Friday, August 19, 2011

SPS: The Holy Grail

In writing about DeShawn Stevenson a while back I leaked my most top-secret player evaluation tool.

I had no idea what would happen after I leaked the information, but since that time at least one well-known writer has silently slipped the tool into his player-evaluation toolbox.

So now that the cat is out of the bag, I might as well go ahead a release the entire system.

It is called situational performance splits, or SPS.

SPS is the reason why I keep praising Baron Davis while others keep ripping him and why I had TJ Ford rated as my top free-agent point guard before I did away with the list.

Any good player evaluation system should look to find players who will lead their teams to championships.

And SPS does that by paying less attention to games which take place outside of championship environments and focusing more on players and how they perform against the best players and best teams and in the biggest games.

And if it involves the final two minutes of a one-possession game, that's even better.

Whenever I'm watching games and breaking down players I pay very close attention to the following things, which make up the essence of SPS.


After a team reaches the second round of the playoffs, you can pretty much bet that the teams are going to be loaded with the best players in the league.

In order to know if a guy can lead a team past this point, you have to know how he performs against the best players in the league at his position.

So whenever you're watching games, pay particular attention to specific matchups against top players. And at the very least you want to see specific matchups against very good players.

Some guys are better on one side of the court.

In these cases, for example, you would want to watch how a small forward plays on offense against Ron Artest and Tony Allen and on defense against Carmelo Anthony and Paul Pierce to get an idea of what he is capable of in big games.

Carmelo Anthony is someone who scores highly on my board.

For all of the stat-geek criticism, Anthony sports a 10-1 career record against Kevin Durant and a 9-4 career record against LeBron James.

And based on the swag and audacity Anthony plays with when he plays against James, I would never know James was the better player.

Anthony goes hard at James like James is some scrub in a pickup game at Rucker Park.

And that's what you look for in these individual matchups, guys who don't back down against the best competition.

And guys who are productive and win against the best competition and relish the opportunity to go against it rate highly in SPS.

It must also be pointed out that good performance is very much a product of good coaching and good teammates.

To find out why someone is underperforming you may have to look beyond his individual shortcomings and look at the coach he's playing for, the system he's playing in and the teammates he's playing with.


The next thing you want to look at is team matchup.

Specifically, you want to see how someone does on offense against the best defensive teams and how they do on defense against the best offensive teams and how they do against the teams with the best records and coaches in the NBA.

All information against teams like Minnesota and New Jersey must be excluded.

Based on last year's coaches, rosters and records the following teams would be of best use: Boston, Chicago, Dallas, Los Angeles, Miami, Oklahoma City and San Antonio.

And of course, things change.

Denver, Memphis and Portland were all on fire toward the end of the season. And Oklahoma City played much better after Kendrick Perkins and Nazr Mohammed were integrated into the lineup.

So games against those teams during that time would be useful as well.

But I wouldn't go much further than that because you start getting into serious mediocrity at the coaching and/or talent levels after that point.


All games are not created equally. Some games are simply bigger than others.

And SPS looks to identify those big games and draw more conclusions from them than the other games on the schedule.

What is at stake here?

That is what you want to find out in determining the magnitude of a game.

Is it a NBA championship? A series? A division title or playoff berth? A reputation? Nothing?

It goes without saying that in order to win it all a team is going to have to win some big games, and at least some of them on the road.

And you can't win big games without big-game players.

As best as I could I built a list of the biggest games, in order of importance:

1. NBA Finals games
2. Conference finals games
3. Second round playoff games
4. First round playoff games
5. Elimination games
6. Playoff position games during the last few weeks of the season
7. Games after the All-Star break
8. Games against top teams and players
9. Playoff position games before the last few weeks of the season
10. Road games, especially those in hostile environments
11. Games during long winning streaks (in general, 10 or more wins)
12. Games on national television

Some of those categories blend into one another.

In short, what you're trying to do is turn the pressure, the intensity, the competition and the attention levels way up to get them to match what someone will face as he tries to win deeper in the playoffs.

And playoff games clearly get bigger the further you go into a series, with game seven of the NBA Finals being the biggest game possible.

Watching guys under these circumstances tells you all you need to know about them.

Coaches too.


When I say game situation I am obviously talking about the clutch, the last two minutes of one-possession games.

Getting back to TJ Ford, he led Indiana to a win over the Lakers in Los Angeles six days after beating the Heat in Miami. And Baron Davis also has a few wins over the Lakers and Heat (one with the lowly Cavaliers) on his resume from last season.

Ford led the charge in Indiana's win over Los Angeles by flawlessly executing clutch plays down the stretch.

His play was what basketball is all about.

Let's take a look at the SPS factors and try to put some sort of value on the win.

There are some qualifiers.

Andrew Bynum didn't play. He was matched up individually against Steve Blake, who isn't a very good player. Although it clearly became a big game by the end, this wasn't a big game for Los Angeles coming in.

So you could say Indiana was able to sneak up on them and catch them by surprise with their strong play. And because it was so early in the season and the teams' records so far apart, there was very little at stake.

All that being said, this was still a significant accomplishment.

TJ Ford led lowly Indiana to a victory over the two-time defending world champion Los Angeles Lakers by outdueling Kobe Bryant and Pau Gasol in the clutch, on their home floor.

I don't have a problem with that victory.

In addition, Ford led Texas to the Final Four in college and took Milwaukee and Toronto (Milwaukee and Toronto!) to the playoffs two times each.

And there is no doubt he would have helped Indiana to the playoffs with a much better record last season had he not been benched in the middle of the season for no good reason.

Mike Bibby is the only one who can match Ford's level of accomplishment among free-agent point guards, but Ford is younger than Bibby and significantly more athletic.

