One of the things that Ben Raymond loves to do at his blog is answer interesting questions about volleyball using data instead of conventional wisdom. One of the topics we discussed along with Jim and Adrien has been crunch time. More specifically, when is it. For ease, in his apps we have defined crunch time as the a score after 18-18 (i.e. both teams above 18) with a difference of two points. This seems logical. One of the problems we encounter with this definition is that it happens so infrequently in actual games as to make the use of it in analysis and game planning more or less irrelevant. So he set about investigating the idea in more detail. You can read his original article on PlusLiga here (the charts that follow are taken from this article) and his article on other leagues here. You can read Jim’s article here. The following is my precis. The
Finding crunch time – The basic method used was to take large dataset, 400+ matches, take every score, and calculate the number of times the serving team won the next point AND the set. That gives us the value of winning any point in a match. Using the 10% most valuable points (the white dots), the outcome is that high value points occur as early as 14-14 +/- 2, and after 20-20 +/- 1 in sets 1-4. So actual crunch time is a bit different than we might imagine. And 5th sets are so volatile that virtually all points are ‘crunchy’. hahaha
Accounting for rarity – How can I think about crunch time if my team never gets to 20-20? It can’t be that none of the points we played are important. Maybe there are points early in the set that are important too. So the next method is the look at value AND frequency. When we look at it that way, crunch time occurs in the first six or so points and after 21. When you are serving with a deficit in the early part of the set, e.g. 3-5, those points are extremely valuable.
Timeouts – You all know that I love to talk about timeouts. Do coaches’ timeout strategies reflect the value of the point? The answer is no, especially at the beginning of the set. In the graph the white dots are the most valuable points and the colours are the timeouts. Coaches take their timeouts too late, especially at the beginning of the set. Perhaps the reason we find timeouts to be ineffective is because coaches don’t use them at the most important times.
As always, I am interested in your thoughts, comments and ideas.
Tagged Volleyball Analytics, Crunch Time, Timeouts, Science Untangled
The total of 82 practical Coaching Tips can be found here and here.
Read about the great new Vyacheslav Platonov coaching book here.
Reblogged this on CoachLife.