Skip to content

Kentucky Sports Radio

University of Kentucky Basketball, Football, and Recruiting news brought to you in the most ridiculous manner possible.

Better Lucky Than Good?

When it comes to competition, luck is too often a factor.  Anything from a bad hop in the infield to a blown call by an official can be defined as luck in either a positive or negative light. Perhaps no photo illustrates that point better than Gordon Hayward’s last second heave for 2010’s National Title, which came ever so close to falling for Butler.  Some believe that “you make your own luck,” and others will tell you that luck is nothing more than a random occurrence. However you like to slice it, luck exists in some form or another. It’s sometimes viewed as a supernatural force that randomly chooses its victims or beneficiaries, but now we have a tool to measure it. has numerous tools in evaluating college basketball teams, many of which are well known to the masses by now.  But, there’s a little known column on the main page that actually measures luck.  How so?  As defined by Pomeroy, luck is “a measure of the deviation between a team’s actual winning percentage and what one would expect from its game-by-game efficiencies.  It’s a Dean Oliver invention. Essentially, a team involved in a lot of close games should not win (or lose) all of them. Those that do will be viewed as lucky (or unlucky).”  Or to break it down more simply with a literal example, if a team is involved in ten separate 1-point games, they, in theory, shouldn’t win all of them, nor should they lose all of them.  Eventually, a shot will fall, a foul-out will occur, or a blown call will be made swinging the outcome in favor of a team. This brings us to the all important question; how does this affect Kentucky? Surprisingly, I’ve made a chart to tell you.

Fret not if the numbers in the “rating” columns don’t make sense, it’s simply a fancy statistical way of telling you that a team is better or worse than their record suggests.  A negative rating further away from zero indicates a team is better than their record suggests. Likewise, a positive rating farther from zero suggests a team is worse than their win/loss record. Finally, a team with an average luck rating is about as strong as their record indicates.

The first thing you should notice, Kentucky is among the nation’s most unlucky, ranking 314th out of 347.  This, of course, suggests our Wildcats’ current 9-4 record is not a true indication of how they’re capable of performing.  Given Kentucky’s current Kenpom rating (9th), one should expect for the Cats to have a much better record than they currently do.  Critics of this system could argue that two of Kentucky’s four losses (Notre Dame & Baylor) weren’t competitive, and they’d be correct, but two of those losses were highly contested until the final horn sounded (Kentucky was in position to lead late in the Duke game).  However, with the way Kentucky has played against a majority of their opponents, they are, at least statistically speaking, better than their record shows.

So now we know that Kentucky is statistically unlucky, but what does this mean in the grand scheme of things?  The most important thing to keep in mind when viewing a statistic like this is its implication come March.  To get a clearer idea of what I’m talking about, here’s a brief history lesson.  In 2010, John Calipari’s first team was considered by some to be the best team nationally, but they eventually fell short to an upset minded West Virginia team in the Elite Eight.  Those Cats won numerous close games over the season and finished ranked 41st in luck.  2011’s Cats lost numerous close games in the SEC leaving many to question if they could muster any kind of post-season run. They exceeded all expectations and went on to the Final Four while being the 284th luckiest team.  Last year’s National Title squad won some close games and lost only two, but a large portion of games weren’t even close.  They finished 84th in luck.  It should be said that this stat does not determine national champions, after all the luckiest team could win the title or the unluckiest could fall in the first round.

One team won more than they should have and it caught up with them. One lost more than they should have, but they were eventually rewarded. The other performed to expectation and was rewarded as well.  This year’s team has been unlucky thus far, and that can certainly shift over the course of the season, but in my mind that’s an advantage working in our favor. Sure the Kentucky name on the front of the jersey gets the best from almost everybody, but many teams will see a “down” Wildcat squad due to its seed, only to be surprised with how well they can actually perform.

Article written by Jonathan Schuette

20 Comments for Better Lucky Than Good?

  1. Track Coach
    9:45 pm January 7, 2013 Permalink

    It continuously amazes me how many Kentucky “fans” bash their own for writing a piece that they obviously took a little time to write, and I’m assuming gets next to nothing in return.

