"Finally, Evidence for a Momentum Effect in the NBA"

[Crow]

http://www.bepress.com/jqas/vol7/iss3/13/

by
Jeremy Arkes, Naval Postgraduate School
Jose Martinez, Universidad Politecnica de Cartagena


It would be interesting to check whether momentum effects are generated largely from top minute lineups or if they often come from small minute lineups and how consistent or random the momentum leaders are. And if looking at plus-minus synergies are a useful additional guide tool for finding these momentum yielding, over-performing lineups and then trying to identify the context that unleashes them best. The distribution of the momentum effects should affect whether a coach rides best recent performing lineups real heavily, and the recently hot ones vs. the season's best, and how much he searches for unique match-up solutions.

[Mike G]

... how success over the past few games leads to a higher probability of winning the next game. The model takes into account the home vs. away strengths of the teams in the current game as well as their opponents in the previous games ...

It isn't within-game momentum but game-to-game.
There could be obvious lineup changes, such as a player getting injured or returning from injury, that drives 'momentum' thus defined.

[Crow]

I was referring to game-to-game momentum.

Injuries will affect the studies results but with players getting injured and most returning from injury I don't think it will bias its findings much.

[Italian Stallion]

I think both intra game and game to game momentum clearly exists. I'm glad some statistical evidence of game to game is coming to light. Anyone that has ever played any game or sport at a reasonably high level has experienced some of the psychological and physical differences you can feel when things are going especially well or poorly for a period of time. Things can change from a random distribution of events into a kind of a self fulfilling sequence of events based on your mental and ultimately physical state of being.

[Mike G]

Haven't previous studies indicated there was not much benefit in weighing recent games more heavily, in predicting the next game?
What did these guys do differently?