colin

History

Member for

4 years 46 weeks

Blog

View recent blog entries

Karma

Current value

2

Recent Comments

Date	Title	Body
1 day 6 hours ago	Anyone else think Speight	Anyone else think Speight looked pretty mobile there? Surprising athleticism from a guy his size and (insert salt grains) recruit ranking.
1 day 9 hours ago	Scrambling gets dealt with a	Scrambling gets dealt with a couple ways. For one, you'll see that DEs don't get upfield much and look to knock down passes. The idea is to contain the passer and make him actually throw the ball. Teams did this a ton against Denard, for obvious reasons. Another option you'll see  is a spy. So maybe they rush 3 rather than 4 and have a James Ross type waiting to track down the QB leaking out or the dump off. I think this is the worst option. It never seems to work that well. There's also the zone blitz, which I like as a complement to the first option.  You get 5 rushing instead of 4 so you cover more gaps in the OL. Usually scramblers aren't great passers, so you're not giving up much to get more pass rush. And the one thing you always hear announcers talk about is playing zone pass coverages rather than man coverages. That's oversimplified for the audience a bit. Zone is so ubiquitous that the distinction is pattern reading vs. spot dropping. Here's a solid rundown from a blog I like: http://runcodhit.blogspot.com/2010/03/pattern-reading-vs-spot-dropping.html
1 day 12 hours ago	Ugh I still screwed up. 5 OL	Ugh I still screwed up. 5 OL plus 1 TE plus 2 ball carriers is 8, not 9. Plus a few annoying other grammar/syntax errors.  Whatever I'll take it.
1 day 22 hours ago	Yeah sorry I don't think I	Yeah sorry I don't think I was as clear as I would like. But WLA2 interpreted me correctly.  Hopefully this clarifies: It's not a simple matter of there being 8 in the box or whatever.  When the offense lines up and it's not an obvious passing down, the defense will set up with some number of run-first defenders to account for the run. In order to get a good run fit, you need one run-first defender for every blocker and every possible ball carrier. If Devin is in, we likely would not count him as a possible ball carrier based on what we've seen from Borges' playcalling with him at the helm. If Denard is in, he's of course included. So if M lines up 3 WR 1 TE 1RB, that's 2 ballcarriers and 7 blockers.  You need 9 in your run fit in this case. If MSU were lining up against us, you'd get the front 4, the 2 middle LBS, the LB lining up over the slot and the 2 safeties. That's pretty oversimplified, actually, but the point is you can usually discern what teams are trying to take away from you and what they'll give up.  From what I've seen, teams generally don't like to switch up their run fits a whole lot. They'll just bring in blitz packages and formations like Mattison does if they don't think you can run enough to be worth worrying about given the down and distance. So if teams are staying in base when you're in base, you can usually be sure normal run fits are in place and that they're playing you more or less straight up. Tangent: basically, college defenses are very modular. You see this with, for instance, what they're doing with Dymonte Thomas or how they in general give defensive backs a chance at corner first. Safety requires somewhat less athleticism, but a lot more know how.  With Thomas, they figure if they can teach him to play flat/seam pass defense and blitz, he can at least be a weapon on 3rd down. If they can add playing force defender to that, then he can probably play all three downs. From there it's just a matter of what formations they're seeing and what he shows he can do.  If it turns out he can also play over the top of the #2 receiver and MOF safety, then he'll be a full fledged starter who can be slotted in wherever you might want to play a guy of his talents.  I think this is basically what they were hoping they could get from M-Rob or Furman, but it turned out not so much the case. But, if, for instance, if Thomas never added anything beyond the flat/seam and blitzing, then him being in the game on 1st down would tell the offense a lot about the defense thinks of its ability/willingness/need to run. In the OSU case, I don't think they got out of base or were dialing up exotic blitzes or anything else that might betray an obvious lack of committment to stopping the run. My recollection aside, defenses will tell you what they think of your ability to run or pass. They will try not to, but they pretty much have to.
2 days 4 hours ago	I think this is right. And	I think this is right. And it's verifiable. All you have to do is check and see how many guys were in the box.  IIRC, usually teams were committing enough to make a proper run fit. I.e. they weren't playing a man down in the box because they figured that would be enough.
