At an NHSDLC tournament we give awards not only for advancing further in the tournament, but also to those with exemplary speaking style. We use speaker points from the rounds to determine which students will be the recipients of such awards. This blog will explain how speaker points are given and exactly how the recipients of speaker awards are determined.
Speaker points are excellent for how clearly and persuasively someone speaks in a debate. Debates are won and lost based on arguments, but speaker points are a way to be rewarded for style. Judges consider if the debaters compel their audience to listen to them and agree With their points. This is not a measure of your English ability - no part of a debate is but any speaker points are tied to your confidence in speaking and being persuasive while explaining your points.The same time as the above-
Speaker points go from 1-30, but the conventional amount of points debaters will see ranges from 22-30. Points below this range are reserved for special circumstances, like a 1 theoretically being someone who doesn’t speak at all. At the top of the range, a 30 is awarded for hearing a speech which is the best you could expect from a high school debater.
Public Forum debate is all about convincing the judge in front of you, and some judges are going to prefer different things, especially when it comes to style. For example, some judges have more tolerance for people speaking quick than others. The NHSDLC speaker scale Is adapted from name scales used in the largest USA tournaments, and you can see it below:Add a text paragraph here.
By convention we don’t go below 20, as scores below that are incredibly rare.
When we determine who the top speakers at an event are, we have to consider a few things. First, an average speech at a tournament is about a 25 or a 26. If we gave speaker awards to the students who are ranked in the top 10 total points, there would be a lot of ties. This would make the award feel less meaningful, because it might work out that 3 or 4 or 5 people would all receive an award for being second place speaker due to a big tie.
The second thing we need to consider is taking into account different judges subjectivity. Although everyone agrees on some main ideas of what makes a persuasive and convincing speaker, there are variances in personal preference. Just like some people may like sweet desserts and other like sour ones, judges have different ideas about what stylistic elements can help make a “good speech”. Similar to your teachers, some judges might also have a tendency to award points a bit higher than others. In order to account for this human variation, we don’t use a straight average to calculate top speakers.
To determine speaker rankings, we use a combination of a debater’s points, ranks, and opponents’ scores to calculate their position in the tab. We use a combination of points and ranks because it helps account for variance among judges.
Tournaments all over the world (including NHSDLC tournaments) use a set of formulas to determine speaker awards and avoid ties.
1. High/low points.
a. This is the score obtained by subtracting the highest and lowest points from a competitor's total. So if a debater had six rounds with a 27, 27.5, 27.5, 28, 28, and 29.5, his or her H/L score would be 111, which is the total speaks (167.5) minus the highest (29.5) and lowest (27) speaks.
b. The goal is to minimize the effect of outlier speaks when determining awards. This means consistency is rewarded, so debaters shouldn’t worry about having one bad round.
c. This avoids anyone's ranks being affected too heavily by any one judge who might give out very high or very low speaker points.
Once the debaters are ranked in high/low points, there might be some ties. In order to break these ties, we move on to the next statistic:
2. Total Speaker Points
a. It’s best to use high/low points first to reward consistently strong speakers and correct for a debater having just one bad round. However, in the event of a tie we move on to total speaker points because they are a more direct evaluation of an individual compared to rankings.
Once the ties from high/low are broken by using total points, it’s possible the rankings will now be finished. However, if there are still ties, we move on to using ranks to break them.
3. Total Ranks
a. The purpose of ranks is to help break ties in speaker points. After a debate, the judge will give each speaker in a room a rank with the the highest speaker as a "1" and the lowest speaker as "4." The judge must also break ties through ranks. So if all four debaters in a round were to get 28.5s, the judge must still rank them 1, 2, 3, and 4 to break the tie.
b. On average, those with a lower number in total rank will be the best in their room.
c. Unlike all other tiebreakers, a lower rank indicates a better performance (1st is better than 4th); in other situations, a higher score indicates better performance.
If there are STILL ties in the speaker rankings, we have to move on to another statistic. Up next is Opposition wins.
4. Opposition Wins
a. The sum of wins of all opponents. The idea here is to prefer those who faced tougher opponents to those who faced weaker ones. If two debaters both have a record of 2-2 and have tied speaker points and ranks, to break a tie it is fair to evaluate how hard it was for them to earn those wins.
b. If a debater lost 2 rounds to a team that won no rounds, they are probably performing worse than a team that lost 2 rounds to a team that won all their rounds. In order to account for this we sum up the win records of everyone that a debater faces and rank them accordingly. If you were defeated by really strong opponents with lots of wins, you will be prioritized over someone who is defeated by weaker opponents with fewer wins.
5. Judge Controlled Variance
a. Judge Variance (JVar) is a measure of how many more (or less) speaker points a debater receives from a judge, given the average of that judge’s speaker point distribution over the course of a tournament. This is a statistical analysis that adjusts speaker points on the basis of patterns in how the judge gave out speaks. For example, if a judge is really lenient with speaks and has a high distribution of scores (think 28-30 range) despite seeing average debaters, then Judge Variance will have a downward effect for all debaters he saw. If a judge is pretty harsh and has a low distribution of scores (20-27 range, or less), then Judge Variance score has an upward effect - so high scores in front of that harsh judge would be rewarded much more.
b. JVar attempts to compensate for the fact that a 28 in front of a lenient judge is much different than a 28 in front of a harsh judge - it's a way of normalizing disparate standards. In practice however, JVar is rarely used for anything but breaking multiple ties.
6. Opposition Points
a. Opponent points: The team speaker point totals for all opponents, calculated in the same fashion as opponent wins. Again, the idea is to give credit to those who faced tougher opponents even if they lost. However, at #6 in the list, this option is very rarely needed to break ties.
7. Random Number
A. Normally there are no ties at this point, however if two debaters are still tied the computer generated a random number and whoever's number is higher is given the higher rank.
B. This criteria is only used as a last resort in a case where even after every single metric we have just described is used, two debaters are still tied. This is almost ok happen, but it is good to have a system in place, just In case.
In almost all cases after share debaters by by high / low points, total points, and room ranks we have removed all ties. Still, very occasionally the additional criteria are needed.
If a team receives a bye, they have their speaker points and ranks averaged and receive the full benefit of a win. If a team forfeits (due to failure to appear or misbehavior) they receive zero speaker points, maximum ranks (remember in this category high numbers are bad), and the full disadvantage of a loss.
These tiebreakers should not give any debater a reason to change their speeches; your number one job is still to win the debate by using good arguments and analysis, and explaining yourself in as clear and convincing a way as possible. Questions about these standards can be directed to info@NHSDLC.cn.