You Are Here > Sports-Reference.com > Blog

Blog Archives: MLB · NFL · NBA · NHL · CFB · CBB · Olympics

NCAA Tournament Game Previews

Posted by Neil on March 21, 2013

Want a handy guide to each matchup of the tournament? Check out our game previews. Each contains the individual & team stats you need to know before watching the game. You can access them for the first round in the table below, and they're always available on the previews page & the main page as well.

This entry was posted on Thursday, March 21st, 2013 at 9:10 am and is filed under Announcement, Basketball-Reference.com, CBB at Sports Reference, Features, Playoffs. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

8 Responses to “NCAA Tournament Game Previews”

  1. Steven Vickers Says:

    Neil-

    Great stuff. I have a question about the way individual ORtg for college basketball players is calculated on this site if you don't mind. The ones here seem slightly, but systematically, above what I've seen at Ken Pomeroy's site. When I've gone through the Dean Oliver formulas and replicated them in a spreadsheet, I get your values, so the only thing I can think of is that there is a difference in the multiplier for free throw attempts because of different free throw rules. Eyeballing players by how often they get to the line does seem to bear this out--a high free throw rate player like Mason Plumlee rates 117.2 here vs. 113.0 there, whereas a low free throw rate player like Quinn Cook is 112.8 here vs. 111.6 there, a much smaller difference. Does this make any sense--just trying to get a better grasp on the advanced stats used here.

  2. Pomeroy does use a different FTA multiplier for his possessions formula, so that may be the source of the discrepancy. Or he could just be using slightly different data. College hoops is sometimes questionable in whether certain games count toward totals or not.

  3. Steven Vickers Says:

    Neil-

    OK, that's what I thought. It shouldn't be a games included issue for the teams I looked at, though I know there are sometimes small data discrepancies (e.g. with "team" rebounds), so I suspect it's FTA. Thanks.

  4. Really handy stuff all in one place; as I always like taking an opinion when a bookmakers line is put up against an event, and then comparing it to what the maths comes up with. I've run the numbers for Fridays games, (assuming a normal distribution of 10 - Winston) using the Sagarin (Predictor) ratings from before the tournament started and most games against spreads come out at around 50/50 chance of covering (or close enough to not want to get involved and I can't even over the counter in the UK).

    The only one that stood out for me was Florida vs Northwestern (+20.5). Sagarin had it marginally less (18.6), SRS has it marginally more (23.6) but the telling thing for me is the NRtg/A that if I'm understanding it right has it at 36.8 (which I think is the probably more likely event, even though it is only a first round game and you have to question whether they will run the score up). I think it is also interesting that the lines this year seem to be around commonly available ratings and in this case it is probably high to try and offset action from Florida fans.

  5. #4 - It's important to remember that NRtg/A is denominated in point margin *per 100 possessions*, not per-game (as in the case of SRS). These college games typically run in the 65 poss/g range, so a 36.8 NRtg/A difference translates to pretty much what SRS is saying, a PPG margin of like 23, 24.

  6. Hey Neil,
    I've got a great topic for you next post - the Trailblazers bench. It looks like it could be one of the worst ever. If you remove Meyers Leonard, all bench players have PERs below 10 and their win shares sum to about 0.

  7. I mean PERs below 12

  8. Thanks - I misunderstood the rating somewhat as I didn't read it through fully (I just presumed it was a tempo adjustment with weighting on recent games). I definitely find it interesting the very slight variations between formulas for ratings that result in similar ratings for some games (e.g SRS and Sagarin Predictor both have Louisville and Colorado State as a 10 point game allowing for rounding up - the line is 10.5) but different for others (e.g. Oregon and Saint Louis - Sagarin has it at around 5 points, SRS at around 2, given the slight home edge - possibly worth a point - with a lot of Oregon fans being able to make the trip you'd have to question the line being at +3.5 Ducks). Given they had the slightly more difficult schedule (SOS) I like the possibility of another upset.

Part of the Network