Posts from Alternative rating.

Hello everyone.

Why is it so difficult to decide on the coefficient to award for tournament wins? Once you decide (if you decide), and the points are added for past tournaments, the ranking will be more objective, and the names of strong tournament players will appear at the top of the list. Also, you need to decide how many points to award for the places taken in tournaments. And whether to award them at all? After all, the winner will already receive a good number of points due to the coefficients. In any case, in my opinion, the winner of a tournament should not be awarded more than 100 points, which is about 7 wins in regular games. If you give more, it will be unbalanced.

Suggestion: determine the tournament rating based on the rating of the participants.
Award points depending on the place taken, or award/deduct points for wins/losses depending on the opponent's rating, i.e., for example, winning against the same player, but in tournaments of different strength, gives a different number of points.

Hello everyone.

Why is it so difficult to decide what coefficient to give for wins in a tournament? When (if) you decide this and points are added for past tournaments, the rating will be more objective, and the names of strong tournament players will appear at the top of the list. Also, you need to decide how many points to award for the places taken in tournaments. And whether to award them at all?

I am against awarding points specifically for tournaments.

I suggest a coefficient of 1.5 for tournament matches and 2 for the semi-final and final. The number of points awarded for winning a tournament should be slightly reduced to a reasonable level. If we differentiate between the levels of tournaments (their categories), then two levels seem sufficient to avoid overcomplicating things. For example, if the participant pool cannot be considered truly representative, the coefficient could be reduced to, say, 1.3 or 1.8.

To avoid making things too complicated...
The tournament rating is calculated (e.g., as the average of the participants' ratings).
The rating for each win/loss is calculated using the same formula as the official rating, but the calculation takes into account the average values between the opponents' ratings and the tournament rating (or another formula).

However, if fixed coefficients are used, what should they be? 1.5, 2, 3...? What are these values based on?

P.S. IMHO.

For example, a tournament victory is 1.5 times more significant than a seasonal victory.

For example, a tournament victory is 1.5 times more significant than a seasonal victory.

:smile02: Why not 1.4 or 1.6? Or 2?
Tournaments have different opponents. Accordingly, some tournaments are "stronger" and some are "weaker." And the coefficient is fixed. It's a kind of average.
However, it's better if, at least in some way, all tournament games are taken into account.

There's a much simpler system for tournaments... introduce a betting system! Since the regulations state in advance who is playing against whom and when, simply implement an automatic voting system. An expert group (or the organizer) sets initial odds for each player to win in each match, and then the players themselves can vote for one or the other, thereby slightly adjusting the odds. And, of course, the number of points earned in case of victory is multiplied by the final odds of the player's victory. For example, Round 4 of "Best of HeroesWorld": Britvo4ka vs numOne (error in the table) P1 = 1.75 N = 2.00 P2 = 1.9 (where P - victory, N - draw) Let's say Britvo4ka wins and receives 25 points according to the rating calculation, and we multiply it by the coefficient of 1.75 = 44 points.

The phrases are much simpler, and the betting system doesn't quite fit. The perception of a game's strength is very subjective and depends on the system itself. Moreover, in such a case, it's possible that a player might receive double or triple the amount, which is clearly unfair. Considering that a certain percentage of wins are not based on gameplay.

Why offer odds at all? It's both an additional way to attract players and a fairer assessment of their skills. Because in a season, some prefer to play against demonstrably weaker players or within a specific circle. A tournament offers a chance to test your level not only among familiar people but also among those who are leaders on other platforms.

On the other hand, it's reasonable not to offer too high of a multiplier. In my opinion, double the amount is the limit. I proposed fixed odds for simplicity and to attract people to tournaments. My proposal regarding them is described above.

For example, a tournament victory is 1.5 times more significant than a seasonal victory.

A conditional statement.

It's possible that it's 2 times, and it's possible that it's 0.5 times, depending on the tournament and the opponents.

Maximum - for the betting system, if you really want to calculate points for tournaments differently.

Against the coefficient system.

For example, I have the same attitude towards both seasonal and tournament games. I believe that tournament games should be included in the overall ranking along with regular games without any increasing coefficients (with the exception of the special Heroes of Might and Magic III tournament, which should not be included in the ranking at all). And as a reward for the tournament winners (1st, 2nd, and 3rd places), a small number of additional points can be awarded in the ranking.