Evaluating each KU player's defense in Big 12 play


If you haven't checked it out yet, David Hess is doing some great Kansas defensive analysis on his blog, Audacity of Hoops.

Hess has started a "Project Defensive Score Sheet" where he breaks down every defensive possession of KU games to give us a better understanding of the Jayhawks' individual defensive performances.

(More on "Project Defensive Score Sheet.") I have used his data to compile each Jayhawk's individual defensive stats during Big 12 play.

The Jayhawks huddle around Kansas head coach Bill Self during a timeout late in the game against Missouri on Monday, Feb. 7, 2011 at Allen Fieldhouse.

The Jayhawks huddle around Kansas head coach Bill Self during a timeout late in the game against Missouri on Monday, Feb. 7, 2011 at Allen Fieldhouse. by Richard Gwin

Below is every KU player's Big 12 defensive stats, followed by a short summary of what we can take from the numbers.

Before we get started, here's a look at the advanced statistics we will be looking at:

Defensive rating — A measure of how many points an individual defender allows per 100 possessions. The lower the number, the better the defender. Used in conjunction with defensive possession percentage.

Defensive possession percentage — This lets us know what percentage of a team's defensive possessions that are credited or blamed on a defender.

Stop percentage — In my Recap blog, I often talk about floor percentage — the percentage of an individual's ended possessions when the team scores at least one point. Stop percentage is the opposite of that: What percentage of a defender's individual possessions does the opponent score zero points?

Let's take a look at KU's numbers. For reference, KU's team average for stop percentage is 0.53. KU's team average for defensive rating is 101.7.

(Here are the game-by-game individual defensive numbers in an Excel file, for anyone who wants a more in-depth look. Please note that some of Hess' numbers will alter slightly from mine due to rounding differences.)

Tyrel Reed

Reed's defensive numbers are almost exactly what you'd expect: Right about average. Reed's individual defensive rating (101.5) is barely better than the team average in Big 12 play (101.7). His stop percentage (0.53) is exactly on the team average.

The average defensive possession percentage for KU players is 17.2 percent (taking into account that on some plays, a player is unguarded, so no player is responsible for the possession), so Reed's 15.2 possession percentage indicates he isn't as involved in defensive possessions as some other Jayhawks.

Tyshawn Taylor

Taylor's defensive numbers reflect KU coach Bill Self's comments from Thursday: That Taylor has been better offensively than defensively as of late.

The guard's defensive rating of 102.7 is a point above the team average. That isn't horrible, but it also isn't good enough for a guy considered the Jayhawks' best on-ball defender. It is important to note, though, that Taylor often gets KU's toughest defensive assignment, which will negatively impact his defensive rating.

Taylor's stop percentage of 0.50 also is below the team average and also fifth-worst on the team. There's definitely room for improvement from the junior.

Elijah Johnson

These numbers seem to indicate one reason that Johnson's minutes have been limited lately.

Johnson is KU's worst defensive guard according to these stats, posting a defensive rating of 104 (well above the team average of 101.7). His defensive possession percentage is the highest of any guard as well (18.8 percent), meaning he's also affecting an above-average number of possessions, which isn't what you want from a player with a poor defensive rating.

It only makes sense that guards' stop percentages should be better than big men's stop percentages, as guards are oftentimes defending three-pointers — which are lower-percentage shots.

That fact hasn't helped Johnson much, as his stop percentage (0.47) still registers as third-worst on the team.

Brady Morningstar

Morningstar's defensive numbers were the most fascinating to me.

As you can see, the senior's defensive rating isn't good. He's second-worst among all guards in defensive rating (102.9). His stop percentage also isn't good (0.48), as it's well below the team average (0.53).

But here's the interesting part: Morningstar's defensive possession percentage is extremely low at 12.7 percent. In fact, it's the lowest on the team by a wide margin.

So what does this tell us?

For me, the low defensive possession percentage indicates that Morningstar is doing a great job of denying his man shots. This makes sense, as Morningstar has proven to be excellent at chasing his man through a series of screens.

The problem is, when Morningstar's man is able to get a shot, that person has been scoring a high percentage of the time. The low stop percentage indicates Morningstar's man has been getting high-percentage shots as well, meaning the senior is probably getting beaten off the dribble.

In summary, these numbers show Morningstar to still be a great off-the-ball defender — probably KU's best. But they also indicate that Morningstar isn't playing to "defensive stopper" level, as he's struggled to keep his man from scoring during those possessions when he allows a shot.

