Point Guard Rankings & Rookies: What a Difference a Weekend Makes...

On Friday, I wrote a post singing the praises of Briann January with the intent to later post statistical rankings of her rookie point guard counterparts relative to the rest of the league.

I then set out to crunch the numbers and reflect on the best means with which to evaluate point guards. And those numbers were supposed to support my point...

As of Friday afternoon, Renee Montgomery and Briann January were neck and neck statistically, with Kristi Toliver ranking 24th in the league among my chosen point guards.

And then more games were played.

Hours after my Friday post, Indiana Fever coach Linn Dunn nailed January with a DNP-CD. Renee Montgomery struggled and slipped significantly statistically. And Kristi Toliver put together two consecutive solid games, despite the Sky losing.

In other words, everything I was going to write was rendered worthless within hours.

This of course is the fun of following sports and trying to apply statistical analysis before teams have even played five games.

I put all of that out there to make an obvious point about these statistical rankings: I fully acknowledge that they’re not only imperfect but also subject to change, either because of a coach’s change in strategy, a significant injury, or natural player development. All of that is exacerbated when looking at rookies who are constantly growing as players, figuring out their role on the team, and responding to the team’s demands of them.

It’s messy. And that makes it fun.

These numbers are not meant to be the final word on the hierarchy of WNBA point guards; instead, I would hope that they serve as a conversation piece. I personally love them exactly because of how my initial thoughts were ruined this season: they allow me to track trends in players across the league and hopefully help me see things I might have missed through standard observation.

To be clear, it would be silly to claim that statistics tell the full story of basketball. However, I find it equally silly for people to pretend that their own observations are perfect. The numbers complement what we see and what we see can complement the numbers. Whichever way you approach a sport, the more information we have, the better we are able to understand what we are looking at…and hopefully form grounded, substantive opinions.

With that in mind, the main purpose of this post is to present an analysis of the first three point guards drafted in the first round – Kristi Toliver, Renee Montgomery, and Briann January. At the end, I’ll look at their performance relative to the rest of the league’s point guards.

Ultimately, I think I can say this about these three right now: judging who’s best among these three point guards is more a matter of taste than any sort of objective standard. And in the end, the goodness of fit with their team might matter more than the isolated skills they bring to the court.

A refresher on the Rethinking Basketball point guard statistics…

(Feel free to skip down to the section titled "Who’s the WNBA’s best point guard after two weeks?" if you hated high school math as much as I did and have a PTSD reaction anytime you see numbers in sequence beyond your bank account or a digital clock. Near the end of the post is a narrative summary of the rookie point guards and the top 10 WNBA point guards overall)

Last season, I used the following statistics to evaluate point guards:

Pure point rating
Net plus/minus rating
Points per zero point possession
Hollinger assist ratio
Usage rate
Boxscores
True shooting percentage

I generally like the results – by the last rankings I posted last season (click here to see those), Lindsay Whalen and Sue Bird were on top…and you’d have a really difficult time convincing me there is something wrong with that outcome.

Nevertheless, I am going to continue to tweak things, with the intention of making the outcome even stronger.

As it happens, the NBA draft is point guard heavy this year so there have been a few stories floating around the web recently about point guard analysis. Draftexpress, ESPN’s TrueHoop (observation, not statistical), HoopsAnalyst, and the Wages of Wins have all had in depth analyses of NBA point guards that I took a look at before launching into my own analysis of WNBA point guards.

First, the Arbitrarian blog no longer posts its “Boxscores” so I am not going to use that (sorry, just extra numbers for me to crunch right now). So for now, I will go back to Dave Berri’s Win Score, simply because it’s easy to calculate (petrel’s description from last year is pretty solid). I will think further on this for future iterations of these point guard rankings.

Update: I settled on Tendex ratings for now. More on that here:
http://rethinkbball.blogspot.com/2009/06/point-guard-rankings-update-surprises.html

Second, after reading through all the other articles, I realized that my framework did not really have any way of accounting for a point guard’s ability to break down the defense and get to the basket, whether to score or set up a play for someone else. The importance of this was especially evident in TrueHoops account of Minnesota’s recent point guard prospect workout – the ability to beat a defender and score either on pull-up jumpers or getting to the rim. It says something about quickness, ball handling, and a player’s instincts.

