[Cracked Sidewalks] Trying to explain Jae Crowder’s greatness

[CrackedSidewalksSays]

Trying to explain Jae Crowder’s greatness

Written by: jpudner@concentricgrasstops.com (bamamarquettefan1)

Jae Crowder was ranked as the 64th best player in the country by the new Basketball Prospectus (DJO was 41st).  When I crunched the numbers last month for the several thousand Division I players who will be on the court, I ranked Crowder even higher as the 48th best.

For someone like me who studies the 8 key player stats at www.kenpom.com, Crowder’s greatness is readily apparent because he is simply great at almost everything.  To simplify these 8 stats I sorted by each to show the percentile for each BCS player in each category so that the best 1% of BCS players at shooting were in the 100th percentile, and the worst 1% were in the 1st percentile, etc.

As great as I knew Crowder was, even I was shocked to find that he was the ONLY BCS player in the country that was in the 80th percentile or better in 6 of 8 key stats.  Here is the table for all Marquette players, which has your typical breakdown of players who are great at blocking shots but terrible at assists, or vice versa, etc.


table.tableizer-table {border: 1px solid #CCC; font-family: Arial, Helvetica, sans-serif; font-size: 12px;} .tableizer-table td {padding: 4px; margin: 3px; border: 1px solid #ccc;}
.tableizer-table th {background-color: #104E8B; color: #FFF; font-weight: bold;}


PercentileTORateFD/40ARateeFG%OR%Stl%Blk%DR%[/tr]

Jae Crowder	100%	62%	43%	81%	81%	83%	80%	88%
Davante Gardner	64%	95%	18%	88%	98%	5%	66%	67%
Chris Otule	19%	48%	1%	82%	86%	11%	98%	57%
Darius Johnson-Odom	72%	82%	67%	55%	19%	34%	34%	32%
Jake Thomas (RS)	83%	52%	49%	44%	12%	31%	42%	62%
Jamil Wilson	31%	13%	40%	21%	75%	39%	79%	68%
Vander Blue	17%	44%	62%	3%	53%	83%	52%	47%
Junior Cadougan	2%	48%	91%	7%	17%	31%	21%	23%

There were 18 other BCS players who were in the 80th+ percentile in 5 of the 8 categories, but you can argue that Crowder is statistically the most complete player in the country because he is not only phenomenal in 6 of 8 areas, but even in the other two is above average at drawing fouls and well above average for a front line player at assists.  Here is the breakdown of the 8 categories:

The following walks through the eight things a player can do to win games (that we can measure).  Crowder is …

1. GREATEST at not turning the ball over. This is the least known of the big 8 stats, but an offense can’t do anything else if the ball is turned over.  Jae Crowder was in the 100th percentile of protecting the ball, only giving it up 8% of the time, meaning only Marcus Denman of Missouri and Jordan Taylor of Wisconsin were better in all of D1 baskeball at protecting the ball.

2. GOOD at drawing fouls.  Getting to the line is key, and this is one of only two categories at which Crowder was good but not great, as he was in the 62nd percentile.  Davante was one of the best in the country (95th percentile) and DJO wasn’t far behind (82nd percentile).

3. GOOD at setting up a basket with an assist.  This is the only of the eight where Crowder’s percentile was below an average BCS starter at the 43rd percentile, but he was over the 60th percentile among frontline players so he is a good assist man.  Junior was in the 91st percentile in assist rate, a perfect traditional point guard if he cuts out the early season turnovers.

4. GREAT at hitting the shot.  Crowder was in the 81st percentile shooting based on effective Field Goal percentage (treys count at 1.5 made), due to his ability to score at the hoop or from beyond the arc.  Gardner and Otule were even better from the post, but Crowder’s number is even better than many centers who shoot form five feet and in most of the time.

5. GREAT at offensive rebounds.  Likewise, when a shot is missed Crowder is better than even many centers – in the 81st percentile at grabbing the offensive rebound.  Gardner was just a beast in the 98th percentile, while Otule was also excellent, and this is where Jamil Wilson could be so exciting – already registering in the 75th percentile of offensive rebounds in his freshman year at Oregon.

