Connection ErrorBad RequestThis page went staleSession ExpiredSign in requiredForbiddenNot FoundToo Many RequestsServer ErrorBad GatewayService UnavailableGateway TimeoutHTTP Version Not Supported
Session Expired
Your session has expired. Please log in to continue.
Sign in required
You need an account to do that. Sign in or create one — it only takes a minute.
This page went stale
Your security token was rejected, usually because the page sat open too long or you signed in elsewhere. Nothing was saved. Reload the page and try again.
Request Failed
We encountered an issue while processing your request.
Please check your internet connection and try again.
Error Details:
Share Link
Posts from Player rating for tournaments across all installments of Heroes.
I don't understand why this system isn't suitable for calculating Grail points? It takes into account both the number of participants and their results. And fixed thresholds are also set. The general table is simply to avoid having to browse through dozens of links in search of your results. After all, it will contain not only the overall rating, but also the rating for individual parts, the number of games, the number of prize-winning places, etc. This is simply to simplify the search; it does not obligate anyone to do anything.
No one said that this system is not suitable for awarding Grail points. On the contrary, it is very suitable. But who will do the calculations? Find a chief coordinator, or will you be one? The trackers for each version of Heroes will send you rating tables for each of their versions. You will combine them into a single statistical table and award Grail points. Start doing this, for example, with the Kanighem tournament. Let VDV award Grail points according to its own system in its tournaments. This will reduce your workload :)
Added 3 minutes ago Sage;60760
I can take on one version, or maybe even a couple of versions.
Well, now let's listen to the head of the transport department; he hasn't spoken in a while :), what do you say, AmberSoler? Give your consent, and let's start dividing the versions.
ATTENTION! An example of the influence of the second component in the rating formula, which depends on the place taken in the tournament.
ATTENTION! An example of how the second component in the rating formula affects the result, which depends on the position occupied in the tournament.
I reviewed your calculation method. It is correct. However, it is impossible to call its application conditions a "solution to the problem in a general form." And the caveats "all the same conditions applied, but the leaders didn't show up again" suggest that the specific example is contrived and tailored to achieve a desired effect.
For any tournament table, it is possible to simulate conditions under which its result will cause confusion among "spectators." However, these spectators still need to be explained what the probability is of exactly THAT scenario occurring. The scenario you proposed has a non-zero probability, but it is vanishingly small. :)
Now, I will refer you to my post #193 in this thread, where the situation is analyzed in detail, and it has a significantly higher probability of occurring than the one you indicated.
It is very easy to simulate it in the table that you have in front of you. Let's assume that all players participated in the second tournament, but the leader managed to complete the map much faster, and the others showed their results. Simply take my original calculation and assign the leader a map completion time of, for example, 5 days. And your first coefficient will give all the other players a rating in the range of 1-10. The underdog will be only 5.7 points behind your result... And if you add aging, your result will bring you 2.0 points for the tournament, and the underdog will get 0.4 points! :) All the other players will be evenly distributed between you. Calculate it yourself. This is probably much more correct in your opinion... And this scenario has a significantly higher probability of occurring. But what is the point of competing for such a rating?
And one more thing. I believe that while you and I delve into the depths of the calculation, the others will either wait or, if they get tired, give up. No one will bother to understand, and our dispute will simply cause confusion, and sooner or later, the phrase will be heard - "to avoid disputes, let's leave the simplest calculation option, which everyone understands and does not cause doubts in anyone"... In fact, I think that no option causes doubts in anyone, and only you and I have created the illusion of the importance of this issue. :)
Calculate as you see fit. In any case, I will not be involved in conducting rating calculations. So, the ball is in your court. However, if you need my help, I will not refuse.
Everything I wanted to say about this, I have said, and more than once. Reducing the "significance" of the second component is incorrect. I have already thought about this. Second place cannot have less significance if the leaders did not show up for the game. Otherwise, the result of any tournament can be questioned and called a fluke, considering that there are players somewhere who could have beaten everyone, but "simply didn't show up"... Each specific tournament is considered separately and regardless of whether someone did not show up.