Josh Selby

If I asked you before this blog who KU's best perimeter defender was, I bet you wouldn't have answered Josh Selby.

Still, these numbers indicate — if nothing else — that we have been undervaluing Selby's defense during Big 12 play.

Selby has posted a defensive rating of 97.6, which is better than KU's team defensive rating of 98.8 during the six games he's played.

It's important to note, as with Taylor, that matchups can play a part in these numbers. Selby normally isn't getting the toughest defensive assignment, which will make his numbers look better.

Still, his 0.58 stop percentage ranks second on the team — an impressive number no matter who he's going up against defensively.

Mario Little

Can we all agree that Mario Little plays because of his reputation on offense?

If so, it makes his ugly defensive numbers a little easier to accept.

The 6-foot-6 Little — playing primarily an undersized four — is struggling to keep his man from scoring.

Little's defensive rating of 106.2 is worst on the team, rating 4.5 points above KU's average. His stop percentage of 43 percent also is tied for the team-worst — and 10 percent below KU's team mark.

KU hasn't been able to hide him, either. Opponents have been attacking Little, as he's been responsible for 21.9 percent of KU's defensive possessions while he's been in.

If Little isn't giving KU a boost offensively, it's unlikely that Self will leave him in too long, as the senior just hasn't been reliable enough defensively.

Travis Releford

It's a limited sample size, but Releford has graded out relatively well despite not being 100 percent with an ankle injury. His defensive rating (103.2) is lower than the team rating in his three games played (104.4), and his stop percentage of 0.56 also is above what would be expected.

Releford's reputation as a good defender is most likely deserved. I'll be interested to see if his numbers continue to improve as he regains his explosiveness.

Marcus Morris

Like Tyrel Reed, Marcus falls almost exactly where I'd expect him to be with these numbers: as a good but not great defender.

Marcus' defensive rating of 100.8 is just better the team average, as is his stop percentage of 0.55. His defensive possession percentage also is exactly on the team average (17.2 percent).

Not much to see here. Marcus is an above-average defender, but obviously he doesn't bring nearly as much value defensively as he does offensively.

Thomas Robinson

Though Robinson is fine defensively, these numbers don't show him to be the impact defender that many fans believe him to be.

Robinson's defensive rating (98.9) is slightly better than the team average in the games he's played (99.2), and his stop percentage is hovering right around the team average as well (0.54). He is involved in a lot of possessions (22.7 percent), which doesn't necessarily hurt or help the Jayhawks.

Much like Marcus, Robinson's numbers indicate he's a good — but not great — defender at this point in his career.

Markieff Morris

Look no further for KU's best defensive player. It's Markieff, and it's by a large margin.

Markieff's defensive rating is not only the second-lowest on the team (97.8), it's also nearly four points below the team average (101.7).

Not only that, he's kept his defensive rating low despite being involved in a large number of KU's possessions (22.4 percent).

Markieff's stop percentage also is tops on the team (61 percent), partly because he's a great defensive rebounder (which is factored into the percentage).

Like Selby, Markieff probably doesn't get nearly the defensive credit that he deserves.

Jeff Withey

Withey's numbers don't look good (105.7 defensive rating), but they're also skewed a bit by one horrible game.

Against Texas Tech, Withey's defensive rating was 135.4 during his five minutes. With only 22 total minutes in Big 12 play, an effort like that is going to have a huge impact.

Yes, his stop percentage is low (43 percent) and his possession percentage is high (26 percent), but 22 minutes in mostly garbage time isn't enough for me to make any sweeping judgments about the 7-footer.

Unguarded Opponent

Sometimes, no KU player is responsible for a defensive play. In those situations, defensive stats are kept for KU opponents when they are unguarded. David also uses this category for a few instances that he didn't think should be given to individual players, like technical foul free throws, unforced turnovers and the “we have to foul whoever has the ball” end-of-game free throws.

These numbers aren't what you'd expect. The Jayhawks' defensive rating (100.8) and stop percentage (0.56) in this category are better than the team averages.

I would take two things from this:

1. KU isn't giving up many uncontested shots close to the basket. Though a team never wants to give up wide-open shots, KU appears to be allowing mostly jumpshots away from the rim. If the Jayhawks had allowed more uncontested shots inside, we'd expect the stop percentage to be much lower than 56 percent.

2. KU's defense might be getting a bit lucky. I wouldn't expect the "unguarded opponent" defensive numbers to stay better than the team average for the entire season. Teams are simply missing open shots against KU, though perhaps a tiny bit of that can be attributed to the Jayhawks making opponents feel uncomfortable even when they are open.

