### Solvers’ rating

The rating is an integer numerical value which shows the solving strength of a solver. A solver gains it by solving at two approved / rated tournaments.

A rating can be gained at the WCSC, ECSC, WCCC Open, or other tournaments which fulfil definite criteria.

Ratings are calculated after each tournament. A rating list is published four times a year. It is calculated on the basis of the results of solving tournaments completed (including the report) by the end of March, June, September and December. If a tournament starts in one period and is finished in another period, the ratings of solvers are taken from the rating list valid at the start of the tournament, and the tournament is included in the period when it is finished. In case that at the same tournament different tables are produced (like one extra for juniors) ratings are calculated only for the major tournament.

The Solving Tournament Manager (STM) software is used for the issue of the quarterly rating lists.

The use of the STM software is recommended for all rated solving tournaments and it is mandatory for WCSC and ECSC, as well as for their open solving tournaments and for ISC (central controllers only).

### Criteria for acceptability of tournaments

at which rating can be gained

R1 | The tournaments should be announced and reported to the responsible WFCC representative (committee) at least two months beforehand but not at all later than 7 days before the day of the tournament takes place. |

R2 | At least 10 solvers with a full rating must compete in the tournament. |

R3 | The selected problems should be originals or originally published problems more than 5 calendar years before the tournament. The selected problems should show a clear theme and a good level of quality and difficulty and should represent different styles. |

R4 | At least 12 problems of different types must be presented to the solvers. |

R5 | Problems for solving can be chosen from seven groups: twomovers, threemovers, moremovers, endgames, helpmates, selfmates and fairy chess problems. |

R6 | At the tournament at least 5 groups must be represented. |

R7 | No group can be represented by more than 3 problems. |

R8 | The tournament may be divided into several rounds and/or days. |

R9 | The correct and complete solution or cook of all problems scores 5 points each. Points for partly solved problems can only have a maximum of two decimal places. |

R10 | When fairy problems are included at the tournament it is necessary that the corresponding fairy elements are announced at least two months beforehand. |

R11 | All tournament documentation (name of tournament director, diagrams of problems, solutions, complete list with detailed results per problem and possible complaints, etc.) should be sent to the responsible WFCC representative (committee) within 5 days (10 days for multiple-locations tournaments) after the end of the competition. |

R12 | In addition the solving sheets have to be kept by the tournament director for at least 4 weeks to enable probing possible issues. |

The representative (committee) has the right to confirm or not the acceptability of the tournament and determines the tournament coefficient. The final decision is approved by the WFCC.

### Criteria for gaining norms

N1 | At least 15 rated solvers from at least 3 countries with a full rating must compete in the tournament. |

N2 | Norms for the titles International Solving Grandmaster of the FIDE (GM), International Solving Master of the FIDE (IM), FIDE Solving Master (FM) can be gained in a tournament if the average of the top 5 participating solvers’ current rating is at least 2550 / 2450 / 2350 respectively. Norms for the titles Women International Solving Grandmaster of the FIDE (WGM), Women International Solving Master of the FIDE (WIM), Women FIDE Solving Master (WFM) can be gained in a tournament if the average of the top 5 participating solvers’ current rating is at least 2450 / 2350 / 2250 respectively. |

N3 | Norms cannot be gained at tournaments which take place in multiple locations. |

N4 | A solver obtains a GM / IM / FM norm when he/she achieves a performance rating of at least 2650 / 2550 / 2450. The solver must be placed within the number of solvers with the qualifying rating of 2550 / 2450 / 2350 (i.e. at least sixth place when there are 6 solvers with the qualifying rating of 2550 / 2450 / 2350). A woman solver obtains a WGM / WIM / WFM norm when she achieves a performance rating of at least 2550 / 2450 / 2350. The solver must be placed within the number of solvers with the qualifying rating of 2450 / 2350 / 2250 (i.e. at least fifth place when there are 5 solvers with the qualifying rating of 2450 / 2350 / 2250). |

### Criteria for gaining titles

T1 | International Solving Grandmaster of the FIDE: A solver must gain a norm 3 times (at least once at the WCSC or ECSC) and achieve a rating of 2550. Women International Solving Grandmaster of the FIDE: A woman solver must gain a norm 3 times (at least once at the WCSC or ECSC) and achieve a rating of 2450. |

T2 | International Solving Master of the FIDE: A solver must gain a norm twice and achieve a rating of 2450. Women International Solving Master of the FIDE: A woman solver must gain a norm twice and achieve a rating of 2350. |

T3 | FIDE Solving Master: A solver must gain a norm twice and achieve a rating of 2350. Women FIDE Solving Master: A woman solver must gain a norm twice and achieve a rating of 2250. |

T4 | The obligatory rating may be achieved anytime (i.e. at any official rating list or as a current rating after a tournament). |

(This version of the criteria was accepted at the PCCC Congress in Wageningen 2006. A small change in the criteria was accepted in Rhodes 2007. Further changes were accepted in Crete 2010 and Berne 2014.)

