Basketball-Reference has edited its box score for the Western Division Round Robin Playoff game between the Rochester Royals and Fort Wayne Pistons on March 21, 1954. In addition to adding newly discovered assists for all players in the game, we have edited our previous total of seven assists for Cal Christensen to 10 assists, giving him a 'new' triple-double. We'll explain what we mean by that a bit below.
The NBA never officially tracked triple-doubles until a rookie named Earvin 'Magic' Johnson made the pursuit of this trifecta a 'thing' in 1979-80. There is some dispute as to who came up with the term, but naming it led to tracking it carefully in the future. However, obviously many players had been amassing triple-doubles before 1979.
You can get a triple-double by recording double digits in any of five statistical categories: points, rebounds, assists, blocks and steals. Blocks and steals have only been officially recorded since 1973-74 and double-digit performances in either category are rare and extraordinary enough that they can be found using various sources and record books. The NBA itself has even published official triple-double leaderboards (the career record belonged to Oscar Robertson, who obviously played before 1979-80, for many years). In Robertson's day, you could only get a triple-double via points, rebounds and assists. Had blocks and steals been part of the equation, it's likely that Wilt Chamberlain or Bill Russell would be the all-time leader. Fortunately, single-game point, rebound and assist totals are largely known in Robertson's era (his career began in 1960).
However, there are lots of discrepancies and unknown single-game totals in the NBA of the 50s. This is also true of the 1940s, but since rebounds were not an official statistic in the 1940s it was impossible to record an 'official' triple-double, anyways. Justin Kubatko's recent blog post is a great example of some of the uncertainties with statistics in the 1950s. Therefore, in the 1950s, there are many 'known unknowns' such as a great deal of triple-doubles we know Bob Cousy had for the Celtics, but where we don't know what game it happened in. For instance, we know Cousy had 33 triple-doubles in his career, and we even know which season they all happened in. But we have the particulars on fewer than 20 of them. No other player has nearly as many (Cousy accounts for over half of the 'missing' triple-doubles in our dB).
However, until we recently the Christensen triple-double was not an 'unknown known', but rather it qualified as completely unknown since we had never seen reference to it in any media guide, record book or newspaper. The likeliest explanation is that the original official scorer's report from this game was discarded or lost decades ago and so an accomplishment such as this contained within it can't 'officially' exist. The Pistons and Kings are two of the NBA's oldest franchises (predating even the league itself since they both came from the NBL), and moved cities in an era before the league was popular enough that anyone could imagine statistics would be considered anything close to sacred. Archives such as filing cabinets full of box scores did not often make these types of trips. Combine that with the somewhat lax approach to record-keeping at the time and you could see how this might happen.
Until today, we had Christensen's line in this game as 12 points, 17 rebounds and seven assists. Considering this was a playoff game, it likely qualified as one of the best games of his five-season career in which he averaged about five points, 5 rebounds and two assists per game . This box score line was based upon reporting in the March 22, 1954 Rochester Democrat & Chronicle, which rigorously covered Christensen's Rochester Royals. The box score in the paper (and everywhere) included points for every player, so his 12 points is easily nailed down. It also included:
Along with the scoring, the Royals contributed a fierce rebounding game, snagging 60 balls off the board to the Zollners' 42. Christensen grabbed 17 rebounds and weighed in with seven assists."
Entries such as the above are why many of our old box scores will have rebounds and/or assists for some players, but not others. Simply put, newspaper articles might have included statistics for standout players that couldn't be found in the abbreviated box scores that were found in most newspapers at the time (The Philadelphia Inquirer being a standout exception to this rule, with very full box scores).
Fortunately, websites like newspapers.com and genealogybank.com are regularly adding old newspapers to their archives. One that has popped up in the last year or so is the Fort Wayne News Sentinel on Genealogy Bank. The Pistons were a very big deal in a smaller town like Fort Wayne and generated great coverage. Additionally, the News Sentinel almost always included FGA and assists in their box scores for Pistons' home games (a relative rarity at the time). Since this game was played in Fort Wayne, it means their box score has assists for all players in the game. The box score credits Christensen with 10 assists (not the seven credited in the Rochester paper). Additionally, it contains the following entry:
They had a surprising 60-42 rebound edge, Cal Christensen, once a castoff, setting the pace with 17 grabs. No Piston was in double rebound figures. Cal, who specializes in pestering the Pistons, also scored 12 points and had no less than 10 assists.
So the Fort Wayne paper confirms the 12 points and 17 rebounds, yet also credits him with 10 assists in the box score and in the game story. This seems more authoritative than the Rochester paper saying seven because it is repeated in two places and because they are showing assists for all players in the game and the numbers in the box score add up. Even though it credits Rochester with 29 assists on 33 made field goals, when you consider the strict dribbling rules of the time which emphasized team play it is a believable total. The other factor at play here is that in this era the box score and story for the home city are almost always better than the one for the road city. Most teams did not have anyone from home covering them on the road at this time. This leads us to believe that 10, while not guaranteed to be accurate, is absolutely more likely to be accurate than seven and represents the best available information.
The other thing to consider here is that, while we don't have game-by-game assist totals for Christensen, we do know how many assists he was officially credited with in the 1954 playoffs. In six 1954 playoff games he was credited with 15 assists. Of his five other playoff games, we know he had two in one of them and none in the other. That leaves three other games. If his total on March 21 was 10, then he had three assists in the three other games. If his total on March 21 was seven, then he had six assists in the three other games. Accepting that the March 21 game represents an outlier either way, his regular season average of 1.5 APG makes it easy to believe that he could have averaged either 1.0 or 2.0 APG over three games. So I don't think that his official playoff totals really offer any clues here unless we can find other pieces of this puzzle in the future.
In summary, we have found compelling evidence to credit Cal Christensen with 12 points, 17 rebound and 10 assists on March 21, 1954. This represents a triple-double we had previously never known existed. We hope you found this as interesting as we do and a useful glimpse into the work we do to make our sites as comprehensive and useful as possible. Most importantly, if you're reading this and have access to old NBA scorer's report or box scores, please let us know!