And when you think about it, if a point guard is going to have the ball in their hands at the end of a game, it would be a huge asset to be able to beat a defender and set up easy shots for their team. That was exactly the problem I described when I watched Sacramento playing Minnesota the other day.

So I spent some time thinking about how best to capture that.

Draftexpress suggests that free throw attempts per possession are a pretty good way to assess a player’s aggressiveness driving the lane. And while that’s true, it doesn’t seem to practically capture what one might want. At the end of a game, for example, a player might simply pick up free throw attempts in the bonus as a team fouls to stop the clock. A player could get fouled on a three. And really, if a player goes to the basket for a layup and gets fouled, getting one free throw for being able to complete the play is much better than getting two for not completing the play.

But HoopsAnalyst has used a different metric for years that I have ignored until now: 2 point field goal percentage. Again, when you think about how point guards might score, a more effective way to assess their ability to get to the basket might be their ability to complete two point plays. Point guards are unlikely to be scoring two point field goals on post ups or set shots because, for the most part, they have the ball in their hands making plays.

TrueHoop nicely characterized some of the ways a point guard might score two point field goals:

Anyway, Tyreke was dominant.

Refusing to simply muscle his way to the hoop (which he did handily a few times), he scored in a half a dozen different ways. Pull-ups on the wing, runners, balanced fade-aways, and a particularly spectacular jab-step and crossover that sent Flynn (without question the best defender of the group) reeling. His jumper needs work but I wouldn't call it a weakness, and he'll definitely hit enough to force defenders to respect it.

Lawson and Flynn played about how I expected them to. Controlled the ball well, took high percentage shots and occasionally switched gears to get to the rim.

Given how an elite point guard is likely to get two point field goals – pull-ups, runners, fade-aways, driving lay ups, changing pace in space – knowing that they are able to complete those plays tells me a lot more about their ability to attack the basket and complete plays than free throw attempts. And if they are able to beat a defender in the variety of ways described on TrueHoop, they will draw a defense in ways that enable them to distribute more effectively as well. So ultimately, I went with 2 point percentage.

Third, after thinking conceptually about what I wanted to see from a point guard, I also decided to drop usage percentage and add turnover percentage. The decision to add turnover percentage was simple – turnovers are bad. A point guard who makes a lot of them is not likely to be elite (yes, even though great point guards make a lot of turnovers, their turnover percentage should be low).

Usage percentage is just more of a descriptive statistic than one used to evaluate anything. Knowing that a player is individually responsible for a team’s plays does not help me assess their quality. For example, some teams might be so loaded with scorers at other positions that a point guard using up a whole bunch of plays would be counter productive. Conversely, some teams have Olympians or all-world point guards who might best benefit their team by using a lot of plays. Furthermore, a point guard who can’t shoot or turns the ball over a whole lot is best served by giving it to someone else. A good shooter who is not shooting a lot might hurt their team.

In other words, there’s no way to say whether a player’s usage percentage is good or bad without taking account of a whole lot of other variables. It’s not really something I can individually rank a player on.

Since this may all seem a little unwieldy, I’ve broken down these statistics into three categories: facilitation ability, scoring ability, and game impact. And came up with some interesting results.

Which rookie point guard is best so far?

(Note: Most stats are through 6/20/09, plus/minus stats though 6/21/09)

Facilitation ability

So fundamentally, a point guard needs to be able to bring the ball up the court and distribute it to others. Ideally, that player would be able to facilitate scoring opportunities for others. A first obvious question might be one that can be answered by looking at a player’s assist ratio:

How often does a player create an assist? (Hollinger Assist Ratio)

Here are the numbers:

Hollinger Assist Ratio

January, 29.26%, #7
Toliver, 21.95%, #14
Montgomery, 18.65, #20

Just for some perspective, if we drop two outliers at the top and bottom of the assist ratio rankings, a high assist ratio is 36.72 (Temeka Johnson) and a low assist ratio is 13.24 (Deanna Nolan).