The players who overtook you in your calculation deserve it - they took high places in the game that you did not participate in. And do not confuse this argument of mine with my previous statements about not showing up for the tournament. There, I spoke about the inadmissibility of a player who played "poorly" rising in the rankings due to your absence. There, this opportunity was provided to him by the high degree of aging (1/3), which you proposed, and it was not the player's merit. Here, the player took second place in the tournament, and for this, he should have the opportunity to move up in the ranking table. Your option will not allow him to do this, and the player will not receive the deserved reward. Therefore, your option introduces an imbalance.
My task is solved in a general form, and for the average tournament, it is more acceptable... IMHO, as always.
I've said everything I wanted to say about this, and I've said it more than once. It's incorrect to diminish the "importance" of the second component. I've already thought about this. Second place cannot have less significance if the leaders didn't show up for the game. In that case, the result of any tournament can be questioned and called a fluke, considering that there might be players somewhere who could have beaten everyone, but simply didn't show up... Each specific tournament is considered separately and regardless of whether someone didn't show up. My task is solved in a general way, and it is more acceptable for the average tournament... IMHO, as always.
By the way, in many tournaments (not just Heroes), not participating is automatically counted as a loss, so the option proposed by AmberSoler most fully takes into account the interests of both the first and second place winners... Sincerely.
I reviewed your calculation method. It is correct. However, the conditions for its application cannot be called a "solution to the problem in a general form." And the caveats, "all the same people were invited, but the leaders didn't show up again," suggest that the example is contrived and tailored to a specific effect that you desire.
I wouldn't say I'm tailoring it to a desired effect, but rather trying to test the system specifically in extreme, critical situations. This is what testing anything is all about. Let's say, in the example provided, not all of the players I banned showed up for the tournament :) But only three. What would happen? Do I have to calculate that again? I already have one example where the test parameters were specifically increased.AmberSoler;60858
Now, I'll refer you to my post #193 in this thread, where the situation is broken down, and it has a significantly higher probability of occurring than the one you mentioned.
Why are you so sure that the probability of this situation is higher than the one I presented? In the example I provided, I guarantee that Dmitry will not complete this map in 5 days. And a significant gap between him and the majority of players could only occur if the players I mentioned did not show up. And in the example, we see that a non-linear rating would show a more accurate rating of the participating players. After all, we know the real situation because, thank goodness, we were able to complete the map together with Grandpa Mozai :)AmberSoler;60858
It is very easy to simulate this in the table that you have in front of you. Just take my original calculation and set the leader's map completion speed to, for example, 5 days. And your first coefficient will give all the other players a rating in the range of 1-10. The outsider will be only 5.7 points behind your result... :) This is probably much better in your opinion...
Absolutely correct!!! The rating will be accurate because I won't see many players at this tournament. Aiden, Harvey, and many others. But if no one shows up, and I suddenly become "cool." But at the next tournament, if a cool monster or monsters show up, they will quickly put me and everyone else in their place by completing the map ten times faster than us. And the proposed example for modeling only confirms this.AmberSoler;60858
And one more thing. I believe that while we are delving into the depths of the calculation, ... Calculate as you see fit. In any case, I will not be in charge of rating calculations. So, the cards are in your hands. However, if you need my help, I will not refuse.
We won't delve into it :) I already have a solution. I will have to keep the rating for VDV tournaments in Heroes using two systems to REALLY assess which one is more suitable for Heroes. I know the main points and formulas of yours. So, let's do this - with your system, everyone gets a 10% aging penalty - will there be a bonus or not? With my system, there will be a bonus of 1-2 skips. But the aging penalty will be 33%. And this - since you won't be calculating, but you won't refuse to help, prepare the forms or use the template. I got tired of coloring the example - I don't really know Excel. Therefore, in addition to the forms, I need instructions on how and where to enter the data to calculate the next tournament, because you saw what I'm capable of - no more. Then you wanted to add some more coefficients later, will that be later?
Sage! I'm taking Heroes 3, maybe even the first one. What will you take? I can't take more. I need to keep the rating for two systems in parallel. Will you choose the system you will use? I can't choose yet, so I will find out in practice? AmberSoler;60858
Everything I wanted to say about this, I have said and more than once. It is incorrect to reduce the "significance" of the second component. I have already thought about this. Second place cannot have less significance...
