After reading some opinions about the most recent figure skating controversy, that is, the scores of Patrick Chan and Dennis Ten, I had a lot of thoughts about what is going on with figure skating rules, judgement and fan community, that we see controversies like this fairly regularly. It made me recall a certain well-known problem from the realm of software engineering, namely the problem with quality of software.
Basically, the whole field of software engineering is struggling for years to improve the quality of produced software with all sorts of different methodologies that are supposed to reduce the amount of bugs. The most famous methodology of that type is UML that was a really huge thing in 2000s, but there are many others such as “extreme programming” and so on. Most of those methodologies basically say the same – “follow us, and your software will be excellent”… And yes, those methodologies work well for the people who invented them, but once released to general public we inevitably see projects failing under those methodologies.
Then there is a view from Frederick Brooks, one of the best software engineers in the history, which basically consists in that there is simply no such thing as a perfect methodology that can solve this sort of problems. There are simply good and bad developers. Some methodology may work much better for some developers than others, but it’s up to a particular developer to decide which method works better for him or his team. The “werewolf” is a well-known metaphor for this problem of software engineering and “silver bullet” is a metaphor for methodology that is supposed to solve it. In my personal opinion, there is no such thing as a silver bullet.
This may seem like a topic that has nothing to do with figure skating, but personally I see a lot in common between this problem and problems that many people have with IJS. The reason is very simple – IJS is basically a methodology that tries to evaluate a performance in numeric terms. A performance, however, is not something that can really be evaluated in numeric terms. IJS tries to give numeric values to everything that can be evaluated objectively, such as types of jumps, but there are also more subjective things like quality of certain elements (evaluated with GOEs) and PCS. They try hard to reduce subjectivity with guidelines, but at the end of day people either don’t know about them or just refuse to apply them. As the result, we always end having some sort of controversy every season.
And every season fans make all kinds of proposals to change IJS, so that it can be more “fair” or to overcome “biased judges”. ISU mostly ignores those proposals… And I’m really grateful that they ignore them, because most of them, frankly, are absolutely mindless and, if they were applied, figure skating would cease to exist as a sport that makes any sense. And since we are talking about werewolves, I hope that people will realize one day, that sadly, it’s just an inherent problem in figure skating, that the performance we like the most is not necessarily the best. We have a system which is far from being perfect, but it was made by people who know figure skating better than the vast majority of fans and, more importantly, people who judge competitions also know figure skating better than most of fans.
There is no such thing as silver bullet, in this case, a perfect rules system that will appeal to everyone and will have no flaws at all. All intents to create such a system will miserably fail at some point. What we can do, however, to really improve figure skating, is to become better judges ourselves. This is the job that fans should do – mostly fans are horrible, horrible judges. And we can do this by understanding the rules, appreciating different types of skating and different types of skaters and becoming aware of the history of the sport. If we appreciate skaters other than our favourites and we can place them higher on the podium than our favourites if they deserve so, then we really are good judges and good fans of the sport. And when we are good fans of the sport we can actually start making proposals that ISU can listen to because they actually make sense. But this is not how things are currently.
So, what I’m saying is not that the current system should not be changed… It should be changed and it surely can be improved, but before making propositions to change it, one has to be sure in its own competence and be able to back up any change with good arguments and knowledge.
So, what can be improved in the current system? I personally have seen a lot of proposals, but most of them, in my opinion, are missing the point. Since we are dealing with subjective evaluation, I believe that any change in the system should be about reducing the uncertainty of evaluating any given aspect of performance. In other words, it should be as clear as possible how many scores should be given for a certain aspect of performance that has X, Y and Z qualities. That way not only it will be easier to understand the scores, but it will be actually easier to judge and, therefore, the judgement will be more objective and more consistent.
As an example of a change that reduces the uncertainty, I can propose the following. This is just an example and I’m not saying that this should be really done. Suppose we want to reduce the uncertainty in PCS, the most subjective and controversial part of rules. Currently the scores are in the range from 0 to 10 with granularity of 0.25 points. This granularity is very precise, if you think about it. It is actually so precise that it may not make a lot of sense. Think about it – I can, probably, explain what it is the difference between skating skills evaluated with an 8.00 and skating skills evaluated with an 9.00 and, most likely, this difference will be visible to many fans. Now, I personally would struggle a lot with trying to distinguish between 8.50 and 8.75. Does this 0.25 difference really make sense?
So, imagine that we reduce granularity to 1.00… Effectively, we reduced the uncertainty, because once we determined that certain skating skills can be evaluated up to 8.00, but not to 9.00, we don’t need to determine whether it’s 8.00, 8.25, 8.50 or 8.75 – it’s just 8.00. It’s possible that with this we effectively reduce certain “noise” from scores and make it much easier both for judges and audience. But I’m not saying that this is a right thing to do. It’s up to judges to decide whether they really need that sort of fine granularity as 0.25 or not. Personally, I think we could safely go for granularity of 0.50 points and I’m not so sure about 1.00 points. It may be a bit extreme. In any case, this is just an example.
On the other hand, if someone makes a proposition to start evaluating PCS with granularity of 0.01 points that would add a lot of uncertainty, because this would demand from judges to distinguish between skating skills of 8.37 and skating skills of 8.38. I’m not sure this makes any sense.
I’m not sure if I should talk about certain particular controversies because, frankly, this is all the same every season. This time it’s Dennis Ten/Patrick Chan, but in a year or so, we will see another one with different names. Once again there will be a bunch of more or less crazy proposals to change IJS, talks about “weak” and “strong” federations, maybe some more petitions… Instead I prefer to point out things that are inherent and will remain in time regardless of details of some particular controversy. What I hope for, however, is that eventually the skating community will seriously change for good, because, frankly, figure skating deserves a better community.
Recent Comments