6. GREAT at steals.  When we get to the defensive end of the court, the only two returning MU players who are above average at steals are Crowder and Blue, but they are both WAY above average at a phenomenal 83th percentile (Crowder was actually one spot ahead of Blue in the national rankings).

7. GREAT at blocking shots.  If you can’t steal it, you can block the shot and Crowder is once again great in the 80th percentile in blocking shots.  Another reason to get excited about Jamil is that he was in the 79th percentile his freshman season, so he was already a great shot blocker and offensive rebounder even before his red shirt year of working out last year.  Of course, Otule was unreal, finishing in the 98th percentile in the percentage of opponents’ shots he blocked.

8. GREAT at defensive rebounds.  Crowder is BY FAR our greatest defensive rebounders and one of the best in the country, ranking in the 88th percentile.

Is Crowder the best player in the country?

No.

You can’t measure everything with stats, and some stats are more important than others, so you can’t weight these stats equally.  Kemba Walker was one of 18 BCS players in the 80th+ percentile in 5 of the 8 categories, but when these are weighted in my Value Add system he is obviously better for the number of minutes he plays, the possessions he takes over, and his ability to put the ball in the hoop, even if he is not great at offensive or defensive rebounding, or shot blocking.

Since it’s baseball season for those of you who don’t live and breathe MU basketball all year like me, let me make a baseball analogy.

Jae Crowder = Craig Biggio

In the late 1990s, when Barry Bonds was clearly the greatest statistical player in the game, Bill James shocked everyone by saying that Craig Biggio was the 2nd best player in the NL.  While everyone respected Biggio’s hard work and would love to have a guy who would give you a .300 batting average, 20 HRs and 80 RBIs a season, he hardly seemed to be one of the best two players in the league.

James explanation goes a long way to explaining Crowder’s greatness.  James said that he usually ignored the league leaders in a lot of “little stats” like doubles and triples, being hit by pitches and stealing bases because by themselves they didn’t have a lot of impact.

However, Biggio was unique because he was near the league leaders in so many little stats, stealing 97 of 115 bases (84%) in 1997-98, leading the league in being hit by pitches and in doubles several times, etc., that when you added all of the “little things” he did to his triple crown numbers he was better than all of the great home run hitters of the era except Bonds:


table.tableizer-table {border: 1px solid #CCC; font-family: Arial, Helvetica, sans-serif; font-size: 12px;} .tableizer-table td {padding: 4px; margin: 3px; border: 1px solid #ccc;}
.tableizer-table th {background-color: #104E8B; color: #FFF; font-weight: bold;}


BiggioBAHRRBI2B3BSBBBHPB[/tr]

1997	0.309	22	81	37	8	37	84	34
1998	0.325	20	88	51	2	51	64	23

Biggio was not as great a player as Bonds, who was hitting twice as many homers a year already to slug 100 points higher.  Likewise, Crowder does more things well than Jared Sullinger, for example, but Sullinger is so dominant on both sets of boards and can take the game over so much of the time, that he is by far the best returning college player this year despite being excellent in 4 of 8 key stats vs. 6 of 8 for Crowder.

However, Crowder’s greatness is among the best in the country, and hopefully he won’t go down as one of the most underrated players by our fan base as they come to appreciate all of the things he does.

http://www.crackedsidewalks.com/2011/07/trying-to-explain-jae-crowders.html

[GGGG]

Jae is a very good all around player, but I think statistics might overrate him defensively.  Yeah, he gets steals and blocks shots, but he was oftentimes out of place defensively (eg, Vanderbilt) in a way that negated any positive defensive impact he had.

Hopefully, with another year in the program, he will be able to be better defensively.

[bilsu]

Citing a play in the Vanderbuilt game as proof Crowder is not as good as his statistics is a little over the top. A bad play does not make you a bad defensive player.  Crowder when in the proper position is a very good defensive player.

[mu03eng]

Citing a play in the Vanderbuilt game as proof Crowder is not as good as his statistics is a little over the top. A bad play does not make you a bad defensive player.  Crowder when in the proper position is a very good defensive player.

