FIOFAFI Archive

Regular indoor football, sport and other activity for working adults in Singapore

Rankings & Line Ups: Some Themes

Posted by rajiv on 18 April 2009

As everyone is now getting at least one participation point for every session or game played, I’ll remove the part of the ranking formula giving weight separately to the number of session/games played.

For those who found the way the formula was expressed too difficult, it can now be put more simply as follows:

(TP + TV) / (P + 1.5)

Where:
“TP” is the total ranking points earned in the current participation period
“TV” is the total number of votes garnered in the current participation period in the player polls.
“P” is the total number of games played in the current participation period.
The “current participation period” comprises the current and the preceding quarters

The formula provides an average, to cater for the fact that the number of sessions/games played by each participant may vary greatly.   As an average, the difference between wins and losses is far more significant than the total number of wins.  This has consistently been the case since Tom first started keeping stats for just over a week in August 2008.

The total number of votes can be a significant factor too.   Players earn points for a good performance, regardless of whether they finish on the winning or losing side.

The “+1.5” in the denominator is a fixed weight, to prevent new players from coming straight in at number 1 or in the top 10 just by winning their first couple of games.

The rankings are therefore based on each participant’s tweaked average ranking points (“TARP”).  Hopefully, it makes more sense than the US government’s “Troubled Assets Relief Program”.

As points were earned differently in the preceding quarter, in order to iron out the discrepancy between the preceding quarter and the current quarter, all wins in the preceding quarter have been re-classified as 2-pointers.   As I was not distinguishing between a 1-point draw and a participation point in the preceding quarter, all other results in that quarter are now re-classified as 1-point draws or participation points.

Once all the places in a session are filled, anytime before the session starts, any two regulars who are playing in that session can contact each other and pick their respective teams.  Like everything else on FIOFAFI, this is entirely voluntary.

In what order the two selectors pick their sides, or what criteria they use to pick their sides, is left entirely to them.  I won’t be involved in the selection process.  All that is required is that either of the two selectors put up the selected line ups on behalf of both of them as a comment to the “Schedule” post.  Both selectors need to be named in the comment.

I will check the blog between 4 and 5 pm each day.  If no line ups have been put up, I’ll put up the “default” line ups based on the rankings.  As far as determining line ups before each session is concerned, that is all I am able and willing to do.

In preparing the “default” line ups, I have been following the seedings strictly, except that I try to keep a new player on the same team as his introducer.   If there is a new player, the session will be a 2-pointer.  In a 2-pointer, I will also try from now on to divide newer players and players who haven’t played for a fortnight or more between the two sides.

Even if “default” line ups are put up between 4 and 5 pm, any two regulars can still select their sides any time before the session starts, or there can be more limited player swaps. Any further changes in line ups after the session starts should be agreed to by those playing.

To allow for all of this, it is useful if everyone brings both colours for the day.

If any participant gets on a winning streak, he will move up the rankings.  Under the “default” rankings/seedings method, when he is higher up the rankings, he will be a higher seed for each session he plays, and there will be more lower seeds on his side, so his matches will get tougher, and he’ll probably end up losing a few, with the likelihood of slipping down the rankings.

Likewise, if any participant is on a losing streak, he will move down the rankings.  Under the “default” method, when he is lower down the rankings, he will be a lower seed for each session he plays, and there will be more higher seeds on his side, so his matches should get easier, and he’ll probably end up winning a few, with the likelihood of climbing the rankings.

It balances itself out in the longer term.   After all, FIOFAFI is about sustainability, and fun.

The player poll votes should do more to push the better players higher up the rankings.  I can’t stop anyone voting for themselves, but I would hope that players have sufficient humility not to vote for themselves.  However, if it makes anyone feel better about themselves, there’s no harm.  One vote alone doesn’t make much difference, either to the player poll, or the rankings.

However, voting for oneself can be counter-productive in the longer term – over time, it pushes the player higher up the rankings than he would otherwise be, which means he is more likely to go into his side as a higher seed.  As such, he likely to face tougher matches, with a greater chance of losing.   More losses pushes a player down in the rankings, and as such, it is likely to act as a counter-balance to consistently voting for oneself in the player polls.  With tougher matches to boot.

Advertisements

18 Responses to “Rankings & Line Ups: Some Themes”

  1. […] Rankings & Line Ups: Some Themes […]

  2. […] Rankings & Line Ups: Some Themes […]

  3. rajseran said

    Hi Rajiv

    where is the link to the latest standings ?

  4. […] Rankings & Line Ups: Some Themes […]

  5. […] of the two can then put up the line ups decided by the selection method as a comment […]

  6. […] a 5-a-side or 14 for a 7-a-side), the session is confirmed.  Line ups are determined by either the selection method or the default […]

  7. […] the profile of the participants for a session so allows, sometime in the future, I might try line ups based on the players’ favourite clubs, for example, Manchester United fans v Liverpool […]

  8. […] new, from now on, I’m not including anyone who hasn’t played at least 5 sessions in the current participation period in the top […]

  9. […] they’re actually part of the system now, because the results and votes feed back into the rankings and seedings, on which the line ups are […]

  10. […] the players for a session are finalized, any two regulars can pick their sides using the selection method.  If there are no line ups selected 4 to 5 hours before the session, I will use the default method […]

  11. rajiv said

    As players with less than 5 sessions/games are usually just starting to move up the rankings, I may seed those who have played 5 sessions/games or more separately from those who have played less than 5 sessions/games before determining the default line ups.

    As the latest rankings now comes out on Sunday mornings, the seedings for any Sunday afternoon or evening session will be the first to use the latest rankings.

  12. […] The ranking formula can be found here. […]

  13. Steve F said

    Did I drop out of the list, or am I rubbish enough not to be worth giving a score? 🙂

    Steve

    • rajiv said

      You’ve been out of the top 40 for a while, probably because you went a while without playing. It changes week to week. You’re still on the full listing (Excel document), at 57.

      A winning run can push you up quite fast, a losing run can push you down quite fast. Playing more games also tends to push you up. It’s more about form than ability.

      The full listing has a lot more details, such as highest position reached etc.

      You’ll move up at the end of this week after the win on Wednesday. It’s all for fun anyway.

      Personally, I’d rather not be too high up.

      • Steve F said

        So, it’s confirmed – I’m rubbish. At least under no illusions, like Andy M…

  14. […] by rajiv on Sun 19 July 2009 The end-of-week rankings has been re-named end-of-week table, and will no longer be used to determine the following […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

 
%d bloggers like this: