Assessing Tyshawn Taylor's All-American Candidacy

by fetch9 on Jan 26, 2012 9:00 AM CST in Kansas Jayhawk Basketball 2011-12

If time travel was a thing, people reading Rock Chalk Talk a year ago would be shocked to see the title of this post. But despite his reputation as a careless, problematic player, there is no doubt that Tyshawn Taylor has been one of the best players on the team, in the conference, and perhaps even the country.

Because neither we as KU fans nor voters can be biased, for entirely different reasons, I instead present a blind item. Here are the lines of some players generally thought to be the best point guards in the country, along with Tyshawn Taylor.

A couple notes: this mostly leaves defense out. I debated using defensive win shares from BB Ref but I'm not really a fan of trying to boil down defensive contributions to one number. (I think Aaron Craft is the best defensive PG in the country and Kendal Marshall the worst among the elite offensive ones. I would put Tyshawn close to the top, so take that for what it's worth.) Lastly, I would personally look at the numbers that are components of ORtg and PER, but that is mostly your discretion.

Player	Min%	Usage rate	PER	ORtg	eFG	Assist%	TO%	assist - TO%
A	78	15.9	14.8	106	47.9	42.5	30.8	11.7
B	73.8	17.7	18.3	108.4	50	27.2	22.7	4.5
C	57	22.8	22.1	111.4	56.9	37.5	22.7	14.8
D	80.1	28.5	20.1	105.1	54.4	30.7	23.8	6.9
E	76.8	22	19.9	110.9	44.8	31.7	19.2	12.5
F	68.2	26	24	113.6	61.9	40.2	29.7	10.5
G	88.2	25.2	21.3	112.4	46	27.6	11.4	16.2
H	86.1	23.6	22.7	118.3	53.4	45.8	23	22.8
I	83.6	31	27.3	115.4	47.4	35.3	11.3	24

First off, let's cut anyone with an eFG under 50. That means good bye to A, E, G and I. Next, bye bye to C. Someone who plays in only 57% of his team's minutes either fouls too much for my liking or isn't probably as good as we think he is. It is sort of punishing a team for having depth, but I think the best PG in the country would be one who is relied upon a lot more, no?

With those caveats in place, I think F, H and D are probably your best bets, in that order. Agree?

Let's see the answer key

A: Kendall Marshall, UNC

B. Aaron Craft, Ohio State

C: Scoop Jardine, Syracuse

D: Tyshawn Taylor, Kansas

E: Phil Pressey, Missouri

F: Pierre Jackson, Baylor

G: Jordan Taylor, Wisconsin

H: Scott Machado, Iona

I: Nate Wolters, South Dakota State

Right now it looks like Jackson is the runaway first teamer to me. He's carried a pretty big load of the offense, and oddly enough has gotten a lot better in terms of assist to turnovers lately, which corresponds to Baylor's dip in performance as a whole. But it's not his fault, as he's shooting almost 50% from three in addition to his ability to get the ball to his teammates.

But if I may stump for Tyshawn quick: Tyshawn has basically Thomas Robinson to pass to in terms of assists, where as Pierre has four or five shooters who are more efficient than anything Kansas has to offer, whether it be Quincy Acy or Perry Jones inside or Brady Heslip shooting threes. Does Tyshawn's edge in playing time and larger role in the offense, plus the worse talent around him, outrank Jackson's more efficient play? Maybe, maybe not. But I think it is pretty easy to argue that he's at worst the 2nd team PG on the All Big 12 team and a serious candidate for All-American honors.

You know, if voters don't go on twitter that much.

Tweet Comment 25 comments | 0 recs |

Do you like this story?

Comments

FWIW, in the blind test, I had Machado first by a decent margin over Jackson

I had trouble picking out a 3rd and eventually gave up.

by jayhawk3r on Jan 26, 2012 9:26 AM CST reply actions

Yep... on numbers alone, Machado is an easy choice

by hiphopopotamus on Jan 26, 2012 9:46 AM CST up reply actions

but he's played against the 187th worst group of opponents in the nation

Of course he can have good numbers, but to be the best you have to play the best, and Iona has not – at all. No games against top 25 teams, and only 1 top 50 team. Sorry, but minor league players don’t win major league awards.

by dagger108 on Jan 26, 2012 11:33 PM CST up reply actions

Tyshawn looks like Cookie Miller in that picture

What a B.A.

by 2.1 seconds left on Jan 26, 2012 9:57 AM CST reply actions

Is there any place to get season-long plus/minus?

I would think that would tell you more about a point guard than any other position on the floor.

by KSinDC on Jan 26, 2012 11:04 AM CST reply actions

Eh, not a fan of it at all in college.

Here

Glad I came, just wish I hadn't stayed so long.
Rock Chalk Talk

by Warden11 on Jan 26, 2012 11:31 AM CST up reply actions

Why are you not a fan of it for college?

by KSinDC on Jan 26, 2012 12:15 PM CST up reply actions

The variablity is so high as to make the stat meaningless.

Nice article by ken pomeroy

I always drink the Kool-Aid
Going for #8 in a row!

by bt01 on Jan 26, 2012 3:02 PM CST up reply actions

fail

forgot the link

I always drink the Kool-Aid
Going for #8 in a row!

by bt01 on Jan 26, 2012 3:03 PM CST up reply actions

I never understood that article

It seems to cast just as much doubt on margin of victory / points per possession (PPP) as it does on plus/minus but PPP is the whole basis of Pomeroy’s ranking system. But I don’t understand why variability in PPP is something we can work with but variability in plus/minus is a dealbreaker.

by KSinDC on Jan 26, 2012 5:49 PM CST up reply actions

Thanks for the link

Statsheet lacking the play by play for the KU-Baylor game makes an otherwise worthwhile site sort of useless for these purposes. I may see what they have for the PGs aside from Pierre Jackson and Tyshawn Taylor.

by KSinDC on Jan 26, 2012 12:38 PM CST up reply actions

Even if plus/minus is rather flawed (and it is), it's still better than nothing

Basketball statistics capture so little of the game and of the value that a player can offer. They capture almost nothing on defense (charges drawn, shots blocked and steals) and aren’t even very good on offense outside of scorers. Plus/minus gives us some sense of the whole. Even if there’s too much variability and not enough data, it still seems like it’s worth taking a hard look at.

by KSinDC on Jan 26, 2012 5:57 PM CST up reply actions

I think you can take a hard look at it and maybe that's why I'm not a fan.

I think to draw any meaningful conclusions, you almost have to have to video matching up with the +/- to make a determination about why points were or weren’t scored. Just a very simple hypothetical, Taylor comes out for one minute but Teahan hits to threes in that time. Taylor reenters and the opponent hits two threes on their first two possessions. Four possessions that have nothing to do with Taylor but to +/-, it looks really bad for him because the Jayhawks scored 6 quickly when he went out and as soon as he came back in they give up 6.

Glad I came, just wish I hadn't stayed so long.
Rock Chalk Talk

by Warden11 on Jan 27, 2012 8:38 AM CST up reply actions

Wouldn't we assume that most of those unrelated performance shifts will cancel out in the long run

Again, it seems similar to points per possession (which can be unfairly influenced by half court shots or depleted defense when the walk-ons are in or various other factors). There’s certainly plenty of noise in the signal, but there’s still a signal there.

by KSinDC on Jan 27, 2012 11:37 AM CST up reply actions

My thoughts about raw +/-

You’re looking at slices of playing time so there is more variability than in total team score.
To say a team is 2 points better than another team is to say they are very close. Yet when you ascribe that kind of margin to one player it seems like a lot. They are on different scales, which is why I’m okay with point variability for team efficiency scores (and their predictive power is demonstrable anyway), but don’t like it for individual players.
People claim it accounts more for the ‘total picture’, yet I think it narrows it a lot by ignoring teamwork.
There’s a perception thing where a player with a low positive seems okay and a player with a small negative triggers “OMG GET THIS BUM OFF THE COURT” feelings. Even if they’re within a range of error of each other.
It’s trivial to think of situations where +/- is giving you totally the wrong idea.

I think +/- numbers can be useful but they have to be cooked a lot.

In the NBA they play more games and play more individual ball. The players stick around longer, face more regular competition, don’t commonly go through big development shifts, have weaker home/away swings…. Plus there’s more motivation to find individual player value because of trades. So I think they’re good for the NBA but they seem a lot less useful in college ball.

by sax solo on Jan 27, 2012 6:26 PM CST up reply actions

Interesting thoughts

Thanks for laying them out.

I certainly agree that +/- per minute or per possession would be more useful than raw +/-.

by KSinDC on Jan 28, 2012 12:41 AM CST up reply actions

Is T2 having a stroke in that picture?

Only half his face works…

The name of that song played at KC Comets games in the '80's was Giorgio Moroder's "The Chase." Now you know.

by labbadabba on Jan 26, 2012 1:08 PM CST reply actions

The bottom half?

by KSinDC on Jan 26, 2012 1:35 PM CST up reply actions

very easy to pick out tyshawn

worst assist/TO numbers in the group

Not everything that counts can be counted, and not everything that can be counted counts."

--Albert Einstein

by Home Run Tony Cogan on Jan 26, 2012 10:04 PM CST reply actions

Well, second worst.

But he also has the second highest TO%, the lowest O Rating, and the third lowest PER. If he does make it as an All-American, it won’t be because of his numbers, IMO.

Gotta say though, Fetch, I really enjoy your work on here. I get excited every time I see one of your new posts come up.

R.I.P. DOM

by KennyGregoryRockThaCradle on Jan 27, 2012 12:30 AM CST up reply actions