Our CRA is being upgraded!

When  TFIM  implemented the  CRA, there was no existing system to respect alignment on the courts. Over the years, TFIM has upgraded its CRA system and has worked with Tenniscores to merge the two systems. Tenniscores was using the CRA to meet our needs but was not managing the system.

Tenniscores has now elaborated on a system similar to our CRA (thanks to TFIM), which is an improvement over our creation. TRS (Tenniscores rating system) evaluates more accurately the caliber of doubles playing than our current CRA. 

From now on,  TFIM will use the new TRS to calculate the CRA. It will still be called CRA, but will use the Tenniscores formula. Tenniscores will manage the system entirely, which means significantly less work for us, especially for our Webmaster at no additional cost. 

If we compare CRAs of  2019-20 and CRAs for the next season, the differences are as follows:

  1. The new CRA numbers are between 1 and 6 as opposed to 2 and 26.
  2. The number of points attributed is the same (in absolute value) for the winners and the runner-ups.
  3. For the matches, regardless of the outcome (win or loss), the change in CRA will be based also on performance and not only on victory :
  1. If Exp %* > % actual**, performance is below expectations, therefore the CRA goes down 
  2. If Actual % > EXP %  performance is above expectations, therefore the number of points attributed is positive and the CRA increases.
  3. If EXP % = Actual % the CRA remains unchanged.

*EXPected % = probability of winning against opponents
**% actual = outcome of the match or performance


  •     Tenniscores has assigned a CRA to all the players that are already in the system according to the level in which they play and the court on which they played their first match.
  •     New players will have a CRA assigned by Tenniscores according to the court on which they play their first match. 

For the coaches: Tenniscores has used match data since 2016 to assign a CRA to each player according to their level. A certain number of matches are required to balance a CRA. An initial adjustment might be necessary for the players that have not played 10 matches over the last 4 years. However, aside from these players, we do not expect any other required adjustments. While this information may seem a little overwhelming, rest assured that we will communicate more information to you before the start of the season.

In the meantime, take advantage of the nice weather, play single, keep in shape and stay healthy!

Michelle Aubé
Présidente TFIM

How does the "CRA" formula work?

Whatever the level, the variations in the CRA factor per match are between 0,14 and 0,50 for a win and -0,14 & -0,50 for a loss.

The following include examples taken from level II where basic CRA factors are between 12,00 & 17,00 in the early season.

The formula generating a variation in the CRA factor has 2 parts: the probability of winning the match "Expected %" and the outcome of the games. As a first principle we want a fair game where the odds are 50-50. However, we all know a match is not always 50-50; this is the "EXP %" to win part. The second half of the CRA formula is directly related to the % of games won.

  • 1st half of the formula: regardless of the court, the CRA totals from both teams are subtracted from each other : EX 34.54 - 32.69 = 1.85. The table of EXP % * gives these pairs, 70% and 30% probabilities of winning. The team with a probability of 70% is supposed to win. For a win, the higher the Expected %, smaller is the increase in the CRA factor and the lower the Expected %, greater is the increase in the CRA factor. It's the opposite in a loss; the higher the probability, the greater is the decrease in the CRA factor and the lower the EXP % the smaller is the decrease in the CRA factor.
  • 2nd half of the formula takes into account the % of games won: EX: 6-4, 5-7, 6-3 gives 17 games for the winners and 14 games for the losers or 17/(17+14) = 55% versus 14/(17+14) = 45% of the games.

EXAMPLE: This 70-30 match with scores of 6-4, 5-7, 6-3 (55% & 45%) has 2 possible outcomes:

  • If the team with the 70% wins, the variation in the CRA factor is 0,22 and will be added to each winner's personal CRA factor.
    Calculations: 0,28# X (100% - 70%) + 0,25 ** X ( 55% of the games) = 0,084 + 0,1375 = 0,22. Therefore a decrease of -0,22 per player in the defeated team.
  • If the team with the 30% wins, the variation in the CRA factor is 0,33 added to each winner's personal CRA factor (-0,33 for the defeated team)
    Calculations: 0,28 X (100% - 30%) + 0,25 X (55% of the games) = 0,196 + 0,1375 = 0,33

As previously stated, the variations in the CRA factor per match are between 0,14 and 0,50 for a win and -0,14 & -0,50 for a loss.

  • 0,14 = a match won (7-6, 0-6, 7-6) by the team having 90% chance of winning
  • 0,50 = a match won (6-0, 6-0) by the team having 10% chance of winning
  • -0,14 = a match lost (6-7, 6-0, 6-7) by the team having 10% chance of winning
  • -0,50 = a match lost (0-6, 0-6) by the team having 90% chance of winning.
    *The EXPECTED % is a statistic worked out in an exponential table
    # 0,28 is a constant established by the CRA committee
    ** 0,25 is a constant established by the CRA committee

CRA committee report

For more on the CRA read the "Rapport du comité CRA" tabled on December 1, 2014 (in French only)

NB: amendement to the report tabled in December 2014:

While trying the new numbers we realized that the margin of "4" between 4 & 8, 8 & 12 etc.. was insufficient. It is more convenient to leave the initial extreme minimum of 2 and a margin of "5" per level. This leaves more room to manouvre within a team of 12 players. The particularities within the level 1 makes it difficult to not increase the extreme maximum.

for the 2015-16 season the CRA per level are:

  • Level IV: between 2 & 7
  • Level III: between 7 & 12
  • Level II: between 12 & 17
  • Level I: between 17 & 24


CRA (rule 2.2.6) & MNP (rule 1.8).

The CRA (calculation for the respect of the alignment) and MNP (My Number of Point system) are two distinct formulas. CRA counts in all games played and generates without exception even during retired matches and substituting. The CRA can increase in a win and decrease in a loss. The MNP only adds up the wins and only increases with a win. Although the CRA and MNP give us an idea of the level of play of a player, the two are independent.