The Most Outstanding Rookies: How do we compare rookies…fairly?

. Friday, August 1, 2008
  • Agregar a Technorati
  • Agregar a Del.icio.us
  • Agregar a DiggIt!
  • Agregar a Yahoo!
  • Agregar a Google
  • Agregar a Meneame
  • Agregar a Furl
  • Agregar a Reddit
  • Agregar a Magnolia
  • Agregar a Blinklist
  • Agregar a Blogmarks

Yesterday I looked at rookies from the perspective of potential instead of production because ultimately, the best players will exceed their first year production.

However, in the comments from Monday’s post about Crystal Kelly, Kbailey3131 asked a great question that I think begs a more balanced approach to evaluating rookies: how can we compare rookies with different strengths knowing that they will develop in the future?

A recent discussion Monarchs fan had over on their fan board had to do with comparing Crystal to her teammate Laura Harper asking if Kelly is "better" than Harper. I don't think you can adequately compare the two right now because they are different types of players. How does Harper rank and how would you go about statistically comparing the two in a way that measures what they each do well and make projections long term?
It’s a difficult question, but one worth tackling because that is essentially the comparison made to determine the rookie of the year. And I think Harper and Kelly are a good starting point...and I'll look forward to input on some other good pairs.

Honestly, my instinct is always to favor the player who just seems to be involved in more positive plays when in the game – that would be Crystal Kelly, who is primarily a scorer, whereas Harper is a rebounder that stands out less. That tendency highlights not only the problem of comparing across player style that KBailey points out, but also the tendency to favor players who grab our attention by scoring.

The WNBA.com rookie rankings very much favor scoring if you look at them closely – their rankings are essentially the rookie scoring leaders adjusted slightly to account for team winning percentage (and Fowles’ injury). While the results seem to match common sense -- Candace Parker is consistently #1 – it makes it difficult to fully appreciate second or third tier players who might influence the game in more subtle ways. And that's what I'll set out to do here...

*** Long post warning: If you would like to cut to the chase and see the final rankings and/or the verdict on Kelly vs. Harper, scroll down to the section titled "Overall Most Outstanding Rookie(s)" ***.


The method

As we know, there's more to basketball than scoring – ball movement, shooting, offensive rebounding, and turnovers are all key elements. So perhaps one way to figure out which rookie is best is to compare their production in those important elements of the game. Evaluating rookies using these elements seems to evaluate players' all around games without favoring any particular position – perimeter and interior players benefit differently from each area.

Another area of interest is defense, which can obviously influence an opponent’s ability to produce in any one of those key areas. It’s tough to make an objective assessment of a player’s defense without counting the actual number of times the player is responsible for stopping the opponent they’re guarding. But I think there is some work being done that is moving in that direction.

So with these principles in mind, let’s try to answer KBailey’s question by comparing Harper, Kelly…and A’quonesia Franklin for good measure. In the process, I will rank all the rookies rotisserie style by scoring them 1-25 in each category below. While #1 may not be surprising, some of the others might be.

Ball movement – how “unselfish” is she?

We’ll start with ball movement to give Franklin a head start. I’m using assist ratio – the percentage of a player’s plays that end in an assist – to account for a player’s contribution to their team’s assisted field goal percentage. Of course it is flawed in that it does not account for assists lost, but I think it’s a good enough metric.

Assist ratio would of course strongly favor point guards who are given ball handling responsibilities and tend to rack up more assists, but it’s also good for sorting out the passing skills of bigs like Harper and Kelly.

Here’s the top ten, plus our Monarchs of interest:

Assist Ratio
1. Mitchell, Leilani 38.66
2. Franklin, A'Quonesia 36.00
3. Bobbitt, Shannon 30.25
4. Swanier, Ketia 26.37
5. Hornbuckle, Alexis 20.17
6. White, Erica 19.87
7. Holt, Amber 19.10
8. Atunrase, Morenike 17.65
9. Gardin, Kerri 17.47
10. Wiggins, Candice 16.30
23. Harper, Laura 6.13
24. Kelly, Crystal 3.88

Well here’s one instance where Parker is actually out of the top 10 (she’s #12). No surprise that Mitchell is at the top here, especially since she has been so efficient with the ball during the month of July, posting an insane 7:1 assist to turnover ratio. Franklin is quietly right behind Mitchell although Mitchell plays more minutes and scores more meaning that her ratio of assists to everything else she does is even more impressive.

Regarding Harper and Kelly, they are both near the bottom of this list, but it’s interesting to see the distance between them. All this says to me is that when Harper gets the ball, she’s more likely to score – since the equation takes into account shots and free throws as well, Kelly’s proportion of assists relative to everything else is smaller, but perhaps not that much “worse” than Harper. So that’s pretty much a wash.

Turnover ratio: How well does she take care of the ball?

Obviously, turnovers are costly so a player’s ability to keep those down are critical to their individual and team success. This might seem to benefit guards who generally have better ball handling skills and get the ball less in traffic, but they also generally have the ball in their hands more. That’s why we look at the percentage of plays that end in a player turning the ball over. The top ten, plus Monarchs:

Turnover ratio
1. Wiggins, Candice 9.22
2. Anderson, Jolene 9.43
3. Anosike, Nicky 9.92
4. Holt, Amber 10.79
5. Carson, Essence 12.00
6. Ajavon, Matee 12.09
7. Parker, Candace 13.14
8. Humphrey, Tasha 13.35
9. Mitchell, Leilani 13.88
10. Young, Tamera 14.14
13. Kelly, Crystal 14.98
18. Franklin, A'Quonesia 18.00
20. Harper, Laura 20.07


A quick note on Wiggins here – I find it amazing that she turns the ball over so infrequently given the way she plays, which borders on reckless abandon at times. It’s a testament to her skill with the ball that she ranks so highly, even just among rookies.

We see some separation here between Kelly and Harper that I think is notable. When you consider how little Harper touches the ball and the fact that a lot of those touches likely come from rebounds, turning the ball over on one in every five plays is somewhat discouraging. Ditto for Franklin who is expected to take care of the ball. You have to imagine that this is why she plays so little…in addition to playing behind Ticha Penicheiro.

Offensive rebound rate: What percentage of available rebounds does she grab?

Offensive rebound percentage is one of Dean Oliver’s four factors because offensive rebounds extend offensive possessions, thus putting more pressure on the defense. So it’s something I would use in judging the quality of a player, especially forwards like Harper or Kelly. And this turns out to be one of Harper’s strengths. Here are the top 10 rookies, with Monarchs:

Off. Rebound Rate
1. Pringle, LaToya 15.33
2. Larkins, Erlana 14.86
3. Harper, Laura 13.40
4. Langhorne, Crystal 12.30
5. Gruda, Sandrine 10.86
6. Kelly, Crystal 9.73
7. Anosike, Nicky 9.52
8. Humphrey, Tasha 9.35
9. Fowles, Sylvia 9.05
10. Hornbuckle, Alexis 8.95
25. Franklin, A'Quonesia 1.81

It’s interesting to see Hornbuckle crack the top 10 here and I imagine it’s due to the number of hustle plays she makes just beating bigger players to the ball.

So this is where Harper shines – her size in the post allows her to get more offensive rebounds than Kelly while she’s on the court. And a difference of a few offensive rebounding percentage points could mean a difference of around 20 players, so it’s significant.

True shooting percentage: How good a shooter is she?

I also used this yesterday in my rookie potential rankings. I’m using it here because I think shooting efficiency is essential to being a productive player. It’s possible to be a productive player without shooting well, but it means one must be exceptional at something else. But here is the top ten and where the Monarchs fall:

TS%
1. Kelly, Crystal 63.26
2. Langhorne, Crystal 62.5
3. Humphrey, Tasha 59.84
4. Parker, Candace 56.84
5. Wiggins, Candice 56.62
6. Pringle, LaToya 54.84
7. Mitchell, Leilani 53.80
8. Harper, Laura 53.64
9. Larkins, Erlana 53.12
10. Anosike, Nicky 52.88
25. Franklin, A'Quonesia 30.24

The Monarchs know a thing or two about poor shooting point guards, but Franklin has yet to show the playmaking abilities of Penicheiro. The thing is, Franklin is an 80% free throw shooter so it’s her shooting from the field that’s dragging her down – 24% from the field and 23% from three point land. That’s not very promising for a turnover prone 5’4” guard.

Kelly’s free throw shooting is a really huge asset for her as it makes her an extremely effective post scorer – when she knows how to create space and hold a position and if she’s fouled she can make the defense pay from the line.

Valuable Contributions Rating: How much does she contribute to the team?

Valuable contributions rating is yet another David Sparks creation that provides us with an idea of an individual player’s ability to make contributions to the team, independent of team success. It’s a ratio of player production to team production and he describes the basis for the “production” aspect of this statistic at Hardwood Paroxysm.

I add it here as the best thing (readily) available (that I know of) to account for a WNBA player’s production, how they help their own team, and how well they disrupt their opponent’s production. I didn’t use it as a standalone here because the weights were set for the NBA and though they may be close to WNBA weights, I’m content to use it in conjunction with these other statistics for now. Here are the VCRs:

VCR
1. Parker, Candace 1.74
2. Wiggins, Candice 1.42
3. Kelly, Crystal 1.39
4. Anosike, Nicky 1.34
5. Langhorne, Crystal 1.34
6. Humphrey, Tasha 1.33
7. Fowles, Sylvia 1.26
8. Pringle, LaToya 1.2
9. Gruda, Sandrine 1.14
10. Hornbuckle, Alexis 1.05
13. Harper, Laura 0.95
23. Franklin, A'Quonesia 0.51

The league average VCR is .76 and VCRs of 2.0 or greater are almost unheard of in the NBA, according to Sparks. So that should give you a sense of the scale and how good this rookie class is. I think it underscores the point I made from the outset – scoring is not the only valuable element of basketball and we overvalue it far too often. Langhorne, Pringle, and Hornbuckle are definitely not big scorers.