Well that's a rather large assumption isn't it?  When in position.....I agree with Sultin these stats don't pass the eye test.  I love Predator, but he is out of position on defense pretty frequently.  Obviously we have no way of proving it but there were a lot of times I remember Crowder being out of position forcing someone to help and the help defender's guy then beating us.

[🏀]

I agree that Crowder has a ways to go defensively still by the eye test. However, he did show improvements throughout the season. I expect he will drastically improve this season.

1.) Transition from JUCO to Big East.
2.) He was late coming to Marquette.

[GGGG]

"When in position" yeah he is a very good defensive player.  The problem was last year he was out of position...slow on rotations, etc.  That is to be expected somewhat your first year playing at this level.  As I said, hopefully that will change.

[bamamarquettefan]

Jae is a very good all around player, but I think statistics might overrate him defensively.  Yeah, he gets steals and blocks shots, but he was oftentimes out of place defensively (eg, Vanderbilt) in a way that negated any positive defensive impact he had.

Hopefully, with another year in the program, he will be able to be better defensively.

And I do ackowledge (and did when I did a piece on defensive stats) that we don't measure the majority of what happens on defense.  I guess my point here is that of the 8 key factors that are measured he is the only guy in the country who is great at more than 5 of them, then even in a worst case where we said, "there are 2 other factors we can't measure and he is TERRIBLE" at both of them, he would still be near the top of the country in 6 of 10 things a player can do to win.  Even in this worst case, that would still make him one of the dozen most complete players in the league.

I always perceived more defensive holes with our guards letting players penetrate too easilyy than I saw with Jae making mistakes on on rotation, etc., but that is getting into the subjective evaluation realm, where I admit you and others may have a much better eye for what is happening than I do.

[Golden Avalanche]

"Greatness" is a ridiculously strong word to describe Crowder in any context.

[bilsu]

Well that's a rather large assumption isn't it?  When in position.....I agree with Sultin these stats don't pass the eye test.  I love Predator, but he is out of position on defense pretty frequently.  Obviously we have no way of proving it but there were a lot of times I remember Crowder being out of position forcing someone to help and the help defender's guy then beating us.

I think that is the problem with MU's defense as a whole. The defense relies on pressuring the ball, which results in players being out of position. Where it is seen most is the oponents' three point shooting. What surpsises me the most about this type of defense is that it does not result in more steals.

[bamamarquettefan]

"Greatness" is a ridiculously strong word to describe Crowder in any context.

Ah, so hard to argue with the old blind assertion.

So basketball prospectus and I used different approaches but both determined that jae crowder is in the top 1 1/2 percent of all division 1 players in the country (top 45 players would be the top 1 percent.  From your statement there is no way to know if you:

1. Don't believe top 1 1/2 percent is great,
2. Don't believe crowder is in the top 1 1/2 percent or
3. Are from the old school that simply doesn't think particular players are good because you just don't, and really don't like evidence confusing things so don't present any.

But without which of those 3 are the basis for your statement, there's really no basis for disputing or confirming your assertion.

[HouWarrior]

Jae is a very good all around player, but I think statistics might overrate him defensively.  Yeah, he gets steals and blocks shots, but he was oftentimes out of place defensively (eg, Vanderbilt) in a way that negated any positive defensive impact he had.

Hopefully, with another year in the program, he will be able to be better defensively.

Agree. He needs to improve positioning(mental), and foot speed(physical) to improve defensively, and eliminate the dumb reach in fouls caused by his lacking these two skills. The NBA will look for this too, as only 4s and 5s are allowed these problems...the 3s need more speed than he's shown so far.
He is now, and will be a very good college ballplayer...but, with the upcoming stronger draft class, I dont see him NBA drafted, except maybe late 2nd.

BTW, I watched Biggio for years. He is a HOFer, but..The stats dont show his clubhouse politics, getting Larry Dierker fired, his disappearing in the playoffs,...or that both he and Bagwell were at the very least Andro cycling. Bidge will get into HOF, but few in his time played totally clean. It repopularized BB after the strike, but permanently skewed statistics of that era viz other times.
Imagine Babe Ruth on Steroids---both his numbers, and the off field antics, fueled with roid rage, would have been amazing--although maybe his women performance totals would actually have been far worse--there's shrinkage(lol).