You thought poorly. The place in tournaments in Heroes should not have such a strong influence. And certainly not dominate over the component that is formed by the approximation of the result to the leader in days. The significance of the SECOND "free" place does not decrease. Let the player be recorded in the Hall of Fame, and let them be showered with full pockets of Grails. AmberSoler;60858
if the leaders did not participate in the game. In this case, the result of any tournament can be questioned and called a fluke...
The main thing is that when they show up - at the next tournament - they will dispel our doubts about the fluke results of the past, when they will outperform everyone by tens of times and immediately get a worthy rating for it, and with your component, how long will they have to prove and catch up with the supposedly cool ones who have accumulated a ton of points over five years? So what if someone worked on their rating for a year, but can't play? A monster comes along who can play and immediately restores the status quo with their super-fast completion times, far ahead of everyone else. Why should he spend a year if he can play? And if he leaves, he will also quickly fall off the rating :) We are talking about the CURRENT rating - have you forgotten?AmberSoler;60858
... if we take into account that there are players somewhere who could beat everyone, but they just didn't show up...
And does this mean that my game rating should become better if I still take a long time to complete maps compared to even the players on this resource? The point is that this system, in the long run, for the GENERAL rating, both between resources and for different versions, but let's not talk about that yet. They didn't show up, so what about your statement that you need to defeat a bunch of Masters to achieve something? A player who takes a long time to complete maps cannot have a high rating, regardless of what place they took. He will only be lucky if he takes first place :) And the same goes for everyone else :) But at the next tournament, speedsters can throw them so hard that it will be noticeable :)AmberSoler;60858
Each specific tournament is considered separately and regardless of whether someone did not show up. My task is solved in a general form, and for the average tournament, it is more acceptable... IMHO, as always.
Therefore, if someone did not show up, you will get a bag of Grails for second place. So, second place will be rewarded. The monsters who didn't show up missed out on a bag of Grails. But what does this have to do with the game rating? A monster doesn't show up - his rating sharply decreases - he returns - he plays well - he will quickly catch up with the imaginary leaders by making super-fast completions, far ahead of everyone else. Do you get the idea? So, what you want to present as a disadvantage of this system is actually its advantage :)
Added 2 minutes ago Kyle;60861
By the way, in many tournaments (not only in Heroes), non-participation is automatically counted as a defeat, so the option proposed by AmberSoler most fully takes into account the interests of both the first and the second... With respect.
Maybe there are super-cool players who don't even know about this resource! Let's not look ridiculous if they accidentally drop by here :)
... Maybe there are some super-cool players who don't even know about this resource! Let's not look ridiculous if they happen to stumble upon it.
That's already funny... what do "super-cool" players have to do with it if the ranking isn't for all players at all, but only for players on our portal? And even if they come (which would be very good), in what way would we look ridiculous (please explain)?
Well, this is getting ridiculous... What do "super-cool" players have to do with it if the rating isn't even for all players, but only for players on our portal? And even if they do join (which would be very good), in what way would we look ridiculous (please explain)? Sincerely.
It's very easy to provoke me, but I'll try to be more polite. A super-fast monster arrives. He looks at the results of tournament runs. He looks at the rating and the formulas. He figures out how quickly he can climb to the top of the rating, assuming he completes maps ten times faster than everyone else, exaggerating. Let's say the regulars have very good current ratings, accumulated over years. And he realizes that he'll have to spend a year with his kids, a year taking first place in tournaments to prove that he's incredibly good - is that right? And if it's clear after the first tournament who's who? However, the rating, as a reflection of the situation based on the places taken, will reflect this lazily. Therefore, the second component will strongly affect the player who took, for example, second place, even though he is much slower than this monster who takes first place with a super result.
PS. And also, if you haven't noticed, the oscillator hits the underdogs hard!!! See the file I uploaded.
Damn - the underdog's already small rating is halved. AND WHY? And again, the second component from the place taken plays an active role in this. At least explain this point to Izual, not to me :) With this system, the player in last place will always get it the worst :)
Added after 15 minutes AmberSoler;60858
I looked at your calculation option. It is correct. However, the conditions for its application cannot be called a "solution to the problem in general." And the caveats "all the same conditions apply, but the leaders didn't show up again" suggest that it's an artificial and contrived example tailored to a specific and desired effect.