In the last four games, no KU opponent has scored on more than 40 percent of its unguarded possessions.

In other words, KU's riding a lucky streak defensively — something that can't be relied upon in every Big 12 game.


ParisHawk 10 years, 7 months ago

Mark Twain mentioned the "three kinds of lies: lies, damned lies, and statistics".

When you do this kind of analysis in business, there are three kinds of reactions:

1) The analysis "confirms" what we already thought, ho hum. Waste of money. 2) The analysis goes against what we already thought. The analysis is flawed. Waste of money. 3) The analysis confirms much of what we already thought, but with some thought-provoking twists.

This article falls in the third category. Really interesting stuff.

I gather that Selby and Releford are two of our more promising defenders. Once they're back and 100% healthy, there's a good chance our defense will improve.

Michael Leiker 10 years, 7 months ago

Don't the unguarded numbers tell you that maybe you are looking at the wrong statistics in evaluating defenive effectiveness? Kind of hard to start with the idea that a defensive player negatively affects the offensive player's ability to score and that these statistics measure that effectiveness, get these results and run with the idea that you are using the right measurements. It seems like a large enough sample size that chalking it up to luck is off base as well.

What about steals, forced turnovers, help side defense, deflections...while these stats probably took a lot of work to put together, I'm not buying this as reliable analysis or anything we should judge someone based on.

It also completely eliminates the factor of who the individual players are guarding, which you do address in the article, but to me that seems like a significant enough factor to avoid publishing this as being anything meaningful in the first place.

Hard work, and appreciated as always but seems to me to be pretty weak analysis.

Alec White 10 years, 7 months ago

I agree completely. It's not that I don't buy some of these statistics, just that they don't tell the whole story. Mario and Elijah are the two worst defenders---and the stats back it up. But while Selby has made some incredible strides defensively, he isn't our best perimeter defender. I think defensive statistics are fun (just like any advanced statistic) but you can't read too much into them.

In theory, a good defense consists of each player harmoniously reacting with their teammates to take away the opponent's highest percentage shot at that specific time (ignoring other factors like shot clock and score). That's why we miss Cole the most...he could single-handedly erase multiple mistakes by his teammates with a blocked/altered shot. I think advanced stats tell us more about offense than defense, but usually it is an offense that imposes their will and forces the defense to react accordingly.

And one more question...who would get charged with the defensive "possession" if say an opponent blew by our guard (like that's ever happened) and scored on an And1 against one of our bigs? Also, thanks for the hard work.

Jesse Newell 10 years, 7 months ago

Rock_Chalk_25 — David explains his method in this entry:

To answer your question, oftentimes players get shared credit for a defensive statistic. It's obviously subjective (as some stats are, like assists and turnovers), but with enough data, some trends seem to distinguish themselves.

Jesse Newell 10 years, 7 months ago

leikness — Thanks for the comment. I think the unguarded numbers simply could be teams missing open shots. Not every open shot is going to go in, especially if it's a long jumper and especially if it's a poor shooter taking it. Yes, the number seems high, but I wouldn't say that it makes the whole study irrelevant.

Just as an aside, Dean Oliver (the creator of the stats above) set the groundwork for offensive rating, floor percentage and other individual statistics, which is much of the basis of the KenPom site. These sorts of advanced statistics are used frequently by ESPN's Jay Bilas, SI's Luke Winn, and many of the other highly respected college basketball analysts.

Oliver admits in the book that no statistic is going to be perfect. Heck, our current statistics aren't perfect. ERA in baseball isn't perfect.

I can say without hesitation, though, that Oliver's method are well thought-out and that his work is highly thought of. The reason it isn't seen more is because we rarely have people like David Hess doing the ground work to compile the individual defensive stats that he did.

Are the stats above perfect? No. Are they useful — more useful than any other defensive statistics that we have available? I would say so, and I hope others would agree.

Michael Leiker 10 years, 7 months ago

After sleeping on it, my biggest problem by far is the fact that it doesn't take into account who each player is guarding and their offensive effectiveness. Weight it for that and you give it some validity, but you just can't give it any credence the way it is structured.

KGphoto 10 years, 7 months ago

I appreciate the effort. But this is massive over-quantification. They are human beings, not robots. A million things happen on that court that no stat keeper in the world could ever see. Only a handful of these on-ball match-ups could be accurately judged as a stop or a fail. The stats are subjective. Meaning I could argue whether a man was being guarded or not, on any given play.

