{"id":390,"date":"2012-10-31T01:09:20","date_gmt":"2012-10-31T07:09:20","guid":{"rendered":"http:\/\/67.227.157.91\/~kenpom\/wp_blog\/outliers-and-sleep-deprivation\/"},"modified":"2012-10-31T01:09:20","modified_gmt":"2012-10-31T07:09:20","slug":"outliers-and-sleep-deprivation","status":"publish","type":"post","link":"https:\/\/kenpom.com\/blog\/outliers-and-sleep-deprivation\/","title":{"rendered":"Outliers and sleep deprivation"},"content":{"rendered":"<p>As I would do in previous seasons in the printed version of <i>College Basketball Prospectus<\/i>, I want to document a few of the perceived outliers in my preseason ratings for handy reference later in the season. This season, I&#8217;m including a breakdown of how much sleep I\u2019ve recently lost over each team.<\/p>\n<p><strong>1. Kentucky:<\/strong> The Wildcats are not really an outlier, but it needs to be pointed out that the system is a prediction of end-of-season data. It may seem a bit odd then that I use them to populate the ratings at the start of the season. In most cases, I don\u2019t think it matters much, but a team composed almost exclusively of freshmen would expect to improve as the season progresses. So Kentucky may not be the best team in the country right now. <\/p>\n<p>I\u2019d also point out that their current rating puts them fifth in last season\u2019s final list. The way to interpret this is that whomever ends up being the best team in the land will have exceeded expectations. (In other words, the top team at the end will have better numbers than Kentucky\u2019s expectation provided by the system.) And I think that makes sense in a year with so little preseason certainty at the top.<br \/>\n<strong>Sleep lost:<\/strong> 1 minute, watching All-Access Kentucky, a personal-best for me watching any reality show besides Big Brother.<\/p>\n<p><strong>2. Ohio State: <\/strong>The Buckeyes have arguably the best offensive player in the country and the best defensive player in the country at separate positions. Sure, perhaps the complementary parts are unproven, and maybe losing Jared Sullinger will cause the house of cards to collapse. But I think the Buckeyes have a reasonable shot at being the best defense in the land. (The <a href=\"http:\/\/grfx.cstv.com\/photos\/schools\/osu\/sports\/m-baskbl\/auto_pdf\/2012-13\/box_score\/stats_20121030aaa.pdf\">ugly showing against Walsh<\/a> gives me a teensy bit of concern, though.) If the offense is reasonably competent, watch out.<br \/>\n<strong>Sleep lost:<\/strong> 3 minutes, writing &#8220;Kentucky&#8221; and &#8220;Ohio State&#8221; into championship game of 2013 tourney bracket.<\/p>\n<p><strong>5. Wisconsin:<\/strong> My overrating of the Badgers last season is fueling their rise here. But even if one assumes they were in the 10-15 range last season as Vegas odds indicated, they\u2019d still be in the top ten now. I guess the question to people who don\u2019t want to rank the Badgers (or who do so apologetically) is: How much of a dropoff do you expect from Jordan Taylor to a redshirt freshman point guard? It\u2019s going to be big, sure, but it would have to be huge to justify the pessimism towards the Badgers. I don\u2019t think they\u2019re a top ten team, but something in the 10-20 range is reasonable unless you think last season was a total fluke.<br \/>\n<strong>Sleep lost:<\/strong> 5 minutes, staring longingly at the autographed framed photo of Bo Ryan that sits on my desk.<\/p>\n<p><strong>13. Texas:<\/strong> The Longhorns have lots of returning minutes and most of them are from freshman. Assuming Myck Kabongo plays, I don\u2019t get why they aren\u2019t getting more attention.<br \/>\n<strong>Sleep lost:<\/strong> 10 minutes, re-learning that Myck is pronounced like Mike.<\/p>\n<p><strong>19. Kansas State:<\/strong> I\u2019m willing to listen to non-numbers arguments for a team, but at least give me some logic. K-State was a borderline top 25 team last season and returns nearly everybody. They don\u2019t return their coach, and that seems to be the only reason to not rank them. If you think so little of Bruce Weber, then put that in your lukewarm previews of the Wildcats, because the players have proven they are good enough. Now I suppose you could actually like Bruce Weber and still say that it will be difficult for him to implement his system using Frank Martin\u2019s players. Either way, we saw both cards played a year ago in Missouri previews, so it will be interesting to see what happens here. I say Weber\u2019s the early front-runner for coach of the year, just because expectations have been set too low.<br \/>\n<strong>Sleep lost:<\/strong> 15 minutes, to shoot an e-mail to <a href=\"http:\/\/bruceweber.com\">Bruce Weber the photographer<\/a> about purchasing his domain.<\/p>\n<p><strong>24. N.C. State:<\/strong> The humans see a late Wolfpack run over the final three weeks of last season. Was that the real N.C. State or were they playing over their heads? If the former, then I could see pushing them ten spots higher. And there\u2019s enough human certainty about them being even better than that, that I\u2019m willing to admit there\u2019s a non-zero possibility of Richard Howell standing on the rim at the Georgia Dome on April 8, even if it lowers my algorithm\u2019s sense of self-worth.<br \/>\n<strong>Sleep lost:<\/strong> 20 minutes, to dig up youtube video of Cozell McQueen standing on the rim at The Pit.