clock menu more-arrow no yes

Filed under:

How a coach reads a box score

New, comments

Which parts of the box score matter, and which are fluff. Which statistics matter when trying to figure out why a game was won and lost? Pounding the Rock's resident ex-coach sets the record straight.

Russ Isabella-USA TODAY Sports

One of our very strong teams at Claremont opened league play at Redlands one season.  Redlands was running the old Loyola Marymount offense which involved firing up 3 point shots early in shot clock.  That night, those shots went in, and we lost by 10. 

When the coaches got back to locker room after the game, upset with how our team had played, we started examining the box score.  Redlands shot 18 for 25 from the beyond the 3 point line.  Put another way, they had scored 54 points on those 25 shots - the equivalent of shooting 108% from the field, even before factoring in lay-ups and other 2-pointers.  If they had shot 13 for 25 from the 3 point line, a still incredible 52%, we would have won the game by 5.  If they had shot a good but not incredible 10 for 25, 40%, we would have won by 14 - and we would have been much happier with the players.  Put another way, making 8 more threes than a "normal" 3 point shooting percentage of 40% meant 24 additional points for Redlands.

What the box score really told us was when the other team shoots 18 for 25 on their three-pointers, even a good team is unlikely to win that particular game.  The same rule applies to NBA games.  In Game 5 of the 2011 Finals, Mavericks vs. the Heat, the Mavs went 13 for 19 from 3 - the equivalent of shooting 103% on two-pointers. The Mavs won that game to take a 3 - 2 lead.

Since I stopped coaching, I have continued to closely examine box scores after games.  What I look for are not the fairly useless statistics that many national basketball announcers focus on.  Mark Twain famously referred to "lies, damned lies, and statistics" [Editor's note: Twain's put his list in order of increasing evil - jrw].  That applies to basketball too.

A good example of a useless statistic is "total rebounds".  Since the defensive team recovers most missed shots, total rebounds often just reflect which team shot and defended better.  It does not tell us very much about which team actually rebounded better.

A much more meaningful statistic to measure rebounding is the percentage of a team's missed shots it recovers through an offensive rebound. The team that rebounded better in that game will be the team that got a higher percentage of its own misses back.

The other side of that coin may even be more meaningful, since (I am told) it more often correlates with winning percentage: The percentage of the other team's misses a defensive team rebounds, thus preventing the other team from getting a second chance to score.

A normal NBA team will get between 25% and 33% of its own misses back as an offensive rebound. 

Watch for offensive rebounding percentages outside of this range.  If a team gets more than a third of its misses back as offensive rebounds, it has rebounded well -- and the opposing team will hear about it after the game, and in the next practice.

This is something you can measure at home. Look at the box score to see how many shots a team missed. (For simplicity, I ignore missed free throws since the defensive team rebounds virtually all missed free throws.)

If a team shoots 40 for 85 from the field, there were 45 misses.  Each miss represented an offensive rebounding opportunity. (I have had many a game where my shooting supplied my team's big guys with many offensive rebounding "opportunities".)  If the offensive team gets 15 or more of those 45 misses back as offensive rebounds, that is very good -- and the opposing coach is upset with his guys.  But if the offensive team gets less than 10 offensive rebounds on those 45 misses, that will make it very hard to win.

Applying this statistic to last year's Finals is telling. In Game Five, the Heat actually had more total rebounds than the Spurs.  As you may recall, the Heat still lost - by a lot.  One reason not mentioned on the air:  The Heat had only 5 offensive rebounds in the entire game, on 45 missed shots.  This was a dreadful "offensive rebound percentage" of 11%.

This was a series-long problem for the Heat, and an important one.  For the Finals, the Heat got an offensive rebound on only 15% of their misses, about half of what a good rebounding team will get.   (The Spurs, generally not a strong offensive rebounding team, were at 23% for the Finals.)  When the Heat missed shots in the 2014 Finals, 85% of the time the Spurs rebounded the ball and could immediately attack, feeding into their offensive dominance.

National announcers also love to mention "points off turnovers".  This statistic tells us virtually nothing.  Unless the turnover is a steal, it is actually harder to score off a turnover than a missed shot.  The turnover that goes out of bounds, or results from a violation or offensive foul, stops the clock and requires the referee to handle the ball. This stoppage allows the defense to retreat and set up, which is the hardest time to score. 

Last year in the NBA, teams shot an effective 61% after live ball turnovers, but only 46% after dead ball turnovers.  While watching the game, watch for live ball turnovers.  Blocked shots that are recovered by the defense (and don't go out of bounds) are similar. They often lead to easy baskets the other way.  (This is also a rarely mentioned problem with the Hack-a-Shaq.  Putting the other team on the free throw line essentially eliminates any chance of a steal or blocked shot and the resulting easy transition basket.) Whatever you do, ignore the "points after turnovers" statistic they throw on the TV screen, since it ignores the crucial distinction between live ball and dead ball turnovers.   

Also be wary of the "4 point swing" announcers love to tout.  This happens when a team misses at one end, and the other team comes down and scores.  We eliminated the center jump after every basket in the 1930s.  The "four point swing" ignores the fact that the other team would have still gotten the ball even if the first team had made the shot. All that being said, total turnovers do matter: Put simply, the team that turns the ball over has a zero percent chance to score on that possession.

Since most teams score about a point per possession, each turnover costs that team about a point. Put another
way, if the Spurs have 10 turnovers in a game, and their opponent has 15, that is generally worth about a five point swing in the final score.  (Of course, since the Spurs are averaging over 16 turnovers per game, having a game with only 10 would be cause for much rejoicing.)

Technical fouls, including those from defensive 3 second violations, matter too. Each is worth about a point too, since normally the team puts their 90% free throw shooter at the line for the free throw. Since teams should shoot about 75% on free throws, any made or missed free throws above or below that 75% line is also a worth about a point. (As an example, Wednesday night the Warriors went 20 for 20 from the line against the Clippers - which is 5 points more than the "expected" 15 for 20. Warriors also went 15 for 25 from 3, which is 15 points more than if they shot a more normal 40% and went 10 for 25. Not coincidentally, the Warriors won.)

Finally, a team's raw shooting percentage can also be misleading, which is why people with better math skills and more time than me have created all-encompassing stats like "true shooting percentage" or "effective shooting percentage".  These new-age numbers factor in free throws percentages, number of times fouled while shooting, and 3 point shooting.  If you have that number, go with it.  If you can't get that, there is a shorthand method you can do at home.  For every three pointer a team made during the game, add ½ of a made shot.  A team that went 40 for 80 shot 50% in raw shooting percentage.  But if it made 10 threes, that is the equivalent of going 45 for 80, which would be 56%.  With that number, the number of free throws shot and a team's free throw percentage, you can get a pretty good idea of which team actually shot better. 

Add the shooting information to the other box score factors discussed above.  This will give you a better idea of why one team won, and the other didn't - certainly a better idea than relying on the meaningless statistics many national announcers like to throw at us.