And that is my ultimate player evaluation tool, and it is the Holy Grail of player evaluation systems.

So when you read my work and see that I seem to be speaking a different language and operating on a different wavelength than others who cover the NBA, now you'll know why.

We are looking at two totally different things, at least.

I put little stock in 82-game performance. I put my stock in situational performance and try to pick the situations which most resemble championship caliber.

And I seek out and praise players who excel in those situations, among others.

Note: HBIQ is notorious for jinxing players and teams.

So now that I have big-upped Carmelo Anthony, Baron Davis and TJ Ford don't be surprised if they totally flame out and suck beyond all recognition for the rest of their careers.

But don't let that discourage you from using the system.

It works fine as long as I'm not the one issuing the praise.

Carmelo Anthony's career record against Kevin Durant was posted by Sarcastic on Inside Hoops. His records against Durant and LeBron James are from Basketball Reference.

And TJ Ford had Milwaukee in the playoffs during his rookie season before he got injured and missed the rest of the season. Milwaukee went on to make the playoffs without him.

I counted that in his favor because there is little doubt they would have done the same with him healthy and in the lineup.

Some of the additional Ford information I learned from Mark Medina.

To wrap it up, I would like to say that I am only one man and can't possibly keep track of all of this information by myself.

But I do the best I can with what I have.

Comment or e-mail:

Wednesday, August 17, 2011

Manu Ginobili: Let Me Entertain You

This is a Manu Ginobili highlight reel to "Let Me Entertain You" by Robbie Williams.

Although he doesn't get much fanfare, Ginobili is well respected for his game.

And he deserves every bit of it and more.

So let him entertain you.

Comment or e-mail:

Thursday, August 4, 2011

Advanced Passing Stats: The Nickel And More

It's been too long since the last time I brought out some advanced statistics.

But it's that time again.

Passing is the part of the game I've been working on most recently.

So I want to show everyone what I've been working on and get you all familiar with these concepts.


A nickel is when a guy makes a nice pass only to have his teammate (1) miss an open jumper, (2) get fouled, (3) turn the ball over or (4) miss a shot close to the rim.

Nickels will generally be credited only for the fouls and turnovers which take place close to the rim.

And missed contested jumpers aren't nickels because there is no real expectation for a make on contested jumpers, and getting someone a contested shot isn't an indicator of strong passing ability.

Missed shots close to the rim which are contested will be judgment calls.

For misses where it is determined that the contest was a major factor in the miss, no nickel will be credited. For misses where it is determined that the contest didn't play a major factor in the miss, a nickel will be credited.

And when a player is fouled off of a nice pass and goes to the line and makes one free throw, the passer will get a nickel and a penny (or six cents); and when the player makes both free throws the passer will be given a nickel and two pennies (or seven cents).

To see this in action, let's look at Terrence Williams in his debut with the Houston Rockets.

Williams makes three nice passes here: plays starting at 1:42 to Patrick Patterson, 2:41 to Jared Jeffries and 3:08 to Courtney Lee.

But Patterson is fouled. Jeffries misses a layup. And Lee misses a three.

In these scenarios, Williams would be credited with three nickels plus the penny from Patterson's made free throw.

Looking at Williams' box score, you wouldn't think he did much of anything in this game. But the story from the ground paints an entirely different picture.

Looking at the game, you can see Williams was effective and made contributions to the team which the box score didn't capture.

And that is what the nickel is here to do, give what the box score doesn't and properly evaluate every player's passing skills.


This is an extension of the hockey assist, but it takes it a bit further to include not only the pass before the pass but all passes before the pass leading to a made basket.

Let's look at an imaginary scenario to see how it works.

Shaquille O'Neal passes the ball from the post out to Kobe Byrant, who swings it over to Rick Fox, who kicks it over to Ron Harper, who whips it over to Robert Horry for a three.

I know Kobe Bryant would fire up a shot the second the ball grazed his fingertips (I kid, LOL!). That's why I said this is an imaginary scenario.

In this case, here is how the assist would look: team assist: O'Neal-Bryant-Fox-Harper.

The last person listed in the assist string (Ron Harper) is the one who gets credit for the individual assist.

Looking at another imaginary play, let's say Kevin Love gets a rebound and zips an outlet pass to Ricky Rubio and three-quarter court.

Rubio throws a touch pass to Wes Johnson under the basket, and Johnson dishes a behind-the-back pass to Michael Beasley for a dunk.

The assist string would look like this: team assist: Love-Rubio-Johnson.

Calls will have to be made as to exactly where the assist string begins, but this is for rapid-succession passes which lead to made shots.

It's not for guys who catch a pass and hold it for five seconds, then dribble around and pass it to another guy who holds it for five seconds and dribbles it around some more.

It is for quick passes leading to buckets.

And it is designed to reward team play and give everyone credit for their contribution to the field goal.


Extra passes create better team chemistry. And better team chemistry leads to wins.

An extra pass is when someone has an open shot and passes it up to give a teammate a shot.

Guys who make the extra pass are unselfish, team players who make offenses go and who everyone loves playing with.

And the camaraderie they bring with their unselfishness brings the team closer together and encourages others to do the same.

So I'll be looking at guys who make the extra pass and what impact it has on how their teams perform.

And it is well understood that you don't want to get too carried away with these extra passes to the point where it hurts the team.

As I say with all of my advanced statistics, I don't have anywhere near the resources or the manpower to track and log this information as I would like.

But I will track some games or parts of games from time to time.

Note: It took me years to come up with the name nickel for the concept of one-half of an assist.

It finally came from the realization that you need two things to get an assist, or a dime: (1) a pass and (2) a made shot. A pass and a missed shot is half of an assist, or half of a dime.

So I named it the nickel. Sounds simple but it was really hard.

Comment or e-mail: