The spreadsheet I have with the numbers is at home and I'm on the clock so I can't recreate it all, but W&M in conference under Shaver has generally been 8th-last over the last 10 years or so. The 2014 Tribe team was 305th in defensive efficiency and the 2015 team was 262nd. Last year's team was 335th, giving up 115.2 points per 100 possession. Pomeroy and other metrics adjust for tempo, meaning how many possessions there are in a game. They also take opponents in to account.
W&M held their last two opponents under a point/possession, which is good. IIRC they were around 109 points/100 possessions at the start of conference play and have shaved that down to 107.7. (For reference, Texas Tech leads the country at 83.1 points; UVA is at 86.5). Last year's team gave up 115.2 points/100.
"3) What is the recommended amount of time we should be practicing defensive? I gather 90 seconds/practice is not enough."
The only answer I have is "more than what they've usually done."
"4) Where has W&M typically ranked offensively nationally? In the conference? Conference games only?"
From 2013-18 W&M was 1st in offense every year except 2017, when they were 2nd behind UNCW's car shredder of an offense. They were usually in the top-75 offensively; last year they were 27th. This year they're at 105.4 points/100 possessions, which is 122nd. That's 10 points down from 2018. (Last year's games were appointment viewing and hopefully made people money from betting the over.) They're currently 4th in the conference.
"I'm not a big fan of analytics-based basketball discussion but if we're going that way, let's go in all the way. Ideally, I would like to see a metric that tells us which players contribute the most defensively per minute (something like a defensive version of baseball's WAR)."
Those stats do exist:
https://www.sports-reference.com/cbb/sch.../2019.html
Basketball stats are still behind baseball stats, which makes sense- baseball has over 100 years of detailed box scores, the baseball sabermetric movement is older, the sport itself is easier to quantify and basketball teams have been very aggressive at hiring stat nerds and making everything proprietary before they can put too much out on the market. The biggest issue with defensive statistics in basketball is it's not always easy to isolate the player from the team.