A multi-sport day of streakiness-related news...

The Oklahoma City Thunder laid another egg against the Dallas Mavericks last night, in the NBA Western Conference finals. Leading 99-84 with roughly five minutes left in the game, OKC gave up a 17-2 run, which sent the game to overtime. There, Dallas took over and won 112-105, giving the Mavs a 3-1 series lead.

***

St. Louis Cardinal slugger Albert Pujols ended a massive home run drought. According to this article: "Pujols had gone 105 at-bats, and 119 plate appearances, since his last home run on April 23. Both were the longest homerless streaks of his career." A whole month for Pujols without a homer? Amazing.

[Update. Kevin Lai, writing at Hardball Times, provides a sophisticated statistical analysis of Pujols's drought.]

***

In looking over the website for the French Open tennis tournament, I noticed that the statistical features for following a given match include a "Momentum Meter" from IBM (apparently, this feature has been around for a few Grand Slam tournaments already, but I just noticed it). Here's an example, from an early-round French Open women's match. As can be seen, each player's momentum is graphed from the beginning to the end of the match, sort of like the stock market's value during the day. Certain plays are picked out as "turning points."

I did some searching of the web to try to learn how momentum and turning points are defined, but I didn't find a whole lot. According to this article, the Momentum Meter displays "an overall swing of player success, calculated algorithmically from match data... [when] a player is particularly on top of their game, [exhibiting fewer] unforced errors, holding their service games [,] etc." I'm particularly curious about whether turning points are determined by some mathematical analysis of inflection points in players' momentum values or subjectively by a rater.