It's a huge, huge reach with no real purpose other than to fuel meaningless debate and pontification.

I really like most of your analysis Jesse, but you are on the fringe lately. This is a useless mess of roughly guessed numbers thrown together in an effort to out-geek some competitor's stat site.

Joel Thomas 10 years, 7 months ago

I don't know that I'd call it massive over-quantification. We've seen significant progress in baseball, but as with defensive numbers in baseball, they're more difficulty to quantify than offensive baseball statistics, since it's just the batter there in the box. Needless to say, the difficulty is compounded by what is happening on a basketball court. The stats aren't to be taken hook, line, and sinker, but rather with careful contemplation.

Tuskin 10 years, 7 months ago

Are they useful? Depends on how they are used.

Personally, I'd love to see defensive statistics well done. But, as Oliver notes, it's really hard. Even if his method is as good as any out there, is it good enough to rank players?

At the moment, I'm more comfortable using these statistics to highlight the differing styles of play. For example, Jesse notes the defensive rebounding of Kieff, or the off-the-ball defense of Morningstar. This usage of the statistics is insightful and illuminating.

But when "unguarded" does a better job than Morningstar and Releford, both of whom have been complimented by Coach on their defense, the statistical method itself begs for more study and/or improvement. Is "unguarded's" success skewed mainly by pure luck? If so, perhaps the entire analysis is too susceptible to luck. After all, "unguarded" plays more minutes than Morningstar and fields a higher percentage of defensive possessions, too - which means luck would find it easier to skew Morningstar's numbers. And if it's an indication that the shooter has gotten so rattled because he's playing Kansas, well, it seems to me that having Morningstar's hand in the shooter's face wouldn't be a balm for the shooter's nerves.

In all, these statistics don't seem to be a good means of comparing the various players. It would be interesting to study this to find out why, and to find out if its quirks can be corrected.

Perhaps its oddities partly relate to the fact that the best defenders will receive the toughest defensive assignments. (Is that more skewing to the statistics than the inverse effect is to accepted offensive statistics?)

Interesting article, as always, Jesse. You've given us food for thought. Thanks.

Greg Lux 10 years, 7 months ago

Or defensive problem is much simpler then all these numbers .. its very simple .. Our guards are getting beat off the dribble and our bigs ( twins mostly ) are not trying to block their shots in and effort to stay out of foul trouble, since we need their offense and rebounding. So its pretty simple all the opponent needs to do is get past his defender and he usually has a clear path to the basket. We are not doing a good job on help side defense and until we do we will continue to give up far too many points off the dribble. Right now we are covering for our poor defensive play with awesome offensive numbers. Offense can win games where defense plays poorly, but Defense will win games where the offense plays poorly meaning defense can win most of all games were offense can only win those games where we score well. HCBS knows that and that is why he is pushing defense so hard. Let's face the facts ... Offense is the clothes and makeup that makes you look pretty, but defense is the exercise ( not the fun part ) that keeps that trim figure that makes the makeup work it's best.

Rock Chalk

Scott Smetana 10 years, 7 months ago

I have access to NORAD's computers here in Colorado Springs. I ran a complex statistics program that took the computer 0.00000000000000001 second to tell us that Mizzery is pathetic in every imaginable category.

5DecadeHawk 10 years, 7 months ago

Interesting Jesse.

I see a fundamental flaw in the basic philosophy, though.

You're not factoring in the offensive potential of who each player is guarding.

A player's defensive prowess isn't determined ONLY by whether or not some guy in front of them scores or not. Rest assured that most anyone on KU's team could guard me and get a fantastic defensive rating. They'd school me faster than Evelyn Wood.

On the other hand, I suspect that using your system every single player on the team would have a terrible defensive rating trying to guard Kevin Durant or Paul Pierce.

The numbers that would become genuinely informative is the impact each defender makes upon holding down the offensive player's average output. What I'm saying is, that this defensive ratings system is on the right track, but it needs to weight each possession based upon the offensive rating of the opponent.

Stops against Kevin Durant should be rated a lot higher than stops against probably any player on Centenary.

Giving up points to Jesse on his Pearson Hall intramural team.( should hurt someone's defensive rating a lot more than giving up buckets to Paul Pierce.

If the numbers were weighted like that, I'm very confident that they would be a lot more enlightening.

However, you've given a good effort at the numbers. Thanks.

Jesse Newell 10 years, 7 months ago

5decade — Good points. I agree that these statistics could be improved with the information you listed (though it'd be tough to tabulate, I would think).