2 days 4 hours ago	I would really bet that if	I would really bet that if Borges did an evaluation of up-tempo teams that there were two things wrong with it.  First, I bet he neglected to make his results tempo free.  Without that, of course you're going to find that increasing tempo leads to the other team scoring more points. Perhaps more interestingly and maybe less likely: I bet there are selection effects.  Teams that choose uptempo are likely not doing it because they were front runners in the first place.  Those teams tend to be slow to change because their incentives are to stay relatively still. As you can see from Borges, it's relatively easy to let other teams pay the price of experimentation to see what works and what doesn't provided your team is the one with the talent advantage.  Being new or different is most likely to be a direction taken by a new coach and/or an old coach in a poor situation.
1 week 12 hours ago	Padmapper is a nice little	Padmapper is a nice little web app that takes craigslist ads and maps them onto google maps.  In the very least, it'll help you get some sense of what the market looks like.
1 week 4 days ago	J Walter Weatherman: And	J Walter Weatherman: And that's why you always write rap songs heavily featuring your own name.
3 weeks 10 hours ago	It's not that hard to get to	It's not that hard to get to 7-5.  Team is basically the same, but minus half a season of Denard and Jake.  That said, I'd be surprised if the rest of the B1G has enough talent to actually capitalize on our weakness.  We'll be at a talent deficit against ND and OSU.  That's about it no? 8-4 sounds right to me.
3 weeks 10 hours ago	Sounds like you're very well	Sounds like you're very well read and extremely open minded.
3 weeks 2 days ago	Obviously his brother being	Obviously his brother being here helps make the decision to visit, but he's visited here twice as much as ND and we offered well before they did.  That doesn't really scream "slight ND lean".  Regardless of what the 247 guys say, visits are usually pretty telling.
3 weeks 2 days ago	Just to add some more	Just to add some more context, 247's composite ranking has Irvin at 29, Walton at 47 and Donnal at 88.  Supposedly it was updated today, but I have no idea if the update actually includes the new ESPN rankings.  It shouldn't matter too much.  Past Beilein recruits: GRIII finished 17, McGary 28, Stauskas 82, Brundidge 87, Trey 93.  Aaaand that's as far back as 247 goes because they hate backdating. But between Scout, Rivals and ESPN, Smotrycz averaged 80. D-Mo was 77 to Rivals, 100 to ESPN and the 20th best PG to Scout. Vogrich was 100 to Scout, 131 to Rivals and the 40th best SG to ESPN. Obviously the trajectory is pretty good.  Also helps to see that Beilein does have some misses (Vogrich and Brundidge) but everyone else contributed as a freshman and at least showed flashes of things to come.
3 weeks 4 days ago	ugh.	ugh.
3 weeks 4 days ago	it was awful for forever?	it was awful for forever? it's good now though.
3 weeks 4 days ago	Given his FB velo, that's	Given his FB velo, that's most likely a curve.  Having a faster and slower curve is common-ish.
4 weeks 2 days ago	And given the number of punts	And given the number of punts per game, switching from an old-style to a spread punt is worth something like 1-2 points in expectation per game? 8-12 possessions, so, 4-8 punts per game? Maybe only 1 point in expecation?
5 weeks 2 days ago	How about ordering	How about ordering in likelihood of committing, including Ways? Ways, Williams, Scott, Holmes, Andrews?
7 weeks 20 hours ago	I've heard that next year's	I've heard that next year's class is supposed to be amazing?  I kind of just assumed that would somehow convince Craft to go? But yeah he'd be really selling low on himself.  He just had his worst season shooting from 3.  That's got to have him thinking he can do better.   And he's a much better NBA value if he suddenly develops a serious 3pt threat.  OSU could have been #1 this year if Craft can shoot, imo.  So yeah, fair enough. I don't have anything really to say about Ferrell.  I barely noticed the guy.  He was what, the 5th option on the floor?  But yeah there's not a whole lot to like in his numbers.  I don't know how his ORtg is that good, because his TO% sucks and his shooting was pretty meh.  But scouts seem to like him. Overall, I like your tiers.  I think you're underrating OSU a bit though.   I'd put them with Iowa/Wisconsin. The more I watched Thomas, the more I noticed how little he tried on defense.  He gave back a lot of his points.  I think Ross and Thompson will replace a lot of his production even if it looks even uglier than it did this year. Craft will regress to his mean and there will be talk about how he's better than he's ever been even though he'll be shooting at his career averages. I think it'll mostly come down to how much they cough it up. You're probably right to like Purdue.  This was Painter's worst team since his first in '06.