[brewcity77]

It amazes me how down we are on Crowder. I'll agree that he never looks good on defense, but as a team, I seem to recall our defensive efficiency being the best when he was on the floor. Was that because he was frequently playing with Otule and Butler? Maybe, but then why wasn't the team's defensive efficiency better when they were on the floor than when Crowder was. I'm honestly asking, that's one thing I don't get, because he wasn't our best defender.

What's so impressive about Crowder is how much he does. He never gives up the ball, he rebounds well, he blocks shots, he's a good shooter, he's a good passer, he just does a bit of everything. While he's not a 20 ppg scorer (and probably won't be next year either) or a 10 rpg guy (and probably won't be next year either) he's still good at both, and good at so much more.

I think Crowder is just one of those guys you want on your team. He might never be the superstar, but he'll work his ass off and he'll do a lot of little things to help the team win. In addition, he's an emotional leader. I really think as a senior he could be that vocal presence that we seemed to lack at times last year. And while his numbers might not be among the top of the charts in the traditional "big" stats, he'll still have great all around numbers when you break it down from a sabermetric standpoint.

[Henry Sugar]

As good as Jae was in many categories last year, what isn't captured is his inconsistency.  Crowder had a lot of games last year where he had a stinker of a game.  Most of those games, MU lost.

At one point, I had a theory that Crowder was the most important player on the team.  Butler could be dialed in for his 15-9.  After the first part of the season, DJO was probably going to get his points, but it was a matter of DJO shooting 8 times or 15 times to get there.  However, Crowder could just as easily put up 3 points as he could 15-16.  If all three guys were on, MU probably won. 

Check out the game logs on the last group of losses (not including the UNC game)

UL (BET) - 10 points on 3-9
@SHU - 1 point on 0-2
UC - 3 points on 1-4
SJU - 15 points on 4-10
@GU - 4 points on 2-7

Crowder *can* be great, but frequently is not.

[g0lden3agle]

As good as Jae was in many categories last year, what isn't captured is his inconsistency.  Crowder had a lot of games last year where he had a stinker of a game.  Most of those games, MU lost.

At one point, I had a theory that Crowder was the most important player on the team.  Butler could be dialed in for his 15-9.  After the first part of the season, DJO was probably going to get his points, but it was a matter of DJO shooting 8 times or 15 times to get there.  However, Crowder could just as easily put up 3 points as he could 15-16.  If all three guys were on, MU probably won. 

Check out the game logs on the last group of losses (not including the UNC game)

UL (BET) - 10 points on 3-9
@SHU - 1 point on 0-2
UC - 3 points on 1-4
SJU - 15 points on 4-10
@GU - 4 points on 2-7

Crowder *can* be great, but frequently is not.

Is it possible that the full grind of the a Big East regular season got to the guy? I remember there was a point where he tweeted about how Buzz was still running 2 a days during a lull in their schedule.  While pushing his players so hard during the season may have an immediate negative impact, I would be surprised if it didn't push guys like Crowder to the next level conditioning wise.  Crowder should be better prepared for a full season this year, and hopefully that will equate to less poor performances down the home stretch.

[Henry Sugar]

Is it possible that the full grind of the a Big East regular season got to the guy? I remember there was a point where he tweeted about how Buzz was still running 2 a days during a lull in their schedule.  While pushing his players so hard during the season may have an immediate negative impact, I would be surprised if it didn't push guys like Crowder to the next level conditioning wise.  Crowder should be better prepared for a full season this year, and hopefully that will equate to less poor performances down the home stretch.

I'm not certain it was the grind of practice as much as the adjustment to D1.  I believe it was a combination of two things.  First, I think the overall adjustment got to him.  Second, I think that the league started to scout Crowder better as well. 

I largely agree with your point that we could (should?) expect more consistency.  Haven't looked into any data to see if that happens, though.

[Golden Avalanche]

Ah, so hard to argue with the old blind assertion.

So basketball prospectus and I used different approaches but both determined that jae crowder is in the top 1 1/2 percent of all division 1 players in the country (top 45 players would be the top 1 percent.  From your statement there is no way to know if you:

1. Don't believe top 1 1/2 percent is great,
2. Don't believe crowder is in the top 1 1/2 percent or
3. Are from the old school that simply doesn't think particular players are good because you just don't, and really don't like evidence confusing things so don't present any.