You'd better model the same absurd situation using my rating calculation scheme and show the shortcomings in numbers, and make a clear conclusion. This will be beneficial.
I could certainly argue some points, but I won't, as I believe it's just pointless arguing... I'll note that if you look at it from your perspective, then yes, you're right, but if you look at it from mine, then I'm right... which means that there is no absolutely correct ranking! And since you've taken on the responsibility of keeping track, I don't have the moral right to argue with you; as they say, "The ball is in your court."
I could certainly argue against some points, but I won't, because I think it's just beating a dead horse... I'll note that if you look at it from your perspective, then yes, you're right, but if you look at it from mine, then I'm right... which means there's no absolutely perfect ranking! And since you've taken on the responsibility of keeping track, I don't have the moral right to argue with you; as they say, "The ball is in your court." Sincerely.
That's the wrong approach. You not only don't have the right, but you ARE OBLIGATED to use numbers and examples to show the shortcomings of the system I proposed, if you plan to actively participate in Tournaments based on the game rating on this resource! And I suggest that you start by explaining to everyone why the player who comes in last has their rating halved. To "kick them while they're down," so to speak. The proposal to give a minimum score to the ten lowest-ranked players was also not accepted. However, a non-linear rating doesn't set a minimum, nor does it halve the rating of the lowest-ranked player. And this is already a small advantage of the ranking system I proposed. What do you say? All I can say is that the ranking doesn't care who looks at it from what perspective. It should reflect the current state of affairs and the strength of the players participating in the Tournaments, quickly reacting to both very good and bad results that a player has achieved in the current Tournament, removing players who have not participated for a certain period of time or have disappeared altogether from its table, and, upon their return to the resource and Tournaments, if they show very good results, quickly restore their rating to match their playing strength.
It would be better if you modeled a similarly absurd situation according to my rating calculation scheme and showed the shortcomings in numbers, and made a clear conclusion. This would be beneficial to the matter.
I have already modeled it when one strong player resets the rating for all the other 100 participants in the tournament... Isn't that enough for you? :)IronAxe;60889
Incorrect approach. You not only don't have the right, but you ARE OBLIGED to show the shortcomings of the system I proposed in numbers and with examples...
But HOW?, if IronAxe;60102
...the numbers are fair, but I don't know what to do with them, I'm not good at math :(
IronAxe;60441
Well, I said I'm not good at math :) I didn't take into account that...
Forgive me for the quotes, but even simple mathematical calculations cannot demonstrate to you the inconsistency of your beliefs. And you claim to want a detailed analysis "in numbers"... I can, of course, do this, but I'm afraid it will go unnoticed, like everything else... :)
So, what's the point of talking in vain if my arguments, unfortunately, are not considered by you from a mathematical point of view? :D It is mathematics that ensures balance; you appeal to selective observations, as I have already said. It is not correct to do this. And that's all, let's put an end to the dispute due to the lack of meaning. :)IronAxe;60889
The proposal to give a minimum score to the top ten outsiders was also not accepted.
In my opinion, it was you who were against this option... :D :D :D And the reason for your disagreement was that the outsider was given a little "free" rating. And now you are indignant...IronAxe;60889
...that the rating of the outsider is halved for last place. To finish them off, so to speak, with your feet :)
It is halved for the same reason that the rating starts to increase for a higher place... :) The oscillator adjusts the result depending on the place taken. Last place is always worse than any other place. Your coefficient simply doesn't realize this, which is a pity... :D IronAxe;60880
I already have a solution. I will have to keep a rating for VDV tournaments in Heroes using two systems in order to REALLY assess which one is more suitable for the game. I know the main points and formulas of yours.
Okay. Your decision to conduct two ratings for the sake of the experiment is commendable. But here, the main thing is not the correctness of the calculations, but the correct interpretation of the results obtained. I hope you have enough understanding to avoid making mistakes, because, after all, you can argue endlessly.
I suggest suspending the discussion about balance until significant and statistically reliable tournament results are obtained. Is that okay?
I completely agree that arguing is pointless now... we just need to choose one system and stick to it... but maintaining two systems, in my opinion, is like giving two starting saves... you can do it, but how do you reconcile all of that later?