    I enjoyed the article. Thank you.

  2. Dear UL and IU fans, We are national champions; you are not. The end.
    9:51 pm January 7, 2013 Permalink

    I thought this Te’o guy was supposed to be good?

  3. john
    10:04 pm January 7, 2013 Permalink

    Good article. Of course it’s hard to reach people who only care about wins and losses.

    Kenpom rated us 9th so far. We are there for a reason.

  4. Wild Blue
    10:05 pm January 7, 2013 Permalink

    So what do you call shooting 47 % free throws? Unlucky…..I think not .

  5. nassau65
    10:08 pm January 7, 2013 Permalink

    6. I was told the same thing. He is all they talked about the entire UK/ND bball game on tv. Mr. Te’o is a Mr. No-show so far.

  6. Jpo
    10:11 pm January 7, 2013 Permalink

    After seeing these comments why would anyone do that audition to become a blogger here lol

  7. WCAT FN
    10:18 pm January 7, 2013 Permalink

    Great post topic, as it gives some metrics against an issue that is ever-present in playoff/postseason discussions. It loses steam because there seems to be little correlation to actual performance. And as others have noted, the attempt to attach it to UK’s losses so far just doesn’t wash. I do appreciate the work that went into the post, and applaud the topic in these desolate times of only five games in the last month.

  8. Catlogic15
    10:25 pm January 7, 2013 Permalink

    It’s not that I didn’t enjoy the article, it’s just that I’m not smart enough to understand it I guess.

  9. James K
    10:30 pm January 7, 2013 Permalink

    Watch the clip of Gordon Hayward’s shot again some time… not much is said about the fact that Matt Howard COMPLETELY LEVELED Kyle Singler right before that shot. Hilarious. Seriously, watch it again.

  10. W.W.B.B.N.D?
    10:32 pm January 7, 2013 Permalink

    I bet B.T.I. would love this post.

  11. The dude
    11:10 pm January 7, 2013 Permalink

    I think I might be starting to feel bad for Notre Dame……………. nah run them over Bama SEC!!!!

  12. Manti T'eo
    11:18 pm January 7, 2013 Permalink

    I’m scared to tackle Trevor Lacey

  13. Matt Jones
    12:09 am January 8, 2013 Permalink

    Jonathan, this was actually a really smart and well-done post. You dont need to hear me say it, but ignore the folks who dont take the time to understand what you are trying to do. The rest do appreciate it.

    This crystalizes what will become clear as the year goes on. UK is much better than their record and ranking indicates. We all thought this by watching, but these stats showcase that as well.

  14. Matt Jones is king
    12:30 am January 8, 2013 Permalink

    If Matt Jones says it is good then so be it.

  15. bballjoker
    12:34 am January 8, 2013 Permalink

    This was one of the better articles I’ve read that wasn’t related to breaking news on here in a while. I think about this often, especially when Bill Simmons gets into his heavy gambling diatribes on his podcast. When Matt talks about the Louisville game actually giving him some hope for us to making a title run, this is why. The comparison to the 2011 Harrellson/Knight team I think is quite appropriate and I’ve been thinking it the last 2 weeks. Last year’s team certainly made their own ‘luck’ by making most games out of reach by the final minutes, but I would even say they were ‘unlucky’ to just play a bad game against a good vanderbilt team at the end of the SEC tournament, which they mentally did not let happen in the final tournament.

  16. KSRforCongress
    1:20 am January 8, 2013 Permalink

    This is well thought out, well written and a really interesting approach at explaining/quantifying luck. I agree with it to a certain extent, but I would argue against it because of the difference between teams like Butler and UK.

    These statistics would, in some ways, seem to be the basis for the “Butler was lucky” argument. Some would say they were lucky to reach the title game two years in a row because they have “lesser talent” and some would back that theory up with these stats because to reach 2 straight title games Butler had to win a ton of close games. One could easily think that luck simply ran out on the “less talented” team in the title games.

    But this is where Ken Pom’s numbers break down and he would admit so himself. What this really seeks to predict is how a team will execute in 1 point games when we don’t have enough data to make an accurate prediction.