But without which of those 3 are the basis for your statement, there's really no basis for disputing or confirming your assertion.

To each their own. I understand there are "numbers guys" who believe a certain truth with the way those numbers are laid out. I also believe there are "real world" guys who use the old fashioned eye test as their barometer.

Crowder had a very good season. He transitioned well from a different level. He was monstrous in certain spots. That said, its hard for me to accept that Crowder, who wasn't even the best player on his team, should be labeled "great" as if to have a differing opinion is seen as lunacy.

You want "great"? Put down the numbers and watch Kemba Walker play. That is greatness personified.

[Skatastrophy]

You want "great"? Put down the numbers and watch Kemba Walker play. That is greatness personified.

Did you watch Kemba Walker last season?  He was about as average as you can get during the entire conference season.  If he hadn't turned it on for the post season I'm not sure UCONN would have even made the tourney, much less win the championship and get Walker drafted.  He played like absolute crap for most of the year.

[Hards Alumni]

Did you watch Kemba Walker last season?  He was about as average as you can get during the entire conference season.  If he hadn't turned it on for the post season I'm not sure UCONN would have even made the tourney, much less win the championship and get Walker drafted.  He played like absolute crap for most of the year.

Except that he was amazing for the non conference portion of the year.

[Henry Sugar]

You want "great"? Put down the numbers and watch Kemba Walker play. That is greatness personified.

The numbers actually back up Kemba having a great year last year.

Just sayin'

[RawdogDX]

From your statement there is no way to know if you:

1. Don't believe top 1 1/2 percent is great,
2. Don't believe crowder is in the top 1 1/2 percent or
3. Are from the old school that simply doesn't think particular players are good because you just don't, and really don't like evidence confusing things so don't present any.

Pretty sure it is number one.  Golden probably puts about .1% in 'great,' if that.

[MerrittsMustache]

Craig Biggio being the second-best player of the 90s backs up the argument that statistical analysis of athletes can be very, very flawed.

[bilsu]

Craig Biggio being the second-best player of the 90s backs up the argument that statistical analysis of athletes can be very, very flawed.

I think statistics are flawed, because it is usings players statistics for a whole year. I think everyone here would agree that Crowder was a much better player in March, than he was in November. Generally, every player will be better later in the season. The question is whether Crowder improved more than the average player, and I would say that answer is yes. I think many fans had high expectations for Crowder, because his was junior college player of the year and therefore early on he was a disappointment in their eyes. Besides that greatness is generally not determined by one play. Some body posted his stats in five losses, but I think you will also find at least 5 losses that DJO's stats were poor. If you buy into Buzz's idea that every possession matters to MU and we had no margin of error, than any player on the team having an off game will result in a loss.

[Pakuni]

Did you watch Kemba Walker last season?  He was about as average as you can get during the entire conference season.  If he hadn't turned it on for the post season I'm not sure UCONN would have even made the tourney, much less win the championship and get Walker drafted.  He played like absolute crap for most of the year.

And by "absolute crap" you mean 21.7 ppg, 5.3 rpg and a 2.5-to-1 assist to TO ratio, all while playing about 39 mpg.
It's true Walker didn't play as well in conference games as he did in the nonconference schedule,- one could chalk that up at least partially to the higher level of competition game in, game out - but absolute crap seems perhaps slightly harsh.
I'd also guess Kemba Walker didn't go from undrafted free agent to lottery pick on the basis of his postseason alone.

[Henry Sugar]

Craig Biggio being the second-best player of the 90s backs up the argument that statistical analysis of athletes can be very, very flawed.

MerrittsMustache is right and Bill James is wrong!

[g0lden3agle]

It's also a lot easier to use advanced metrics in a sport like Baseball than it is in Basketball.  Baseball is largely a combination of individual performances, while there is way more to a basketball game in terms of the cohesive flow of two 5 man units performing in tandem.  In my opinion basketball players need to pass an experienced eye test before metrics matter.  There's a certain intangible element to basketball that simply cannot be captured in metrics.