7 weeks 23 hours ago	Here's DX's most recent top	Here's DX's most recent top 100 freshmen update, which was last week.  Take whatsoever grains of salt you wish: http://www.draftexpress.com/rankings/NCAA-Freshmen/ 9. Glenn Robinson 41. Mitch McGary 47. Nik Stauskas   Others of note: 14. Gary Harris (State) 30. Yogi Ferrell (Indiana) 31. Sam Dekker (Wisconsin) 51. Adam Woodbury (Iowa) 65. Jake Layman (Maryland) 71. Jeremy Hollowell (Indiana) 78. Shaquille Cleare (Maryland) 84. Hanner Perea (Indiana) 90. Amadeo Della Valle (Ohio)   Looking at the sophs likely to return, OSU has a big class with a lot of talent that hasn't quite put it together (Williams, Scott, Ross, Thompson) but the B1G is otherwise pretty empty on that front.  Presumably Craft and Thomas are gone (2nd rounders, iirc, per DX), but if not they'll have their entire team back plus a year of experience.  Iowa is a pretty good sleeper pick if Marble and Basabe come back.  White, Woodbury plus those two give them a nice core.   For State, a lot depends on what Payne and Harris do.  I'd guess they'll come back, but who knows.  If they do, they'll be very good. Replacing Nix isn't that big a deal.  Not landing Parker was though.  Parker, Harris and Payne likely would've made them B1G favorites. Indiana will have a lot of talent, much of it being asked to handle far greater usage.  I'd guess they'll be fine given the pedigree, but there's some variance there.   M is basically in the same boat.  We (hopefully) lose one fewer lottery pick, but no Vonleh in the incoming class.  If GRIII stays, the talent level will once again be very high but we'll also be once again dependent on young guys.  It would be nice if THJr came back, obvs, but especially because he could help offset the youth at PG. Wisconsin will presumably be Wisconsin.  Goddamnit. Still lotta ins, lotta outs given the draft, but the contenders look like they'll be able to stay on top.  Iowa might be able to force its way into the conversation. Purdue could give it a go if Hammons really breaks out, but who else is going to score on that team? Minnesota loses its two best players.  IL hasn't recovered from Weber yet, but supposedly are in the running for Drew Crawford.  NW, PSU and Nebraska are bad. Picking from the favorites is a real slog because everyone has question marks even if everyone likely to come back actually does.
7 weeks 4 days ago	Actually Syracuse does	Actually Syracuse does remarkably well at defending 3P%.  KenPom has forwarded the idea that teams actually can't control opponents' 3P%, they just control how many they put up.  That's because year to year, the vast majority of teams don't see any correlation between 3P%.  Syracuse, surprisingly for a 2-3 zone team, appears to.  They make you take a lot of 3's and they apparently do a great job contesting them.  Best guess? It's related to their height around the perimeter.
8 weeks 1 day ago	I didn't say anything about	I didn't say anything about play calling, I said play design.  They consistently had trouble throughout the season executing runs that weren't inverted veer.  Whatever you think of the play calling, lack of execution consistently put us in trouble throughout the season. Meanwhile, there's no hint from this article that he's learned anything new.  Just that he is doing what he always does, which is look over what good offenses have done and how they were successful.
8 weeks 2 days ago	"Borges, who does a forensic	"Borges, who does a forensic study each offseason of the top offenses in the nation in an attempt to discover why those offenses worked, has a much greater understanding of the allure of the spread option after working with Robinson for two years. The offense can always scheme to overload the defense on the play side, overwhelming with superior numbers. Still, it has one major drawback. “There is a case for spread offense. Because it looks so good on the board,” Borges said. “There are no runs that look bad."" Not that it wasn't evident in the play design, but this is good to hear. Further evidence that Borges definitely gets X's and O's. Execution was the problem last year and that's obviously the other half of it. But there are plenty of B1G coordinators who seem to struggle with both. So cheers to that.
13 weeks 2 days ago	Have you found any papers	Have you found any papers that looked at natural experiments in this arena?  There are probably a decent number but my brief googling has turned up nothing so far, which I thought was weird.