The order of assist ratios among the rookie point guards is identical to the order of their assist averages, with January ranking just above average (23.34%) in assist ratio among the point guards on this list.

Neither Montgomery nor Toliver could be considered effective distributors based upon these numbers. And, if we ask a second question about a player's ability as a facilitator, that point becomes even more clear.

How efficiently does a player create scoring opportunities per others?

To answer that, let’s take a look at these players’ pure point rating:

Pure point rating:

January, 1.43, #11
Montgomery, -2.18, #19
Toliver, -7.29, #23

Again, for some perspective, the high pure point rating is 6.95 (Temeka Johnson) and the low is -7.98 (Shannon Johnson).

So what does this say about Toliver?

Although Toliver’s assist ratio tells us that she is creating an assist on almost one of every four plays that she makes, she has not been particularly efficient at creating those opportunities.

More specifically, if we consider the fact that turnovers are statistically more harmful than assists are helpful, the fact that Toliver has more turnovers than assists on the season means that she is not making very good decisions in trying to distribute the ball.

More generally, whereas we can use assist ratio to describe what a player does with the ball, we can use pure point rating to evaluate how well a player is distributing the ball.

Unfortunately, none of these three point guards does very well when it comes to turnover percentage all ranking in the bottom five of this list of point guards.

Turnover percentage:

January, 16.72%, #20
Montgomery, 16.96%, #21
Toliver, 21.95%, #24

Combined with the pure point rating and assist ratio, what this tells us is that January is clearly doing a better job than Montgomery at picking up assists despite committing turnovers at about the same rate. That Toliver is committing a turnover so often is what accounts for her low pure point rating.

To say that Toliver has been ineffective as a distributor is therefore an understatement. And this supports a point I’ve made previously – players with low pure point ratings in college, typically do not make a big jump in the pros.

Scoring Ability

However, Toliver’s strength in college was her scoring ability and one might hope that she has been a more efficient scorer than distributor in the pros.

And judging from Montgomery’s distributing stats, one would hope that she’s making up for it with point production.

Having examined how efficiently these point guards distribute the ball, we might now want to know how efficiently a point guard is as a scorer. Theoretically, a poor distributing point guard could make up for that by being an efficient scorer.

A first question I might ask is as follows: how well does a player shoot the ball when adjusting for free throws and the increased difficulty of three pointers into account?

For that, I use true shooting percentage and to this point, Toliver has not done well in that regard:

True Shooting %:

Toliver, 57.32%, #5
Montgomery, 51.36%, #8
January, 44.50%, #20

While Montgomery has been an average distributor, she’s been a rather solid shooter based on her true shooting percentage. The opposite is true for January – she’s been an effective distributor, but not so effective as a shooter. And that’s expected – she was not known for her shooting ability entering the league.

Although Toliver is still a rather inconsistent shooter, she’s demonstrated against the Sun on Friday and Mystics on Saturday that she can shoot the ball.

However, in addition to knowing that she can shoot the ball well, we might also want to know how efficient she is in scoring for her team compared to wasting possessions.

That leads to a second question, answered by the statistic points per zero point possession:

How often is a player individually responsible for scoring possessions compared to non-scoring possessions?

This can be used as a proxy for scorer decision making – how well are they balancing scoring possessions with non-scoring possessions? – but also as a way to think about how well your point guard is managing possessions.

So for example, if a player does create a lot of turnovers or end possessions without scores, this stat will help us see if they are making up for that with points, rather than assists.

Pts/Zero Pt. Pos:

Montgomery, 1.55, #12
Toliver, 1.51, #14
January, 1.21, #19

So first of all, it’s worth noting that all three of these players is below average (1.64) in this stat among the 25 point guards I selected for these rankings. But although none of them are particularly efficient scoring point guards, both Toliver and Montgomery are close to average and shooting a high enough percentage that they could probably establish themselves as efficient scorers in the league.

However, a third question we might ask about scoring ability is: how well does the player score inside the three point line?

Where Montgomery really separates herself from Toliver and January in terms of scoring efficiency is 2 point field goal percentage.

