So this might be a thing only I care about but the current way most competitions are run by having people sign up to skill categories based on belts or time spent training seems suboptimal to me.
As far as I can tell most sports have some sort of league system where you can go up or down leagues, season to season, based on your performance. This would generally sort people into somewhat equal skill divisions, allowing for people of varying levels to compete and have fun doing so.
Now for my idea:
Likely not within this year but some time in the future I would like to host small local competitions and have been thinking on how to do the skill sorting more effectively. One system I've come across which seems quite suitable to our weird little sport seems to be the elo rating similar to how they do it in chess. (I won't try to explain it here but if you havent heard of it a quick google or AI search should give you an idea)
Now if we host regular competitions it should be relatively easy to implement this with a few spreadsheets, and we just might do that. But as I was thinking about it, I realized it would also integrate very easily into the Belt Checker app, of which I already am a huge fan. Simply program a little algorithm taking the claimed competition results from verified people on the platform and if both people agree on the result have it count and put a little Elo Rating on their profile.
I can come up with a number of arguments why this might suck and I can absolutely see their validity. But perhaps the community my like the idea, so I figured I put it out there.
Hmm, interesting thought, but I'm not sure it would have much value. It works in chess because even casual players can have plenty of matches to set their skill level. I want to compete, but haven't been able to do much.
I did the local tap out cancer last summer and had a great time. Masters white belt local comp was the perfect level for me to dip my toes into the competition scene and I went 2-2, narrowly missing out on the bronze in my last match. If I was looking for more challenge I would hit the road with some of the guys from the gym to an ibjjf regional. If that wasn't enough, I'd fly out to Vegas for worlds.
That being said, a little bit after the comp I got my blue belt (i.e. relegated to the next league up) and wanted to wait a few months to get it broken in before competing again. But in the fall I got a nasty shoulder injury which required surgery to fix. Now, it's been a year since I got my blue and I feel I need still more time before I put the joint to the test in competition. How can an elo algorithm capture all that?
"Hmm, interesting thought, but I'm not sure it would have much value. It works in chess because even casual players can have plenty of matches to set their skill level. I want to compete, but haven't been able to do much.
I did the local tap out cancer last summer and had a great time. Masters white belt local comp was the perfect level for me to dip my toes into the competition scene and I went 2-2, narrowly missing out on the bronze in my last match. If I was looking for more challenge I would hit the road with some of the guys from the gym to an ibjjf regional. If that wasn't enough, I'd fly out to Vegas for worlds.
That being said, a little bit after the comp I got my blue belt (i.e. relegated to the next league up) and wanted to wait a few months to get it broken in before competing again. But in the fall I got a nasty shoulder injury which required surgery to fix. Now, it's been a year since I got my blue and I feel I need still more time before I put the joint to the test in competition. How can an elo algorithm capture all that?"
This is a very fair criticism, and perhaps an Elo algorithm wouldn't capture all of that and the frequency of competition in chess absolutely helps.
My defense in this case would be the following.
Competing in a masters division of entry level Elo would, I assume, place you with similarly inexperienced opponents as the white belt masters placed you. (This is just a guess though, I haven't thought about this much)
You have now been moved up a skill division by your instructor and perhaps rightfully so but you haven't proved such in a competition setting, and, infact haven't had the chance to win something before moving up.
And as you said yourself you wanted to take time breaking in the blue belt before competing. Had you been competing under an Elo system you could have broken in the blue belt by competing at the novice level and proving your competition chops. No need to wait since you won't be moved up the competing level before proving you should be.
Perhaps if you had moved up in Elo and then had a shoulder injury you would be afraid to compete at that higher level and so I would even consider slowly draining elo for extended lay offs. (That comes with its own set of problems of course)
Perhaps my main point being, how does ranking by belt system capture all of that?
I really appreciate you taking the time to formulate these criticisms by the way.
I think the biggest problem is going to be getting accurate and relevant data. Results are easily enough found and confirmed, but finding out if they are really legit is a problem. Why wouldn't they be legit? I"ll use my wife as an example. She competed heavily from 2017-2019. I'm not sure how many tournaments she did during that time; maybe 20? Should be enough to see where you're at, right? I don't know what her record was, but she was holding her own. The problem is that during that entire time she was able to compete in her correct division one time and that was at Master Worlds. Every other tournament she was in she had to either go up 1 or more weight classes or down 1 or more age groups. Her results in those tournaments would not have looked impressive at all. Then she went to Worlds and finished second in her division. This really isn't much of an issue in the male adult divisions, but as you get away from that it becomes really common.
ETA: And it's also really difficult to accumulate matches when you get out of the adult male classes. If you win as an adult male you probably had 4 or 5 matches. Winning in other divisions you're probably only getting one or two matches.
"I think the biggest problem is going to be getting accurate and relevant data. Results are easily enough found and confirmed, but finding out if they are really legit is a problem. Why wouldn't they be legit? I"ll use my wife as an example. She competed heavily from 2017-2019. I'm not sure how many tournaments she did during that time; maybe 20? Should be enough to see where you're at, right? I don't know what her record was, but she was holding her own. The problem is that during that entire time she was able to compete in her correct division one time and that was at Master Worlds. Every other tournament she was in she had to either go up 1 or more weight classes or down 1 or more age groups. Her results in those tournaments would not have looked impressive at all. Then she went to Worlds and finished second in her division. This really isn't much of an issue in the male adult divisions, but as you get away from that it becomes really common.
ETA: And it's also really difficult to accumulate matches when you get out of the adult male classes. If you win as an adult male you probably had 4 or 5 matches. Winning in other divisions you're probably only getting one or two matches."
This is also a very fair criticism.
Perhaps accounting for weight differences could be done mathematically in the formula to calculate Elo, Im not well versed enough to judge that.
You didn't mention your wifes belt rank during all this so Ill assume it stayed the same. So if she competed in a certain Elo bracket from 2017 through 2019 and held her own there but didn't out perform everyone, she would also have stayed around the same Elo bracket come worlds and would have been able to prove her skill there. Not much changing. Between the systems.
Im gonna make a lot of assumptions based on nothing here but I would even argue there would be more competitors over all since people who moved up the belt ranks, but never performed very well in competition might still be competing since they would be able to compete closer to their actual skill.
I know a person or two who I think stopped competing because they didn't feel up for the task at their belt. But this is pure speculation.
"This is also a very fair criticism. Perhaps accounting for weight differences could be done mathematically in the formula to calculate Elo, Im not well versed enough to judge that. You didn't mention your wifes belt rank during all this so Ill assume it stayed the same. So if she competed in a certain Elo bracket from 2017 through 2019 and held her own there but didn't out perform everyone, she would also have stayed around the same Elo bracket come worlds and would have been able to prove her skill there. Not much changing. Between the systems. Im gonna make a lot of assumptions based on nothing here but I would even argue there would be more competitors over all since people who moved up the belt ranks, but never performed very well in competition might still be competing since they would be able to compete closer to their actual skill. I know a person or two who I think stopped competing because they didn't feel up for the task at their belt. But this is pure speculation."
She was a blue belt in the beginning and got promoted to purple in September of 2019. I think she competed twice at purple before covid ended the world.
