.png)
Rethinking How to Stat Certain Volleyball Situations
Introduction
I designed my own volleyball analysis system to accurately measure the technical skills of the players and the team. While I think my system is highly accurate (I did write 6,000 words worth of operational definitions) and is a good measure of the technical skills, there are still some inefficiencies with it. These inefficiencies are centred around an operational definition that makes a bad performance statted as a perfect performance, e.g. an attacker manages to get a kill from a bad set which means that bad set is turned into an assist. The other inefficiency is I can’t get the exact information from the volleyball analysis system, e.g. how can I measure how well a middle blocker can anticipate a set? Below are are other situations I have encountered that covers these two inefficiencies:
Consider These Situations
Situation: a receiving player correctly judges the service ball is going out of bounds. Current Stat: service error for the server. Issue: the receiving player should be rewarded for correctly judging the ball is going out because it is a part of your reception skill. Stat Change: possibly creating another stat but this would create more issues of where you would fit that in the reception rating.
I want 75% of serves to go to a particularly weak receiver, Player A. This is a common serving strategy. = Player A’s Total Reception Attempts/ Team’s Total Reception Attempts. the current performance indicator assumes Player A was available to receive for every reception but this is not always the case. If the strategy is working and the opposition coach is any good, he or she would take that particular player out of the reception line-up. This could be done in by substituting that player out or getting the opposite player to receive. This means the current performance indicator calculation underestimates the actual percentage of serves going to the weak receiver. have a stat of who was in the reception line-up. Then we can calculate the percentage of serves going to a player when that player was available to receive.
an attacking player hits the ball out of bounds but the second calls for a net violation on the opposition blocker. kill for the attacker, block error for the opposition blocker. it is a the operational definitions that there has to be a kill for every block error soit is only on a technicality that the attack error was turned into a kill. judge the attack for what it is and keep block error as separate. In this case, if it was out, then it is an attack error. If it landed on the floor regardless then it is a kill. If this was adopted, it does represent extra operational definition issues, e.g. what happens if a defender digs the ball and the second ball could be kept in play but play stops because the second ref blew for a net violation. You are trying to stat what happens and not what you think/ should happen.
a blocker blocks the setter sets the ball. blocker gets block error and setter gets a kill. similar to above. Due to a technicality, a setter who was not intending to attack, gets a kill. as above. Separate attack and block error.
I want to know the percentage of block attempts led to a block kill or block touch. = Block Kills/ Total Block Attempts. block attempts are defined when the player touches the attacking ball on a block. But what about the block jumps when they did not touch the attacking ball? They are still attempting to get the block kill but we don’t count it. have another stat called “block jump” when a block attempt but no touch of a ball.
the middle blocker has two jobs – attack and block. We have attack performance indicators but what about how well a middle blocker can anticipate where the set is going. nothing. even with the block jump stat, it doesn’t really say if the middle blocker is able to anticipate where the set is going. I do see novice middle blockers pretty clueless of where to block, e.g. trying to block the back row setter, but they might get two block kills and which we might regard as a decent performance. tricky one. I haven’t come up with anything good.
which outside hitter is the best attacker, especially when comparing across teams. Hitting % some players may have a better hitting % because they play on a better receiving or defending team. This is important because the more perfect reception passes and digs the team can do, the more easier for the setter to create 1v1 situations and stress the opposition block and defence, the more likely for outside hitters to get kills. separate hitting % into after perfect reception/ dig and also after good reception/ dig.
I want to know how well a hitter attacks after a set nothing there are other attack attempts that can cloud a how well a player can attack after a set, e.g. kills from an opposition overpass, free ball or overpasses that results in a kill, second ball attacks when the ball cannot be set. calculate hitting percentage only after a set.
the setter sets the ball a little tight to the net, the attacker spikes the ball and the blocker block kills it. set attempt, attack error, block kill the fault lies in both the setter and attacker but the error solely gets charged to the attacker only. I don’t really have anything good yet.
setter sets a terrible set too tight to net/ way beyond the antenna/ so bad it could be volleyed over but the attacker still manages to get the kill. Assist, Kill so what was a poor set has now been upgraded to an assist. we could use this setting grading.
I want to measure setter – middle attack connection. middle player’s hitting percentage. a middle player’s hitting percentage may be very good but rarely used (I have written previously how middle players should be utilised more). However, a hitting percentage is only when the ball gets to them. What about the times when the ball doesn’t, e.g. set too low/ high, middle player’s attack approach timing is off etc? I don’t really have anything good.
Not Everything That Counts Can be Counted
While the above is true, I do believe with an innovative analysis system we can measure many of the important aspects of the technical performance. This is why I obsess over the little things like these.
Comments