Advanced Stat Talk – A Different Way to Look at Strength of Schedule
By Tim Donahue
In the NBA, Strength of Schedule (SoS) is a relatively simple calculation. As shown on this ESPN Page, it is merely the compilation of the current win-loss records of a team’s opponents. Each opponent’s record is added for each game they played. (As an example, the Pacers have played the Hawks twice, so they contribute 66-36 to Indiana’s SoS calculation.)
It’s not something that gets a great deal of attention. SoS carries more weight in college basketball, where rankings affect seeding in the tournament, and in college football, where rankings largely determine a mythical national champion. Since the NBA playoffs, seedings, and champion are all decided on the court, SoS has less utility.
However, it can be of some use when you’re sitting in the middle of the season, trying to figure out where teams stand, and where they are going.
Exhibit A – NBA Strength of Schedule 2/13/2010
This chart shows the current SoS for the NBA, ranked in descending order. Utah, Houston, and Miami all have had markedly tougher schedules than the rest of the league. Meanwhile, your Indiana Pacers have played the second “weakest” schedule at .483, nestled in between the Knicks and the Sixers. However, there are some limitations to the simple S0S. Specifically, it ignores the question of where the games have been played.
Weighted Strength of Schedule – Home/Road
A couple of weeks ago, I did a rough projection of where I thought the Pacers would finish. That got me to start looking at the types of games they had won, and the types of games they had lost. In the midst of this, and another project I’m working on (and hope to get up in the next week), I decided I wanted to cut a little deeper into this area.
The adjustment I made was relatively simple. Instead of using the opponent’s overall record, as the normal SoS does, I varied the record based on the site. Therefore, if the Pacers played Atlanta in Atlanta, I used the Hawks’ 21-6 home record. If they played them in Conseco, then I used Atlanta’s 12-12 road record.
Why does this matter? Well, the home team generally wins more than 60% of the time. Through the All Star break, there were only six (6) teams with winning records on the road, and another three at exactly .500. Meanwhile, there were 22 teams with winning records on their home floor. Last season, there were 21 teams with winning records at home, but only seven (7) with winning road records. Home court advantage is a real thing in the NBA.
Like I said, this seems so simple that I’m sure somebody, somewhere has done this before, but I can’t find it anywhere. Therefore, this gave me three new (to me) metrics that I get to name. The first is “Weighted Strength of Schedule – Home/Road” or SoSHR. This is exactly as I described in the previous paragraph. I use the same basic math as SoS, but substitute an opponent’s road record for a team’s home games, and their home record for a team’s road games.
Exhibit B – NBA Weighted Strength of Schedule – Home/Road 2/13/2010
Granted, it doesn’t make a huge amount of difference in Indiana’s case (.480 SoSHR vs. .483 SoS), but it does have a dramatic effect on some teams. The New Jersey Nets, as an example, move all the way up to first or “toughest” with a .537 SoSHR. Of course, the Nets’ schedule will always be somewhat tougher than anyone else’ simply by virtue of the fact that they don’t get to play themselves. Exhibit C illustrates the comparative SoS vs. SoSHR for each team.
Exhibit C – SoS vs. SoSHR (Sorted in descending order of SoS)
Here, you can see some of the dramatic swings. The previously stated New Jersey example is on the right, but it’s also important to take a look at Utah on the far left. They were looking pretty good with the 6th best record in the league against what appeared to be the “toughest” schedule. However, digging little deeper shows that their schedule is nowhere near as tough as it first appeared.
In fact, both Utah and New Jersey’s SoS were skewed by an imbalance in home vs. road games. The Jazz have played seven (7) more games at home than on the road. The Nets have played five (5) fewer home games than road games. Of course, some could say that I could get the same answer simply by looking at net home vs. road games. To those people, I say, “Ehhhh…leave me alone.” Besides, this particular metric wasn’t the one that I was trying to quantify when I wandered down this path.
I was really more interested in the next two.
The “LossFactor” (LF)
What I was really trying to do was come up with a sense of quality of wins. The SoSHR is really an amalgam of the other two metrics. The “LossFactor” (LF) and the “WinFactor” (WF) are two sides of the same coin. In effect, they are the SoSHR in losses (LF) or in wins (WF). For the Pacers, it’s the WF that I’m more interested in, but the LF deserves some discussion.