<\/p>\n<p><strong>46. Wichita State:<\/strong> This is here more or less to showcase what happens when a team loses nearly all its key parts. The Shockers are assumed to drop, but not catastrophically. The MVC media <a href=\"http:\/\/wcfcourier.com\/sports\/college\/uni\/mens_basketball\/mvc-basketball-bluejays-perched-at-top-uni-third\/article_466efaa2-2261-11e2-bb88-0019bb2963f4.html\">picks them fourth<\/a> and I\u2019ve got them second, if barely so. They may be overrated by me but there are a bunch of unknowns with the Shockers this season. <br \/>\n<strong>Sleep lost:<\/strong> 30 minutes, because I just can\u2019t live with myself if I don\u2019t get the top of the MVC right.<\/p>\n<p><strong>108. Cleveland State:<\/strong> Digging deep here, but I have to note that the Horizon preseason poll has the <a href=\"http:\/\/www.horizonleague.org\/blog\/valparaiso-chosen-horizon-league-favorite-detroits-mccallum-repeats-as-preseason-player-of-the-year.html\">Vikings finishing fifth in the league<\/a> while I have them winning it. Likewise, Valparaiso is the league&#8217;s pick, but I have the Crusaders fourth. Last season in the Horizon was weird. Valpo won the regular season, but I had them as the fourth-best team. Cleveland State was the best team and finished two games behind. They lose three senior starters, but one of them wasn\u2019t particularly efficient. Besides, they have rising sophomore Anton Grady and he will personally make each and every one of you respect the system by season\u2019s end.<br \/>\n<strong>Sleep lost:<\/strong> one hour, daydreaming about <a href=\"http:\/\/sportsillustrated.cnn.com\/2012\/writers\/luke_winn\/09\/04\/NCAA-basketball-sophomore-breakout\/index.html?sct=cb_t11_a0\">Grady and Alan Williams on the same front line<\/a>. Talk about respecting the system.<\/p>\n<p><strong>135. Maryland:<\/strong> The computer sees Terrell Stoglin as a guy who took a ton of shots and made an impressive amount all things considered, while those in the Terrapin basketball program saw him as a ball hog who drove teammates and coaches crazy. The computer doesn&#8217;t like to see him go. Everyone in College Park does. The ACC media <a href=\"http:\/\/www.bcinterruption.com\/2012\/10\/17\/3518878\/acc-basketball-preseason-poll-2012-2013-nc-state-duke-north-carolina\">has the Terps sixth<\/a> and I have them 11th (behind BC!). I really screwed up the whole ACC if the media poll is taken as the standard.<br \/>\n<strong>Sleep lost:<\/strong> 8 hours, contemplating possible existence of \u201cchemistry\u201d.<\/p>\n<p><strong>309. Utah and 338. Towson:<\/strong> My biggest issue with my own system is that when virtually nobody is returning it treats good teams (Wichita State) the same as bad ones (these). It assumes it\u2019s always bad to lose a bunch of players. Both Utah and Towson put up seasons that were among the worst in their respective conference\u2019s history, and they\u2019re turning the roster over because better players are arriving. Both of these will do better than advertised.<br \/>\n<strong>Sleep lost:<\/strong> None yet, but probably a few hours next offseason figuring out a better way to handle replacement-level concept.<\/p>\n<p>Finally, I have to mention <a href=\"http:\/\/basketballprospectus.com\/article.php?articleid=2426\">Dan Hanner\u2019s model<\/a>. It\u2019s quite a bit more sophisticated than my freshman-level regression work here and likely to be better in a general sense. (Though not in every case. Am I right, Michigan fans?) But like I wouldn\u2019t recommend ignoring the preseason AP poll while solely using my work (assuming my work was better, which it may not be), I wouldn\u2019t ignore my work whjle solely using Dan\u2019s. <\/p>\n<p>What I like about Dan\u2019s effort is that it uses a completely independent method from mine so there\u2019s the potential for wildly different results. Which is something that is becoming increasingly more difficult to find in human polls, lest one wishes to invite public ridicule. So there\u2019s some value in using all three rankings, which each use separate methods. The differences and similarities in teams across those rankings provide additional insight that wouldn&#8217;t be possible from using a single method (at least until Dan cranks out some probabilistic preseason ratings). And on that note, I need to get some sleep.<\/p>\n<p>&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>As I would do in previous seasons in the printed version of College Basketball Prospectus, I want to document a few of the perceived outliers in my preseason ratings for handy reference later in the season. This season, I&#8217;m including a breakdown of how much sleep I\u2019ve recently lost over each team. 1. Kentucky: The [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"_links":{"self":[{"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/posts\/390"}],"collection":[{"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/comments?post=390"}],"version-history":[{"count":0,"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/posts\/390\/revisions"}],"wp:attachment":[{"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/media?parent=390"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/categories?post=390"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/kenpom.com\/blog\/wp-json\/wp\/v2\/tags?post=390"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}