Wednesday, July 30, 2008

Monta on my mind

Part I - Improvement

This fall, Monta Ellis will turn 23 years old and, days later, begin his fourth NBA season. He just signed a 6 year, $67 million deal to stay with the team that drafted him in the second round four years ago, the Golden State Warriors.


Besides the ability he's shown in his three years as a pro, it's Monta's potential, evidenced by his improvement throughout his career, that earned him his contract. After all, he'll be making the same amount of money as Chauncey Billups next year, and more than Manu Ginobili -- even though it's unlikely that he'll produce as much as either of those players. At just 23 years old, though, it's reasonable to expect that his production will increase through the length of his contract.

In the following look at Monta Ellis' improvement, I'm going to look at three different shooting percentage numbers. The first is FG%, which is the normal Field Goal percentage that we're used to seeing in box scores and such. The next is eFG%, or "Effective Field Goal Percentage," which accounts for the fact that 3-pointers are worth an extra point. For instance, a player who makes 4 three-pointers in 10 attempts is similar in efficiency to a player who makes 6 layups in 10 attempts -- they both score 12 points while taking 10 shots. While traditional FG% says the first player shot 40% and the latter shot 60%, eFG% says they both shot 60% eFG. Next, True Shooting Percentage takes into account not only three-pointers, but also free throw attempts. Imagine a player who takes 3 shots and misses all of them, and also is fouled in the process of shooting 5 other shots, and makes all 10 resulting free throws. Traditional FG% and eFG% say the player shot 0%, which is true, but the times that the player went to the foul line were also scoring attempts (they would have been shots if he hadn't been fouled), so TS% looks at the points created there, also. The TS% formula is slightly more complicated than this, but for our purposes you can imagine that in the above example, the player scored 10 points in 8 scoring attempts (3 shot attempts plus 5 shooting fouls), giving him a TS% of 62.5%. Players such as Dwyane Wade and Corey Maggette have learned how to absorb contact and get to the foul line, and that skill has greatly increased their efficiency (since most players shoot a much higher percentage from the free throw line than they do from the field).

Ok, on to the analysis:

Monta Ellis' FG%, eFG%, and TS%, respectively, in each of the last three years:

2005-06: 41.5 45.9 48.6
2006-07: 47.5 49.5 54.5
2007-08: 53.1 53.6 58.0

Notice a pattern? There are notable reasons for his improvement. For one, in each year he has decreased his number of 3-point attempts, which is useful since he only hits 28.5% of them for his career. In his second year in the league, Ellis also doubled the rate at which he got to the free throw line, allowing him to increase his efficiency that way. Last year, Ellis further improved his mid-range game, allowing him to take even fewer three-point shots, which led to the observed increase in effective shooting percentage. Look at his hotzone shooting chart from the last two years:

In 2006-2007:

And then, in 2007-2008:

Ellis improved his midrange FG% from an okay 39% to a very solid 44%. Meanwhile, he took less than 1/3 the amount of three-point attempts per game as he did the previous season.

Finally, in 2006-2007, Ellis turned the ball over almost 16 times for every 100 times he used a team possession (leading to almost 3 turnovers per game), while in 2007-2008, he cut that rate down to below 11 times per 100 possessions (even very talented young players tend to have high turnover rates, so in that sense the decrease was expected. For Monta, his turnover rate was further depressed by the fact that he was given fewer ballhandling responsibilities last year, decreasing his opportunities to turn the ball over) .

The decrease in turnovers and increase in shooting percentage together allowed Ellis to increase the points that resulted from his touching the ball from a pretty average 1.04 points per posession to an elite 1.17 points.

That might be a little abstract, so a better way of saying it: the only perimeter players who produced as efficiently as often as Monta Ellis last year were Chris Paul, Deron Williams, Kevin Martin, Steve Nash, Chauncey Billups, and Jason Terry.


Nash, Paul, Williams, and Billups, as point guards, had a very different role on their respective teams compared to Ellis. Of the remaining two, Ellis fell somewhere between Jason Terry and Kevin Martin (Martin had a severely underrated year, as highlighted below). At this point, I think statistically the most comparable year was Jason Terry's , and Jason Terry is 30 years old while Monta is 22 and steadily improving. Note that Terry and Ellis achieve their results in vastly different ways, as Terry takes over a third of his shots from behind the 3-point line while Ellis almost never shot a three-pointer last year.

The takeaway from all of this? Monta Ellis is positioned to become a frighteningly good offensive player.