    Butler played a ton of 1 point games in 2010 and 2011 and won a bunch of them. You could say they were really lucky. They weren’t. Brad Stevens would tell you he knows he’s going to be in a ton of 1 point games because of their pace of play. So Butler constantly practices executing in close game/game ending situations. Consequently, because they’re prepared to win those 1-point games, they do win them. They know where to be/what to do when they absolutely have to do it. That shot by Hayward certainly, without context can be viewed in terms of luck, but the only reason Butler was in that situation was because one play earlier Gordon Hayward failed to execute on the isolation play Brad Stevens drew up for him. It was rare that Gordon Hayward failed to execute, but I don’t think that means it was unlucky.

    Anyway, the ‘luck statistic’ fits for Kentucky. It doesn’t fit for Butler. Any measurement of luck needs to have preparation factored in. Teams, like Butler, that play a lot of close games aren’t lucky when they win, they’re prepared. They do what we should have predicted they would do. UK doesn’t play many close games and I doubt they practice close game situations very often (although I don’t know) so we don’t have enough data to accurately predict what will happen in close games. Consequently, we have no choice but to say it’s lucky when they succeed and unlucky when they don’t. We don’t have enough data to know.

    This UK team is “less talented” so we’re going to be in a few more close games. If we start to win these close games are we going to say they were lucky or are we going to say they were battle tested and learned how to win in close games? I would hope when these young Cats gain more experience and win close games, we aren’t so dumb as to say “our luck finally turned.” It robs Coach Cal and these young talented players of the credit they deserve when the time comes. We’re going to win some close games this year, I hope we don’t say it was luck because it wasn’t. It was experience and preparation.

  17. bart edwards
    3:30 am January 8, 2013 Permalink

    Pomeroy would have a cow if he read this post, as it completely misstates his luck stat. He has said for years that the term luck is not scientific at all in this usage, but rather a shorthand name for a statistical analysis of which teams underperform or overperform based on what his other stats predict. It is not luck. It manifestly does NOT mean that a team is better or worse than its record shows. I have no idea where you got that. A team is EXACTLY as good as its record shows (unless they’ve been cheated).

    The luck stat simply marks the extent to which the result of any given game tends to deviate from what Pomeroy’s metric had predicted. It exists so Ken can locate and try to eliminate glitches, like trying to see why every year Wisconsin seems to be “unlucky” in that it fails to win games the computer predicts it should.

    You should be more careful to read Ken’s blog before you put something like this out there. It’s bass ackwards.

  18. Paul
    6:30 am January 8, 2013 Permalink

    17- I take from kenpoms site that the luck factor simply quantifies the deviation from the norm (standard or random deviation, he makes no distinction). For example, if you are a 68% FT shooting team but happen to shoot 47% in a game and lose by 3, this would be ‘bad luck.’ Similarly, if your opponent shoots 35% from 3, you normally defend at 35%, but the opponent makes 50%, this is ‘bad luck.’ Doesn’t make any claim that someone wasn’t sick, that the team just didn’t lay a clunker, or any host of non-statistical factors which could very well imply the loss or win was exactly deserved.

  19. RealCatsFan
    7:39 am January 8, 2013 Permalink

    If there truly were a way to measure luck over the course of many seasons, Duke would come out on top as the luckiest team in college BB. Maybe it has something to do with the NCAA viewing them as “untouchable” when it comes to violations. That final shot by Butler not splashing through the nets is a direct violation of the laws of Basketball Karma. If Butler hits that shot, then Duke fans get to see why UK fans hate to see the Laettner shot played over and over again ad nauseum.

  20. Big Whoop
    9:02 am January 8, 2013 Permalink

    Aside from the meaning of “Luck” this was good analysis which is something that isn’t done at some sites around here from those who have a title of Analyst.

    My only complaint with this is that you didn’t include Duke. For me, right now, Duke and Michigan are the measuring sticks.

    Was Notre Dame “unlucky” last night or did they simply get their butts handed to them on a silver platter by the Crimson Tide? Evidence seems to suggest the latter.