13 weeks 2 days ago	I just tracked down	I just tracked down this: http://andrewgelman.com/2008/12/congressional_r/ Even considering this study doesn't (per Gelman...I'm not familiar enough with the statistical techniques in question to know) establish causality, it certainly seems plausible that Congress can do quite a lot to negotiate the specifics of NIH funding given the following: "In the House Appropriations Committee (HAC), the NIH budget request is handled by the Labor, Health and Human Services, and Education and Related Agencies Subcommittee (LHHE). A similarly named subcommittee of the Senate Appropriations Committee (SAC) evaluates the NIH budget request in that chamber. The LHHE subcommittees consider the NIH budget request, amend the funding requests in the presidential budget, and “mark up” the appropriations bills, sometimes specifically for institutes and centers at the NIH, that are ultimately reported to the House and Senate by each chamber's appropriations committee. The subcommittee meeting reports that accompany the appropriations bills to the floor contain additional detail and guidance on the allocation and disbursement of appropriated funds by the NIH. Transfers affecting the level of support may involve (i) reallocation for NIH funding among the agency's institutes and centers, (ii) subcommittee support for specific fields of biomedical research associated with particular diseases, and (iii) project-level transfers that reallocate funding among particular lines of research and/or research projects within a given disease field. (See Supporting Online Material for examples)." Just as a matter of agenda setting, it would seem Congress and assorted lobbies have the power to do what MJ suggests they can on an order of meaningful magnitude.  Again, that's far from establishing causality, but I would like to be convinced both that it can't and that it doesn't.  Assuming this mechanism is viable, I think the incentives for a given congressperson are clear.
13 weeks 2 days ago	Does it perhaps make a little	Does it perhaps make a little more sense to divide funds apportioned to a given state for research on some kind of merit basis? The way you have it is that Pitt doesn't add anything because we have PSU, right? But presumably PSU and Pitt split that money along some lines. It also seems to indicate that if we took Duke we'd be less likely to take UNC? Anyway, if I had to guess, the total is more or less determined by research status but the marginal dollar goes to the more politically connected district of the two. Do I have that right and would altering your formula along those lines change much?
15 weeks 1 day ago	It's worth mentioning that	It's worth mentioning that the NFL's struggles were at least in some part related to the fact that they simply haven't seen much spread to run stuff.  College defenses (like Narduzzi's, for instance) are clearly designed around the idea that the QB will be a run threat.  They know that traditional pass-only defenders need to be involved in the run.  While I'm sure plenty of NFL guys know it too, the number of times when teams just flat out screwed up their numbers just doesn't happen in college these days.  The scheme is too old.  Relatedly, I saw a great screen shot of the Bears lining up against one of the spread teams this year in obvious C2 (turned out to be C2 man, at that) against a 4 wide set.  That left 5 in the box against 7 for the offense.  Guess how well that turned out.  I don't watch the NFL that much but I bet if I went back to the All-22 and checked out the numbers on a lot of defenses that they lost the battle well before the snap.
15 weeks 1 day ago	Given the incentive structure	Given the incentive structure you laid out, why wouldn't cheating be rampant? Isn't that a pretty weird belief? Also: there's another party to add to the interested in seeing Phil and friends get busted. Namely, any non-cheating rivals have a very significant incentive to at least work through back channels in bringing this kind of thing to light. But I imagine they would have to be extremely secretive because there's very likely a lot of anti-snitching codes in place throughout the game. ADs are hamstrung by this, but boosters probably could devote some time/money to it. It doesn't have to be just Yahoo that funds Charles Robinson.
15 weeks 3 days ago	So M has two of the top	So M has two of the top couple players in the league but won't even be above average?  How bad are the rest of our guys?  No offense meant, I just assumed we were average-ish top to bottom for whatever reason.
15 weeks 4 days ago	I'm guessing it means there's	I'm guessing it means there's a decent chance that Jourdan Lewis pulls a Reverse Zion Babb.
15 weeks 4 days ago	That is something that may	That is something that may well be generally true, but a lot of the research that has gone on in biomechanical research in kines labs and the like is starting to be incorporated usefully into training regimens.  Given access to that kind of thing, athletic guys with quarterback stature and a feel for throwing are candidates for velocity gains.   Not saying Borges is that sort of guy, but I would imagine that as the throwing programs that are starting to proliferate in baseball trickle into football that the pool for quarterbacks will grow.

more