that game will move up.
-- Step 3: Each team's best game is ranked against all the other best games, the second-best against the second-best, and so on all the way to the bottom.  The team with the best game in each category gets one point, the team with the worst gets 37 (or however many teams are there that week) and so on up and down.  Guidelines here:
  • It's better to beat a good team by a little than a mediocre or bad team by a lot.
  • If margins of victory and opponent strength are similar, it's better to have a low-scoring game than a high-scoring game.  Example: in considering two 21-point wins, 24-3 is better than 35-14 is better than 49-28.
  • When I get to the point where I have to compare wins to losses, I break it up into four categories, in order from best to worst: good wins, good losses, bad wins, bad losses.
  • In the section where byes are involved, byes are ranked just like a game.  For example, it's probably better to have had a bye than to have beaten Army by one lousy point.  Seven of the 37 teams have had byes so far, and they stretch from #25 to #31 in the ranking.  Those teams get the average of that, which is 28.  Yes, this punishes byes.  It should.  4-0 is better than 3-0, 3-1 is better than 2-1.
-- Step 4: All the points are tallied up - the fewer the better - and that determines the ranking.  Ties are broken by who had the best "best game."

-- Step 5: Fudge factor is applied if I think the system gave me a stupid result.  Most often this happens when two teams are very close, but the lower-ranked team beat the higher-ranked one.  (Yes, I'm aware there's at least two instances of this in this week's ballot.  I'm OK with those.  Stanford is being punished for beating San Jose State by three and Oregon State only has two games on their resume.)

So that is how the ballot is arrived at.  And that's why the Blogpoll is better than any of the "real" ranking polls; nobody in those polls will ever tell you their rationale behind their ballot.  Particularly not the coaches' poll.  I like this system because it fully considers every team's entire resume.


Anonymous said...

"Wins are always ranked above losses, no matter how competitive the loss or ugly the win."

"When I get to the point where I have to compare wins to losses, I break it up into four categories, in order from best to worst: good wins, good losses, bad wins, bad losses."

I like that you have a transparent system, but I'm confused about this part of it. Are you saying in the second quote that a good loss is better than a bad win? But in the first quote you say that wins are always better than losses? Is that just because this is a multi-step process?

I like the spirit of the rule that wins are always better than losses (you don't want to have a 5-5 team be in the top ten based on a lot of "good losses"), but I wouldn't make it a hard-and-fast rule, just a strong bias. You don't want to end up claiming that a 1-point win over Savannah State (on a lucky bounce! on a bad call!) is better than a 1-point loss to Bama.

Looking forward to the QB manifesto. I'd feel better about Sims if his bad throws weren't quite so bad. The phrase "wild pitch" comes to mind.

Brendan said...

Different steps, different guidelines. The rationale is this: from the perspective of the one, individual team, a win is always better than a loss. Always. I guarantee you, if a team were offered the chance to wipe one game clean and play another randomly chosen opponent, they'd pick the 1-point loss to Bama, not the 1-point win over Savannah State.

But then, obviously once you get to the part where you're comparing teams, an 11-1 MAC team is probably not better than a 9-3 SEC team, so the rule is loosened up considerably for that step.

Plus, as I mentioned, it minimizes how much I have to compare wins to losses. You want to compare wins to wins and losses to losses as much as possible. This is the best way to get there.

Anonymous said...

That makes sense, thanks for clarifying.

This is why I like the BlogPoll, it's not just some mysterious votes that invite charges of bias and conspiracy theories. You can see what everyone's doing.