I already modeled it when one strong player resets the rating for all the other 100 participants in the tournament... Is that not enough for you? :)
If this is your strongest argument :) I will make 2 decimal places for everyone and multiply by 100. I'm exaggerating again, but the idea should be clear. If such a situation arises, I will find a way to get rid of the zeros.AmberSoler;60899
BUT HOW?, if Forgive me for the quotes, but even simple mathematical calculations cannot demonstrate to you the inconsistency of your beliefs. And you claim to receive a detailed analysis "in numbers"... I CAN do this, but I'm afraid it will go unnoticed, like everything else... :)
I provided you with the calculations :) And you are just trying to nitpick at the words. Or you are avoiding uncomfortable questions and not giving answers to them. I know this trick :)AmberSoler;60899
So, what is the point of talking in vain if my arguments, unfortunately, are not considered by you from a mathematical point of view? :D It is mathematics that ensures balance; you are appealing to the selectivity of observations, as I have already said. It is not correct to do this. And that's all, let's put an end to the dispute due to the lack of meaning. :)
Yes, I already understand. It's easier to come up with multi-layered formulas than to calculate using them. And it's still unknown what will come out of these formulas :)
Just look at the columns of the outsiders, what comes out for them according to your scheme and what comes out according to mine. Izual doesn't even need to try - according to your scheme, he has already broken away :) And according to my scheme, he has a chance and hope to overtake at least the second to last. Will you answer this question from a mathematical point of view with a formula, or will you offer two answers: 1. I won't understand 2. Let's set a limit of 20 people
Not only I was against the limit; if you can quickly find my quotes, I think you will also find these :)
Besides the correct mathematical formulas, I don't see any logical connections in your system. And I already showed you with an example that they work incorrectly. Show me the shortcomings of my system in that Excel table, based on the first tournament, even with the infamous zeros. And then I will give a specific answer. When you started talking to me about specifics, I spent the whole day today coming up with an example. And when I tell you to now specifically show something, you are just saying that I am a layman in mathematics, although I don't argue, but I would like to see your example in Excel :)
I completely agree that arguing is pointless now... we just need to choose one system and stick to it... Maintaining two systems, in my opinion, is like giving two starting saves... you can do it, but how do you reconcile everything later?...
Maintain the ranking based on the system that will be adopted as the main one. The second option is to keep a separate one for personal use, so that in the future, all the issues raised by this dispute can be resolved.
In my opinion, it was you who opposed this option... :D :D :D And the reason for your disagreement was that the underdog was given a bit of "free" rating. And now you are complaining...
What "free" rating for the underdog? It was about a threshold for a bunch of underdogs. Then you started talking about some kind of relationships, but Sage already disagreed with you. I even gave you examples of aging, like with your 10%, and how those who missed the leaders with a handicap overtake them. And with my 33% aging, they don't pull ahead as much as with your 10%. What's wrong there, give an example :)
Guys, you've posted over 200 posts in just four (incomplete) days, and at this rate, it could continue... :smile15: Based on this, I suggest organizing a poll (preferably an open one) with a duration of, say, a week (so that everyone who is interested has time to vote) and choosing one of these two options... I think further discussions are unnecessary! :smile38:
Besides the correct mathematical formulas, I don't see any logical connections in your system. And I've already shown you with an example that they don't work correctly. Show me the shortcomings of my system in that Excel table, based on the first tournament, even with the infamous zeros. And then I'll give you a specific answer. When you started talking about specifics, I spent the whole day today coming up with an example. But when I tell you to now show me something specific, you're just talking nonsense that I'm a math profan, although I don't argue, but I would have liked to see your example in Excel.
I've already told you that your calculations are correct. But your interpretation, in my opinion, is incorrect. We're not arguing about the correctness of rearranging numbers and coefficients, but about the validity of taking into account the player's position.
You say that the position doesn't matter. I say it does. How else can I show this with numbers? It's already clear from our calculations. You don't want the position to matter and you're surprised why in my calculations, the rating of the underdog is halved and slightly increased for those who are higher in the table. What's so surprising about that, if that was the goal when writing the oscillator - the higher the position, the more significant the bonus...
What should I give you as an example in Excel? What? I'm talking about logic and common sense in evaluating a player's result. Not about numbers.
IronAxe