### Rules for rating calculation

Let us assume that the participants with ratings have ratings R_{1}, …, R_{n}, and the corresponding scores are S_{1}, …, S_{n}. We define the following quantities:

If *CorrCoeff* < 0.6, recalculate the Slope and Intercept, but with both R_{i} and S_{i} sorted in the same order. So the new R_{i} is now the i-th largest rating, and the new Si is the i-th largest result. They may no longer correspond to the same solver.

**A rating for a solver without a rating**:

It is calculated as **performance rating** *(PerfRat)*, i.e. a temporary rating achieved by the solver at the tournament. Performance rating is defined as the rating for which the expected result is the actual achieved result, using the formula:

where*Res* = result of a solver achieved at a tournament Solver’s first rating is called half-rating.

At his next tournament he gets another half-rating. The average of all half-ratings is calculated and put to the rating list.

Expected result for a solver with rating: The expected result (*ExpRes*) is calculated for all solvers with ratings in the following way:

*ExpRes* = *Slope · Rat* + *Intercept*

where

*Rat* = rating of the solver from the last rating list

**Correction**

The expected result might exceed *RMAS*. *RMAS* (round maximum achievable score) is the sum of best results of any solver in each round. This may be higher than winner’s score. In such a case we recalculate the Slope and Intercept in the following way:

where

*MaxRat* = the highest solver’s rating

The performance ratings and expected results are calculated the same way as before, using *CorrSlope* and *CorrIntercept*.

**Change of rating** (*ChOfRat*) is calculated from the difference between the expected result and the achieved result upon the formula:

*ChOfRat = KT × (Res – ExpRes)*,

or in case of correction:

*ChOfRat = KT × (Res – CorrExpRes)*,

where

*KT* = tournament coefficient (from 1 to 4, see Annex)

**New rating is calculated using the formula**:

*NewRat = Rat + ChOfRat*

All calculations are made to the second decimal place. Ratings are published as integers. Note that it is no longer necessary to subtract 1600 from ratings for the calculation.

Ratings are calculated after each tournament. An updated rating list is published four times a year (at 1^{st} January, 1^{st} April, 1^{st} July and 1^{st} October).

After three years of non-participation in rated tournaments, a solver will be removed from the rating (half-rating) list. Should he participate in the future, his original rating (but not half-rating) will be accepted.

### ANNEX

**Tournament coefficient**

Category | Coefficient | Criteria for tournament |

A frame for coefficients for tournaments organised according to the WCSC rules: | ||

W40 | 4 | WCSC, ECSC |

W30 | 3 | WCSC-type tournament for norms |

W25 | 2.5 | participation of at least 5 solvers with a rating of 2300 or higher |

W20 | 2 | participation of at least 5 solvers with a rating of 2200 or higher |

W15 | 1.5 | participation of at least 5 solvers with a rating of 2100 or higher |

W10 | 1 | Other tournaments organised according to the WCSC/ECSC rules |

A frame for coefficients for tournaments organised according to other rules: | ||

O20 | 2 | WCCC and ECSC Open |

O15 | 1.5 | Participation of at least 15 rated solvers from at least three different countries and with at least 5 solvers with a rating of 2350 or higher |

O10 | 1 | Other tournaments |

The representative (committee) has the right to confirm or not the acceptability of the tournament and determines the tournament coefficient.

This version of the rules was accepted at the PCCC Congress in Wageningen 2006. A change in the publication of the rating list was accepted in Rhodes 2007.

A change regarding negative *NewRat* was accepted in Rio 2009.

Changes of the criteria for the acceptability of tournaments at which ratings and norms can be gained were accepted in Berne 2014, in Dresden 2017, in Ohrid 2018, in Vilnius 2019, in Rhodes 2021, in the online meeting of February 2022 and in Batumi 2023.

Rating Rules, PDF (updated in October, 2023)