Not sure if you read this link I had above (, but it explains Dean Oliver's thought process for creating defensive rating.

His goal was to make a defensive statistic like ERA for a pitcher.

Think about it: ERA has plenty of flaws. One guy could be pitching in a small ballpark with bad defense in the AL East where he has to face the Yankees and Red Sox a bunch. Meanwhile, another pitcher could be pitching in a huge ballpark with a great defense in the National League.

Still, for hundreds of years, people have used ERA to compare one pitcher to another.

I'd like to think of this stat like ERA. Is it perfect? No. Can I still pretty much assume that a pitcher with a 2.50 ERA is better than one with a 6.00 ERA? I'd say that assumption is probably safe.

I'd say the numbers above aren't perfect, but used correctly, they can be used to tell us something.

5DecadeHawk 10 years, 7 months ago

Interesting perspective, Jesse.

Where I see a problem is that pitchers are exposed by detailed numbers. Like Batting Average vs Righties or vs Lefties.

There are detailed statistics showing every individual player's batting average vs every single pitcher. If you really want to drill extremely deep statistically you can get into how well John Doe pitches against Joe Blow with the bases loaded and a 3-2 count.

What's more, it's fairly easy to discriminate between relief pitchers that only face Lefties and starters that face the entire batting order. They are different types of pitchers with different roles. Comparing a closer's ERA to a starter's ERA is pretty much pointless and most people recognize that, so they don't bother to try to compare those numbers.

Similar kinds of issues will creep into a defensive performance statistic for basketball just like in ERA. It's just that we aren't familiar enough with where the statistic breaks down so that it doesn't really apply anymore.

Take Elijah Johnson for example.

He has a limited role on KU's team at this point in his career. In fact, one might draw parallels to a relief pitcher. His limited minutes automatially create a smaller sample size, and therefore a larger margin for error. Also... when Elijah enters the game, is the game situation such that representative statistics can be gathered?

If Elijah enters the game in mop-up minutes, is KU going to be playing high pressure defense that will allow him to maximize his steals, or are they going to play a basic defense that may allow the bigs to maximize their defensive rebounding stats?

If Elijah enters the game in the heat of a close ballgame, it's often because someone got in foul trouble, or because one of KU's other guards is not getting the job done. (maybe because an opponent's star guard is torching them?). In which case, it's probably more likely to be a more difficult opponent and more difficult situation for Elijah to be guarding. These phenomonon may skew the statistics against someone that is playing Elijah's role on this team.

I'm not saying that tracking an opponent's offensive efficiency for each defended possesion is a panacea either. Some of the same structural statistical flaws will still exist. I just think it would dramatically improve what the numbers can tell us.

As far as tracking the detailed statistics of exactly which player is guarding which opposing player on each possession, I agree that would be a difficult task to compile all those numbers. However, I think even a few generalizations could be helpful. Even if you did something as crude and grossly inaccurate as assuming the initial starting lineup is always matched up with the same guy from the opposing starting lineup it might be a modest improvement. Yeah, I'd like something better, but that would be fairly easy to compile. It's a compromise ... far from an ideal one... but not completely unreasonable.

5DecadeHawk 10 years, 7 months ago

Another point:

The unguarded ratings are somewhat of a misnomer. Some player on the team is almost always responsible when an opponent is left unguarded. The statastician may not know who to place the blame on, and I understand that, but usually somebody was supposed to be there with a hand in someone's face.

Another factor on the unguarded ratings is that the numbers may be indicating that KU is doing a good job of following the scouting report. It appears that when KU is double teaming, they are not leaving good shooters unguarded to double team bad shooters. If they are going to leave someone unguarded, they tend to leave a bad shooter unguarded.

Michael Leiker 10 years, 7 months ago

Exactly, and if you use the stats above, guys who leave their man unguarded would many times get a bump in their rating. Maybe Selby loses his man every 5th time down the floor for a reason!!

Jesse Newell 10 years, 7 months ago

A couple things to note:

  1. Through KU's first five Big 12 games, the "unguarded" defensive rating was 105.3 and the stop percentage was 0.46. That's more of what we'd expect from the "unguarded" line.

  2. I'd forgotten that David had used his "unguarded" category for a couple of instances that he didn't think should be given to individual players, like technical foul free throws and the “we have to foul whoever has the ball” end-of-game free throws. I honestly don't know where those numbers should go. It doesn't make up a majority of the ranking (especially for a team that's not behind much like KU), but it has some impact. He calls this category "team" defensive rating instead of "unguarded," but I didn't want to confuse folks with too many "team" labels above. Perhaps "team" is the better way to go.

