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 DISCUSSIONS on “Rules for online games in HoMM3 and formulas for calculating player rating.”Auto-translated
My only wish is to add statistics and win rates for each player and castle in the future.
This is already in development. KobRa_RR
At first glance, Maretti is out of competition with a win rate of 85.6%, which is what I always assumed. In addition, there are many other worthy players with a win rate of 70+% and a sufficient number of games played (300+).
Alex has the highest win rate, but Maretti's consistency is impressive. :smile11: matador99
It feels a bit unusual that there is no dynamic change in the rating per game (columns p1 Overall rating p2 Overall rating p1 Season rating p2 Season rating in the previous version).
I was also thinking about this. But the Creator has decided to replace them with graphs in the profile for now. Maybe we'll add them later.
Not all players made it into the new ranking, for example, kamukag3e. He is already in the old ranking (21 games, 1609 points, 82nd place), but he hasn't made it into the new one yet, although the new one should be updated more frequently.
And another suggestion: make it like in the old ranking, so that for each game, it would be visible how many points its result added/subtracted.
"There's no need to execute me,sir Davos . I'll be dead before dawn." ICQ 716748479
I'm also not in the new ranking. At the time it was released, I hadn't played 20 games, and I think that's the problem. The ranking doesn't add new players. And is there a way to create a filter (all players, played more than 20 games, played more than 100 games) like the filter for "active/active + inactive participants"?
I'm also not in the new ranking. At the time it was released, I hadn't played 20 games, I think that's the problem. The ranking doesn't add new players. And is there a way to create a filter (all players, played more than 20 games, played more than 100 games) like the filter for "active/active + inactive participants"?
The ranking on the main page of the forum hasn't been updated for 2 weeks. The same ranking, at the old address (http://heroes3stat.tk/rating), works perfectly, use it.
The rating on the main page of the forum hasn't been updated for 2 weeks.
Hmm... That's strange. It says there that it updates, almost every minute. New reports are taken into account in the "details." In any case, thank you for the notification. We will investigate.
Added 13 hours 37 minutes later korovka1987
I'm also not in the new rating. At the time it was released, I hadn't played 20 games yet, I think that's the problem. The rating doesn't add new players.
I think that's exactly the problem. We'll fix it. It does record the statistics, but it doesn't display it in the table.
Regarding the rating update, I checked, and compared to yesterday, everything has been updated. It seems to be working. But if anyone notices anything, please let us know, and we'll fix it. Antiplayer
It's strange that the rating on two sites calculates using the same formulas and the same source data, but the results still differ by a few points... Why is that? I still don't understand...
....... It's strange that the rating on two websites uses the same formulas and the same initial data, but the results still differ by a few points... Why is that? I still don't understand...
Maybe it would be better to see here the points won or lost for every game, like on the other site. I think we can see after that why the rating is different.
Another thing is about the season; it hasn't been updated to the new season, and there isn't one here.
Not all players made it into the new ranking, for example, kamukag3e. He is already in the old ranking (21 games, 1609, 82nd place), but he hasn't made it into the new one yet, although the new one should be updated more often.
And another suggestion: make it like in the old ranking, so that for each game, it shows how many points its result added/subtracted.
I am not in this ranking, although I already have 20+ games with reports.
It's strange that the rating on two websites uses the same formulas and the same initial data, but the results still differ by a few points.... Why is that? I still don't understand...
I noticed several problems/inaccuracies in the new rating: 1) It records games in the log, but doesn't calculate them. For example, my last 5 games haven't been calculated, although they are in the log. You can see that they haven't been calculated on the graph below the log; it shows the "movement" of rating points - it's very inconvenient to use, but at least it's something. At first, I thought it wasn't calculating games from 2018, but no - for other players, it calculates some of the games from the new year. Games from the last 2/3 weeks are not being calculated, but I can't understand the logic/reason. 2) As an example, with korovka1987, it records and calculates everything, but doesn't display it in the General rating. 3) As an example, with Joker2k, the number of games is the same in both ratings, but the number of wins/losses doesn't match, i.e., some rating calculates some results with the opposite outcome. 4) For some "old" players, the number of games in the new rating is higher than in the old FTS - 382/378, herolwy - 472/440, Gomunguls - 905/787 (!!!) and so on. Probably, the new rating takes into account very old Tournaments that are not in the Nitro rating, which is also why there are small discrepancies when the overall numbers seem to match, but in one rating, for example, they gave 15 points for a win, and in the other, 14.
According to the old rating, I played 22 games, and in the new one, only 20. However, the last game was definitely included. Apparently, the games with noangel on 07/01/2018 (defeat) and with Shanti on 14/01/2018 (victory) were not included.
It's strange that the rating on two websites uses the same formulas and the same initial data, but the results still differ by a few points... Why is that? I still don't understand...
In the old rating system, if reports are written on the same day, it calculates them in reverse order.
I would suggest modifying the rules to highlight with color (even if not red) all the relevant points, not just 2.1.5 – this would be more user-friendly and reduce misunderstandings when rereading.
I would suggest modifying the rules to highlight with color (even if not red) all the points, not just 2.1.5 – this is more user-friendly and reduces misunderstandings when rereading.
For this purpose, there is a changelog at the end of the topic.
Dirty_Player
