##Disclaimer My question **isn't** about how often the FIDE ratings are updated. I know that's done once a month. --- Are ratings adjusted match after match, or after every tournament? Even if the ratings are updated once a month, does FIDE count every match, or a whole tournament? ##An example to make my point clear Consider a three-player, single round robin tournament. **Player** **Elo** Roanin Novel 2798 Rigi Hains 2745 Dan H. Yvasin 2784 #Results **Round 1:** R. Novel (1) beat Rigi Hains (0) **Round 2:** Rigi Hains (0.5) drew with D.H. Yvasin (0.5) **Round 3:** D.H. Yvasin (1) drew with R. Novel (1.5) ##My question If FIDE was calculating the players' ratings, would it first update the players' ratings **match by match**<sup>1</sup>, or would their rating be adjusted by **counting the tournament as a whole?**<sup>2</sup> <sup>1</sup> In this case, rating changes after every match. <sup>2</sup> Here we calculate the expected score of each player at the start of the tournament, and compare it with the **final score only.** No rating changes occur between matches. ##The calculations (all assume K=10) ###Method 1 [Match by match] (Final ratings in bold) - After the first match, R. Novel is rated *2802.24* and Rigi is rated *2740.76.* - After the second match, Rigi is rated **2741.38** and Dan is rated *2783.38* - After the last match, Dan is rated **2783.65** and Novel is rated **2801.97** ###Method 2 [Whole tournament] 1. Roanin Novel - Initial expected score = 1.096 - Actual score = 1.5 - Hence, new rating = **2802.04** --- 2. Rigi Hains - Initial expected score = 0.868 - Actual score = 0.5 - Hence, new rating = **2741.32** --- 3. Dan H. Yvasin - Initial expected score = 1.036 - Actual score = 1.0 - Hence, new rating = **2783.64** --- ###Conclusion - In such a short tournament, there doesn't seem to be a significant difference in the results of the two methods, but maybe there **could be one in a longer tournament**. - Which of them does FIDE use? --- ##Note - If something is unclear, please ask in the comments.