John Reher 10 years, 7 months ago

I would suspect that most "unguarded" shots are perimeter shots rather than points in the paint. That would tend to reduce the likelihood of a make just because of the distance involved. Would also think that Morningstar's numbers are affected by the quality of the players he has guarded. Typically, he draws a tough defensive assignment and, when he hasn't been able to deny the ball, one might expect that player would score at a higher rate than the other players on the team.

5DecadeHawk 10 years, 7 months ago

Nothing wrong with you having an opinion about the difficulty of a player's defensive assignments.

However, when it comes to statistics, I'd like to see a basis in numerical facts that can support or refute such opinions and assumptions.

This ties directly to the point I was making.

If Brady is really guarding more difficult defensive assignments, and if the ratings system weighted things based upon the offensive player's offensive ratings, then Brady's defensive rating would reflect that. We'd be able to numerically calculate how difficult each player's defensive assignments have been, and how they have performed against them.

slowplay 10 years, 7 months ago

I like numbers as much as the next stat freak, but you can't quantify the intangibles of a man-to-man defense. Weakside help, help the help and double teams and all part of the strategy. Example, coaches will instruct defenders to steer the man they are guarding to a wing or the baseline where the weak side will pickup and/or trap. If that's not there, the defender looks rather foolish. Example, if a post player steps out on the ball screen and the pick roll is executed but there's no help, who's at fault? Who gets blamed for the defensive error? Example, a defender gets rubbed off by a hard screen because nobody was talking, who's fault? Defense cannot be measured by stats alone. A complete picture can only be had by watching each possession closely.

jhawklifer 10 years, 7 months ago

Sometimes stats and what we see live line up; for example, Morningstar's skillful denial defense, and subsequent burnage if his man happens to get the ball, are borne out by Jesse's post. I also tend to agree with other posters in that these metrics could be improved if they factored in the offensive capabilities of the specific player guarded, but in that case for sure stats do make a case for reality. Plus, Brady tends to guard skilled offensive players due to his psuedo-stopper status, and often the best way to stop those guys is off-ball denial no matter who you are. Either way, stimulating post from Jesse.

Kevin Huffman 10 years, 7 months ago

Has anyone else noted that while we're ranked 2nd in the country, Bracketologist Joe Lunardi has us as the 4th 1-seed (having to match-up against Ohio State if we make it to the Final Four) and puts our odds of making it to the Sweet 16 only 5th best even though he made us a 1-seed?!

Still working with a lack of respect among nat'l media. What a surprise! We just had the embarrassing choke year in last year's tournament maybe this year could be another one of those - give this team their due kind of years.

Kevin Huffman 10 years, 7 months ago

I think on Monday we saw a LOT more of Releford and a LOT less of Brady not because Travis was hot shooting but just because he's our BEST defensive player. And Travis helped shut down some of what Bowers and Matt Pressey were able to he didn't guard Phil Pressey, the PG as he was still going off, but things improved a little bit in some of the latter stages of that 2nd half from the other scorers. The Morris's were G*d-awful on defending Bowers and Ratliffe.

jayhawktalk 10 years, 7 months ago

Not going to look back at the box score from my phone, but didn't Brady play like 38 minutes against MU?

I watched Travis during the MU game, as I am still worried about his ankle. Could be I imagined it, but Travis still doesn't look 100% to me. He looks fine going straight, but also looked like he struggled to get around screens, moving laterally. I expect it is still painful for him to move side to side pushing off that ankle. Again, I'll be the first to admit that I may have only "seen" it because I was looking for it.

KU_FanSince75 10 years, 7 months ago


I love this piece and I am fascinated with all of the statistics you bring out. Very impressive----well done, sir!

jayhawktalk 10 years, 7 months ago

Going to have to agree (passionately) with 5Decade. W/out a modifier to account for who somone is guarding, for me, these stats, while not completely useless, have extremely limited value.

I mean, the fact that it basically refutes most of what everyone sees by saying, "all those guys you thought are good/great defenders, who are guarding the other team's best offensive players, really aren't as good defensively as you thought because their assignment scored as often as the average/sh***y offensive players on the other team who were guarded by the people you thought were weaker defenders..."

My guess is, you could make a reverse argument using these numbers that Jimmer Fredette only scores a lot because the guys guarding him suck on defense. That makes a lot of sense, doesn't it!?!

Commenting has been disabled for this item.