2 Point%:

Montgomery, 50%, #6
Toliver, 38%, #13
January, 36%, #17

While Montgomery may not be very efficient as a distributor, she is finding ways to get herself high percentage shots, and that bodes well for her future as a point guard.

This would suggest that Toliver is doing the majority of her scoring from low percentage outside shots, which might end up being ok given her true shooting percentage. But the low 2 point percentage seems to suggest that she is not as good at creating high percentage shots inside the three point line, which is pretty much consistent with what I’ve seen of her.

Toliver spends a lot of time with the ball at the top of the key looking for her own shot without really being able to beat her defender. That not only hurts her scoring ability, but also her ability to make plays for others.

Chicago Sky coach Steven Key addressed on possible cause for this problem in a recent Washington Post article:

One of the of knocks on her before we drafted her was that she's not physically as strong as a lot of other people. She's not as big as everybody else. But she also has a larger skill set than most people do. We've been working with her on that, trying to make that adjustment, until she can get a little stronger, until she can gauge how deep she can go into lane before she can still get her shot off, how much body contact can she take and not be off-balance. Until she figures that out, we're helping her get around it the best she can.

Perhaps this explains her low 2 point% -- a tentativeness about scoring in the lane.

So we now see that Montgomery is the better scorer, January the better distributor, and Toliver struggling at both so far. So the next thing to examine is how good a playmaker the player is.

The difference between these two statistics and the facilitator metrics above is that I am now looking at quantifying “intangibles” by using these metrics as proxies – ball dominance, a team’s reliance on a player, and aggressiveness in driving to the basket.

To put a player’s performance in perspective, it is sometimes helpful to look at their usage percentage -- the percentage of a team’s plays that a player is individually responsible for while on the court. And the outcome for Toliver is somewhat alarming given what we found out above.

Usage %

Toliver, 35.83%, #1
Montgomery, 17.57%, #13
January, 16.93%, #14

Toliver is using up more plays than any starting point guard in the league....and yet, she’s a below average shooter and distributor. Which means Toliver has essentially been a ball stopper to this point in her career – she is getting the ball and individually responsible for burning plays.

Just to put it in perspective, the three players closest to Toliver’s usage rate are Nolan (30.51), Becky Hammon (30.66), and Diana Taurasi (27.38)…but these are all-stars whose teams are dependent upon them to make plays. It’s quite perplexing that Toliver is using a larger percentage of plays on the court than any of those three, especially when considering that the Sky should be looking inside to Sylvia Fowles and Candice Dupree.

(Note: Since this is a statistic that does not stand very well on its own without some indicator of a player’s efficiency with the ball, I am no longer including it in the rankings. But it’s still interesting to look at when judging the quality of a point guard)

Game impact

Lastly, given that a point guard likely has the ball in their hands a lot, you would hope that they have a positive impact on the game. For that, I use Win Score and plus/minus.

Here are those numbers:

Win Score:

January, -.39, 15
Montgomery, -.65
Toliver, -.83

Plus/Minus:

January, +5.5, #10
Montgomery, -2.4, #15
Toliver, -8.5, #18

The plus/minus numbers are more clear – January has had a relatively positive impact on her team when on the floor whereas Montgomery and Toliver are not contributing as much as one might like.

Win Score does not look favorably upon any of these players.

Thus far, this confirms what one might essentially infer from the previous statistics: Toliver is struggling to contribute positively out of the gate.

Who’s the WNBA’s best point guard after two weeks?

When compared to other point guards in the league, Montgomery and January have been about average, while Toliver has been struggling to make a case as a potential starter one day.

Every starting point guard was included in these rankings in addition to a few players who often assume point guard roles on their teams (Hammon, Nolan, Taurasi, Candice Wiggins).

(Note: for the complete rankings and numbers of the top point guards so far this season, click here.)

Briefly, the following players from last year’s rankings were left out because they either have not played significant minutes or are no longer with a team:

Nikki Blue
K.B. Sharp
Tan White
Ivory Latta
Jia Perkins (I don’t think she’ll be running point much this season with Toliver around)
Katie Smith (does not do well in these rankings)
Shannon Bobbitt (limited minutes this year, but I like her!)

