Throwing skill - Why it is important for Dodgeball
The basic skills of dodgeball
There are few things in Dodgeball that are as easily identified and as obvious to influence game outcome as throwing and dodging ability. It constitutes the bulk of what one focuses on during training, it sets the best players apart from everyone else, and no amount of tactical skill can compensate for a lack of these skills. During international competitions, teams are always talking about the players that have the “greatest arm”, the “best catches”, and the “best dodges”, and prepare their tactics accordingly. These skills are palpable, visible, and easily measurable, and are an essential part of Dodgeball.
What determines these skills?
There are two main factors that influence throwing ability:
Precision
How accurately you hit your intended target
Velocity
How hard you can throw the ball
This ignores decision making ability, which is harder to measure and include in the analyses. I will show in later posts how these less obvious skills can be partly estimated and included in the analyses.
The case for improving precision speaks for itself, so I won’t say anything more about that. Regarding velocity, the faster you throw, the less time you give the defender to react to your shot and dodge, block, or catch it. However, it is more important than that. The closer you throw to your individual maximal velocity, the worse your aim gets. Consider two players with identical precision. Player A throws at a maximal 80km/h, player B throws 100km/h. Balls travelling over 80km/h are very hard to catch. To avoid throwing a catch, player A will have to throw their hardest, which invariably causes low precision, and 10 such shots may have a spread of about 1.5m around the intended target (such as the left foot of the opponent). Player B can throw at 80% of maximum, where average precision is very high, and instead throw all 10 balls within an area of 30cm around the foot, while having the same risk of throwing a catch. The hit percentages of the players will markedly differ at the end of the game.
Using this logic, it makes more sense to first improve precision. Increasing throwing speed will allow the player to throw with higher precision at a higher velocity, only if the precision is already in place. One can say that precision is a limiting factor and velocity is an enabling factor.
Dodging ability is a bit more complex, as there are many factors that influence it (such as athleticism, reactive ability, catching skill, tactical knowledge of the opponent’s skill and favourite place to aim etc.), and there are many ways in which it can be expressed (one can deal with an approaching ball by jumping, laying down, moving sideways, catching, or blocking). How a player defends will depend on the individual profile, skill set, and athleticism, but more so on the opponent. As such, players usually differ less with regard to this skill, and as you will see later, it also predicts less of net points scored and game outcome than throwing skill.
How does throwing relate to the game of Dodgeball?
Clearly, the more one throws, the more hits one can expect to make. The plot below – number of throws against number of hits (data from Euro2018) – shows a linear relationship with a line outlining the expected hit percentage. The slope of the curve represents the average hit percentage for the team as a whole, and will be 1 for a hit percentage of 100% (36% for this team). The further up and to the left the player is, the better. The distance from the line (also called the residual) will thus show how good or bad the player is at making hits, irrespective of the number of throws or their hit percentage (see second plot). This allows for comparisons to be made from game to game between players of different skill. For example, in the second plot, one can see that Female1 and Male1 underperformed equally, despite different hit percentages and number of throws.
Relationship between number of throws and number of hits. Statistics for the Swedish Mixed team from EuroDodge2018.
However, it is not as simple as the plot above makes it seem. Each match only has a given number of opportunities for throwing the ball at an opponent. If you only throw once during an entire match, you are likely to only throw when you have a very high probability of making a hit. For example, if there is a player 1m in front of you, without a blocking ball, running away from you, most likely everyone will have a 100% hit rate. But, the more often you throw (or are expected to throw, if you are the team’s main attacker), the worse the chances are that you will have to take. Even though the relationship for the team is linear, for the individual player instead it is curved, showing diminishing return of investment for each thrown ball.
In a team where the players know each other well, the players will all be close to the expected line, since everyone is aware of their skill relative to everyone else’s, and thus also the loss or gain that occurs if one is to take a shot. This is something that I will come back to for other stats as well, but in general, the closer the players are to the expected line, the better the players know each other, the better they play together, and the higher the efficiency of that line-up is. If a player over-estimates its ability, and throws 140 times and makes 20 hits, many ball possessions will be wasted. If it under-estimates, throws only 10 times and makes 10 hits, the result is a lot of wasted opportunities, and a lower total number of points scored by the team.
This relationship is made clearer when one plots the number of throws against the hit percentage as below (data from CEC2019), which shows the diminishing effect of throwing more. It can also show which player, given a set number of ball possessions for the team, should have been the one to throw more. One can see that the higher the hit percentage, and the closer one is to the line, the higher the point scoring efficiency (points scored per minute, illustrated by marker size).
In this case Male5 played it safe and didn’t take all the chances he had (something I refer to as ”shot-taking sensitivity”, which is a measure of how good a chance has to be before the attacker decides to throw, and will be outlined in a later post), resulting in a lower point-scoring efficiency. Simultaneously, Male1 over-estimated his attacking ability and threw the ball when the probability of a hit was less than expected. This wastes balls which a teammate may have made a hit with, and gives the opponents a chance to eliminate teammates. The end result is that the team as a whole had a lower attacking efficiency and scored fewer total points. This again illustrates that teams with more spread out data-points are less efficient.
Relationship between number of throws and hit percentage. Statistics for the Swedish Mixed team from the Central European Championships 2019.
To see how hit and defence percentages relate to point scoring efficiency and game outcome, read this post. In it, I also explain how that allows us to calculate an index for a player’s technical skill that will be predictive of point scoring ability.