Exhibit D – NBA LossFactor 2/13/2010
The LossFactor is something that is probably of more interest in looking at top teams, or at least playoff teams. It can give you an idea of how good a team is at “taking care of business.” The league-wide LF is about .580, and no one has an LF of less than .523 (the Knicks). The teams towards the right hand side of the chart, particularly those to the right of the red LG AVE bar, can be said to be leaving a lot of games on the table.
Meanwhile, this chart shows why many consider the Lakers to be hands down favorites to win it all this year. They have an LF of .674, which is well above anyone else in the league. Of their 13 losses, only one has come to a losing team (the Clippers), and that was on the road (yeah, I know). Of the four losses at home, two of the teams (Dallas and Cleveland) have winning road records, and the other two (Denver and Houston) are within one and two games of .500 on the road, respectively.
The Pacers are below average, but not horribly so. They are getting hurt by road losses at Golden State and Minnesota, and home losses to bad road teams like the Knicks, the Bucks, the Sixers, and, just before the break, the Bulls. The real bad news is what comes next.
The “WinFactor” (WF)
The thing that has bothered me all year long about this team is that I have never seen them show any sustainable success. Throughout the first part of last season, there was a sense that they were letting games slip away. All the games, all year long, seemed close and competitive. In fact, they had the fewest games decided by double digits (24) in the league last year. The 2010 Indiana Pacers have already lost 22 games by double digits (and played in a total of 33.)
A quick glance at the WinFactor (WF) chart tells the depressing story.
Indiana’s .317 WF is worst in the league. Even worse, it is way-out-of-whack low. The difference between the Pacers and the Bucks (.043), who sit at 28th, is greater than the difference between the Mavs and the 18th place Bobcats (.042). While the Pacers are not the bottom of the food chain, they’re awfully close. More importantly, they’ve shown precious little ability to defend themselves against the predators further up on the food chain.
Even during the heady days of the now legendary “Five-game winning streak” in November, there was nothing of substance. The one quality win was in Conseco over Boston. That game was an anomaly even for the streak, being won at the offensive end in a rare display of firepower. Additionally, it was Boston’s third game in four nights, and the Celtics have a worse LF than the Pacers, indicating a tendency to “leave games on the table.” The other four teams in that streak presented a WF of .265.
The Pacers are an awful road team, winning only six times in 26 tries. All of their road wins have come against team with losing home records. There are only eight of those in the league, and one of them is the Pacers, so not a lot of sunshine there. Two of the wins have come against the Nets. They are a mediocre-to-bad home team, winning only 12 of 26 games. Nine of these wins have come against teams with road records of .346 or less.
The Pacers have gotten 39% (7) of their wins against teams with worse records than their own, but those teams only accounted for 17% (9) of their games. They’ve managed to win only one time in every four tries against teams with equal or better records.
What Lies Ahead
What do the remaining 30 games of the season have in store for the Pacers? Nothing good.
The SoSHR for the remaining Pacers’ games is .523 – over 40 points higher than the first 52 games. Either the Pacers WF is going to get a lot higher, or they’re going to have a lot of ping pong balls in the lottery this spring.
The sad truth is that the Pacers have won only two games this season where the WF would have been above .500 (Boston & Orlando). Even more depressing is that there were only three other wins where the WF would have actually cleared .400 (on the road vs. the Knicks and the Pistons, at home vs. the Suns). It’s hard to even look forward confidently to a high draft pick, because teams like Minnesota, Golden State, New York, and Philly all have tougher remaining SoSHR’s.
This is shaping up to be a truly lost season.
There’s little question in my mind that Jim O’Brien has been far too erratic with his lineups, and I think he’s got to stop tinkering and go with a rotation for the rest of the year. A large part of it is just for the sanity of everyone involved with the Pacers – players, coaches, management, fans, me. However, the more basic reason is that it’s becoming more and more obvious that their success hinges a whole lot less on names on the back of the Pacer jerseys, and a whole lot more on the names on the front of the opponents’.
We’re gonna need a bigger boat.