As for Kelly vs. Harper, we see that VCR actually reinforces the idea that Kelly is the more productive overall player than Kelly, given that she makes more valuable contributions to the team. After this and the other numbers presented, it would be difficult to argue strongly that Harper is better than Kelly.

Overall Most Outstanding Rookie(s)

So, it should be obvious that the most outstanding rookie of 2008 is Candace Parker – subjectively or objectively, real or fantasy. But what this exercise allows me to do is also present a first and second WNBA All-Rookie team for your criticism (though I know the WNBA only does one All-Rookie team). I went straight by the numbers and there are some surprises. Here are the teams, with numbers:

First Team:
Name Total Rank
Parker, Candace 95
Wiggins, Candice 94
Anosike, Nicky 91
Humphrey, Tasha 91
Kelly, Crystal 83

Second Team:
Mitchell, Leilani 80
Langhorne, Crystal 79
Pringle, LaToya 77
Hornbuckle, Alexis 75
Holt, Amber 74

12. Harper, Laura 63
24. Franklin, A'Quonesia 37


If I were voting, I’d move Mitchell to the first team over Kelly since they’re reasonably close…and I just like what Mitchell has done in her first season. And those two players represent just two of multiple differences between the WNBA.com rankings and these -- Mitchell, Langhorne, Pringle, and Hornbuckle are all players who have been very productive without being big scorers.

Fowles probably didn’t make it at this point because she’s been injured which has hurt her statistics, but subjectively, she is undoubtedly one of the top five rookies of this class. It’s conceivable that she could even accumulate stats to play her way onto the first team with some good games after the Olympic break, but she will likely get an honorable mention instead.

Some people may dispute Humphrey being on the first team in favor of players like Ajavon, Young, Houston, or Carson. But I can’t think of a convincing argument for anyone beneath Humphrey. There are not many rookies able to contribute to a team in as many ways as Humphrey on offense and she has excellent basketball awareness. Her defense is coming along as well and she’ll almost certainly develop as a post defender playing for Laimbeer and Mahorn. I also think it speaks volumes about Humphrey’s talent that she has earned a starting role as part of one of the deepest frontcourts in the league.

Instincts count

People might also disagree with Kelly being the fifth rated rookie, but you know how people talk about great players having that “It” factor? I think it’s hard to find a rookie with better instincts than Kelly’s, especially around the basket. In fact, I think she’s one of those players where her mechanics are behind her instincts – with practice and repetition she should get even better.

When comparing the Monarchs’ rookies, I think that’s what it comes down to – Kelly’s instincts put her above Harper right now and her skill set makes her more productive than Franklin. What makes her a great scorer is her ability to initiate contact and hit free throws despite occasionally giving up an inch or two to her opponents. Overall, Kelly looks more decisive and fluid with the ball and that should pay off in the future.

Defensively, neither is great, but Harper’s ability to extend possessions with offensive rebounds is valuable. However, Kelly’s quick hands and anticipation allow her to pick up steals and that will definitely benefit her. Harper also has a nagging tendency to pick up fouls and I think that’s because she looks a little bit more mechanical inside than Kelly.

But to answer the question about balance, right now, there’s no contest: Kelly is the more productive and more promising player statistically. Yesterday’s look at ts% and plus/minus rating demonstrate that Kelly seems to know how to pick her spots and make an impact on the court efficiently. Her diamond rating further reinforced the notion that she has plenty of room for growth, possibly the most of any rookie.

In terms of first year production, whereas Harper could become a dominant rebounder which will extend possessions for her team, Kelly is more efficient with the ball (higher ts%, lower turnover %) and she makes a much stronger overall contribution to the team (VCR).

So to return to the original point in KBailey’s comment, it might be possible to compare these two players and determine that Kelly is better. However, Monarchs fans probably shouldn’t worry about that – they have quite a dynamic frontcourt duo in the fold that complement each other very well in the front court, in fact, they fill in each other’s blanks in many ways. They could be the core of a very bright future in Sacramento.

Transition Points:

What are your thoughts? Who would you put on the WNBA All-Rookie team and why? Which of these choices do you disagree with most strongly? I'll look forward to the comments...

I interpret the Rookie of the Year
as rewarding the “most outstanding” or “most productive” rookie, independent of their team’s winning percentage. The reason I like to think of it that way is that 1) rookies are inconsistent so they might lose as many games as they win in some cases and 2) the best ones are usually drafted to bad teams so they don’t win. Since Harper and Kelly are on the same team anyway, it’s a nice mini-case to use in figuring out how to compare rookies when winning is equal.

I was looking for some background info on Laura Harper and this page came up. I’m wondering it’s the discussion KBailey was referring to: http://www.kingsfans.com/forums/showthread.php?p=558024

Although Kelly has the better plus/minus for the season, Harper won the plus/minus battle in their loss Sunday against the Storm, which I watched: Harper +11, Kelly -8.

Matt at the Connecticut Sun blog takes a nice look at Jolene Anderson
...who was tied for 20th in these rankings and tied for last in yesterday's rookie potential rankings...with A'Quonesia Franklin...

Found this rap about Candace Parker and posted it in my video of the day box (top right corner of the page). Better than one might expect from YouTube.