In their place are the following:

Our three rookie protagonists
Shavonte Zellous
Kara Lawson
Nikki Teasley
Kristi Harrower

I just did a ranking of 1-25 in each of 8 categories from above (excluding usage rate, which is more of a descriptive stat than evaluative).

One quick caveat: looking at last year’s statistics, using Boxscores along with these stats gives a much more reasonable result than using Win Score. At the moment however, I don’t have the time to come up with Boxscores without the help of the Arbitrarian…I put in an email to him…hopefully he gets back to me. So take these with a grain of salt knowing that the results will be different once I get that additional stat.

Here are the top 10 WNBA point guards overall as of 6/21/08 (total points in parentheses):

1. Temeka Johnson (174)
2. Kiesha Brown (169)
3. Kristi Harrower (153)
4. Sue Bird (148)
5. Lindsay Whalen (143)
5. Diana Taurasi (143)
7. Nikki Teasley (138)
8. Tully Bevilaqua (136)
9. Kara Lawson (127)
10. Noelle Quinn (117)

14. Renee Montgomery (93)
15. Briann January (91)
19. Kristi Toliver (80)

For rankings as of 6/22/09, click here.

What we see is that all three of these rookie point guards are knocking on the door of “average”. None of them is really among the top tier guards right now.

I keep saying “right now” because these statistics are based on a very small sample – you have to imagine that by the end of the season, players like Bird and Whalen will move up while Kristi Harrower and Kiesha Brown will fall back down to earth after solid statistical starts. And Boxscores – a metric that takes account of the player’s contribution to a team’s wins – will help that as well.

But what’s interesting is looking at the point guard styles of our three rookies of interest:

January is looking like she will end up being more of a “distributor” point guard, with a below average usage percentage, an above average assist ratio, and an above average pure point rating. That is great for a team that has scorers to pass to. She might be the closest thing to a “pure facilitator” among these three point guards, but will have to cut down on the turnovers.

Montgomery is looking like she could either turn into a scoring point guard, but I think it’s also obvious that she has all the tools to become an elite combo guard with the ability to penetrate the lane and score or distribute.

Toliver is hard to peg, but right now her only real strength as a pro is three point shooting. She really has not demonstrated the ability to distribute the ball efficiently at this point, which is why I would not put her in the “combo guard” category.

Of course, you could say she’s not that much worse than Montgomery, but from watching the two play, Montgomery is able to get by her defender and make plays. That simply is not Toliver’s game at the professional level. Right now she is a turnover prone three point shooter who seems to be looking for her own shot more than that of others. That doesn’t bode well for her future, but hopefully she’ll get better over time.

So in the end, the "best" of these three will depend on what a team wants...at least partially.

Here are a few other notes about the rankings:

So I’m sure seeing Kiesha Brown as the #2 point guard raises red flags about this process. However, despite the caveat that this is a small sample size, Brown is actually playing extremely efficient basketball right now.

Brown is not exactly a big playmaker but she is doing a great job of protecting possessions – her turnover percentage (the percentage of plays in which she commits a turnover) is 6.23, the second lowest of any point guard in this group (Vickie Johnson has yet to commit a turnover this year). Brown is playing better basketball this season and the statistics show that.

During last week’s game against the sky, Mike Thibault said the following about Brown during a halftime interview:

“Well I think part of her problem was that people were always trying to get her to be a starter where she was…in LA they started her a lot of games. And what happens with a lot of teams is they go get a cheaper backup…and what you get is a situation where we don’t need her to be a starter now. We’re not looking for her to be a starter. She could be if we had an emergency. But we’re looking for a player that can play a couple of roles for us – at both guard spots. And so I think it makes a difference in how you fit in a team when you know what the expectations are. LA needed her to be the starting point guard..and we don’t need her to be that, we need her to be a complement to Lindsay Whalen. And this may be why she can settle in a little bit better.”