Part II - The Next Big Step: Becoming the Offensive Focus

In interviews this summer, Don Nelson has claimed that Monta Ellis will be the starting point guard for the Warriors next year. Personally, I quite liked Monta as a small 2 alongside a bigger point guard. Kawakami wrote earlier the summer that he didn't like the idea of Monta at the point, either. The last time he played significant minutes at the point (during the 06-07 season), Ellis suffered, particularly because he turned the ball over too much. In this post I don't intend to argue the merits of one position vs. another as a good home for Ellis, but with Baron Davis gone, and Monta signing a 6 year, $67 million deal, it would seem reasonable to assume that Monta Ellis is going to be playing a much larger role in the offense, regardless of position. In terms of numbers, we should expect to see an increased usage rate from Monta. (As the name implies, usage rate is a measure of the percentage of a team's posessions that are "used" by a player when he's on the court. By "used" I mean the player is responsible for the end of the possession, by shooting/getting fouled, assisting, or turning the ball over. For reference, the highest usage rate players in the league are players like Kobe Bryant, Lebron James, or Dwyane Wade who have the ball in their hands a lot, while spot up shooters, energy guys/rebounders, and other role players have lower usage rates).

The reason I bring up the question of Monta's usage is to get an idea of what can be expected of him, and what he'll need to do next year. For now, I'm ignoring the complication of playing him at the point. Studies have shown that an increase in usage tends to lead to a decrease in efficiency, which also makes sense intuitively since, presumably, you'll be forced into more turnovers and bad shots when you're trying to be the main offensive threat for your team.

If Monta Ellis were a fully-formed player without any expectation of increased efficiency as an offensive player, and if he were to increase his usage rate from what he provided last year (21.7, which is a little more than average) to what Baron Davis provided last year (25.6), we would expect his efficiency next year to decrease to around 112.1. That moves him from the elite territory highlighted above into a neighborhood with Michael Redd -- still a star but not quite transcendent. Will he continue to improve, thereby negating some of the expected decrease in efficiency? Hopefully . . ..

Part III - Comparison, Hopes, and Avenues for Improvement

About a year ago, Tom Ziller at Sactown Royalty did a very interesting preview of Kevin Martin's upcoming season, comparing Martin to his closest comparisons, Reggie Miller and Michael Redd, and setting reasonable expectation levels given the expected jump in Martin's usage rate (as for results, both at midseason and season's end, Martin came through with absolutely stunning numbers). I was thinking of that post recently when reading about Monta Ellis' recent deal, 6 years and $67 million (compare to Martin last year, who signed for 5 years and $55 million), since for some reason in my head I'm always comparing the two players.

But upon further reflection, these are two completely different players.

Which got me to thinking, who has had a similar season to the one Monta Ellis just completed? Monta just finished his 3rd professional season and is 22 years old.

It looks like Kawakami went through a similar exercise in February and came up with George Gervin and Kevin Johnson. Looking at the numbers, I'm coming up with the following list: Sidney Moncrief, Kevin Johnson, and Dwyane Wade.

What Ellis shared with these players was an ability to be incredibly efficient offensively without using the 3-point line as part of his arsenal. He also has notable differences from each of them. Unlike Ellis, Moncrief was a tenacious defender and prolific offensive rebounder. Kevin Johnson, obviously, was a much more natural point guard and able to create numerous shot opportunities for his teammates without Monta's tendency to turn the ball over when pressured. Wade, meanwhile, has developed a league-leading ability to get to the free throw line.

I'd like to propose these comparisons and contrasts as a point to look for possible avenues of improvement. If Nelson is being honest and Ellis is to play the point next year, then improving his ballhandling and passing skills, a la KJ, should be the main priority. If Monta is allowed to continue to play at the 2, then increasing his free throw attempts should be a goal (perhaps he can learn from new teammate Corey Maggette) -- this avenue is sensible considering Monta's highlight-worthy finishing ability around the rim, which could lead to many 3-point plays. Also promising for Monta -- Wade's huge increase in free throw attempts came during the season in which he turned 23 years old, and Monta will turn 23 in the upcoming season. Becoming a better defender (like Moncrief) should be a goal regardless. The other possible route for improvement is the addition of a 3-point shot -- here Monta would be similar to Richard Hamilton in being an effective mid-range player and then slowly extending his range beyond the arc (Hamilton is now a very effective outside shooter). Given the Warriors reliance on three-point shots combined with their lack of a consistent shooter from that range, this is a possibility (though personally I'd hope to continue to see Monta improve his game inside the arc).

I do not at all expect Monta to become as good a point guard as Kevin Johnson or take as many free throws as Dwyane Wade, but just wanted to throw some familiar names into the mix to give an idea of where Monta is historically.

Part IV - Caveats

Ellis' claim to fame is his inflated shooting percentage (he shot 53% from the field last year), but historically FG% is the most volatile stat year-to-year, and we might be in for some regression to the mean. If he can keep his True Shooting Percentage above 55% that should be sufficient -- he may need to take more free throws to keep that up.

Ellis has also been criticized as only getting such lofty numbers because he was playing alongside Baron Davis. In looking at his player pair data, there is some support for that claim -- Ellis shot 53.8% last year when playing alongside Davis, but only 50% when playing without Davis. Furthermore, 46% of Ellis's shots were assisted last year. Without Davis in the mix, Ellis will not only be shooting more, but he'll be creating his shots off the dribble more often. Can he continue to be efficient without the help of a shot-creator like Davis? This is just one more reason that Monta Ellis's season will be one to keep an eye on in the upcoming season.

No comments:

Post a Comment