The funny thing is that the same could be said about Temeka Johnson’s shift from LA to Phoenix – her role and expectations got clearer and she performed better. Obviously, Thibault was not taking a cheap shot at LA’s system…but…

Speaking of which, Temeka Johnson has been phenomenal this season. She is near the top in almost every category. The only one she is below average in is usage rate…which actually makes her season thus far more impressive – she is distributing the ball efficiently, and taking good shots while also not dominating the ball on a team with Taurasi and Cappie Pondexter. To this point, she has been the perfect fit for the Mercury...

Kristi Harrower #3? Really?

Let me just give you her numbers to this point:

PPR: 6.85 (#3)
Plus/Minus: +12.1 (#7)
True shooting %: 47.79 (#14)
Points per zero point possession: 1.98 (#6)
Turnover%: 5.40 (#1)
Assist ratio: 32.42% (#3)
2 point%: 47.3% (#7)
Usage%: 11.52 (#22)

Harrower is not a superstar point guard. But as of right now, she’s just not making many mistakes and getting the ball into the hands of her superstar teammates. She is not dominating the ball at all as evidenced by her low usage rate and that’s probably a good thing.

Would it be nice for the Sparks to have a more dynamic point guard? Sure. But Harrower is playing some of the most efficient basketball in the league right now, which is something that could not be said for any one of the Sparks’ point guards last year...

Sue Bird has not been scoring very well to start the season (until yesterday), which is really hurting her ranking right now. It is likely that once she starts shooting more effectively she’ll rise in these rankings. But she's been on fire distributing the ball in the last three games after publicly lamenting a 7 turnover performance against the Lynx: 22 assists, 0 turnovers is not too bad...

Lindsey Harding is currently ranked #12 which might be surprising given her relatively hot start. But while she is doing a lot right, her shooting has been abysmal thus far, with a true shooting percentage of 34.58% and a 2 point% of 23.68%...

In watching Whalen in the game against the Sky, she is still doing all she can to make plays for the Sun. The problem is the Sun have not been making shots (until about Friday); in that game against the Sky, I counted at least five assists that Whalen lost simply because her teammates did not make shots after Whalen set them up nicely...

Wondering about my personal favorite from last year, Leilani Mitchell (#18)? Some folks…ahem…some haters... suggested last year that if given starter minutes, Mitchell’s production would go down. Thus far, that’s exactly what’s happened this season. I have not watched the Liberty enough yet this year to say exactly what’s going on, but I can say her shooting numbers are among the lowest in this group and her assist numbers are down considerably as well. Hopefully she can turn it around...

Transition Points:

I was tempted to follow Draftexpress’ lead and add steals per 40 minutes to the mix. While it’s not a perfect proxy for defense, it does demonstrate a measure of athleticism. Something to think about for the next iteration.

I have to wonder if Renee Montgomery will assume a bigger role on the Lynx with Augustus out for the season. The reason is that they will have to replace her scoring somehow or seriously change their offense. Montgomery has demonstrated the ability to score in spurts and they might need more of that now.

I got a look at Shalee Lehning for the first time yesterday. She is not necessarily a player who will “wow” you, but she makes good decisions with the ball and keeps the offense moving. She looks really comfortable within Atlanta’s offense. It was especially interesting comparing her to Leilani Mitchell because they both are very decisive young point guards. The big difference – and an important one – is that Mitchell is ultra-quick. It will be interesting to see how Lehning does as the season wears on.

Nikki Teasley also looked like a pretty good point guard in yesterday’s game against New York. She is extremely patient with the ball and just looks in complete control bringing the ball up the court. She’s decisive, but unlike Lehning or Mitchell she does not commit to a specific decision unless she’s sure it’s right. While Lehning and Mitchell spend a lot of time dribbling back and forth across the court surveying opportunities, Teasley just calmly surveys the situation and lets the game come to her. It’s impressive…and it’s showing up in the stats this year.

If only the Dream had played defense yesterday.

Two WNBA point guards have been signed to endorsement contracts with a new company that makes athletic shoes designed specifically for female athletes, Atlanta-based Nfinity. Will be interesting to watch for any of those promotions.

The Chicago Sky did not look very good against the Sun on Friday…and then dropped Saturday’s game against the Mystics who were also coming off a back to back. I am still a little confused about what they are attempting to do on offense. Hopefully they work that out as the season goes on.