Confirmations and Line-Ups
Confirmations
As stated earlier, confirming carries a degree of commitment to turn up on time and play for the whole session. It also carries a commitment to pay a share of the charges and costs.
If you confirm on behalf of another, you’re “underwriting” that other person’s commitments, so I suggest that you only do so if that other person cannot access the blog at that time, but has confirmed to you by SMS or e-mail that he is confirmed.
Line ups
The line-ups are currently determined as follows – once the list of those confirmed is finalized for each session, I place players in alternating teams. Therefore, the order in which confirmations are received determines the line-up, which makes it fairly random (although if the two sides are very obviously unbalanced, I have on occasion placed players two at a time to change the balance). There is also the recourse of making adjustments at the session itself, but this is hardly ever done.
Nevertheless, the issue of unbalanced sides has come up again in the last month or so, including after a session at Sports Planet on 2 September 2008, and Monday’s session at The Premier Pitch.
More evenly balanced games are certainly more competitive and enjoyable.
It has been suggested before that stats be used to determine line-ups. As I have been maintaining the result for each participant at every session since 1 September 2008, I have included the difference between each participant’s win-ratio against loss-ratio, which works out to figure between 1.00 and -1.00.
Once the participants for each session is confirmed, I will list the participants in order of that figure, and divide them into teams as follows:
- Team A – 1st, 4th, 5th, 8th, ….
- Team B- 2nd, 3rd, 6th, 7th, ….
I’ll start for next week’s sessions, and we’ll see how that works out.
Jiv:
Kin, George, Raj and I has a short discussion after Monday at Mase’s. We thought it was a good game and the teams were finely picked according to the system above.
However, someone made the observation that the system, while assigning players to teams anonymously and fairly, has the disadvantage in not being able to discern the actual needs of each team. For example, it could be that one side is “over-loaded” with defenders, while the other is similarly spoilt for choice – but with forwards.
The alternative alternative is the old, old way of assigning “captains” (rotating week by week) and getting them to choose from the available pool on the day itself. Captain A chooses 1st, Captain B chooses 2nd and 3rd. Etc. Of course everyone will only know the composition of the teams on the evening itself and have to come prepared with Red and Dark jerseys everytime.
Anyway, there is surely no one single method that can get it “right”.
Well, if there is a consensus, we can try that for Mondays and see how it works out. Rotating the captaincy might be hard, especially to keep track, and with changing personnel. It might be simpler if the two highest ranked players are the captains.
As the rankings carry a certain degree of expectation as to performance, from now on, only those who have played in 3 sessions or more since 1 September 2008 will be included in the rankings.
On the other hand, performance diminishes for each week that a player is away and does not participate. Therefore, the win-loss differential ratio will be adjusted by -0.25 for each week that a player does not participate, up to a maximum of 4 weeks. Beyond 4 weeks, it doesn’t make much difference.
By doing so, the rankings also take into account the stated objectives of regularity and sustainability.
To see how the rankings are computed, download the Excel worksheet, and have a look at Sheet 2, on which the computation is shown.
If nothing else, it makes far more sense than CDOs and CDSs.
To avoid several changes in the line ups arising from late withdrawals and additions, where I’m doing the line ups based on the rankings, I’ll only provide the line ups after 5 pm on the day of the session. I’ll also e-mail the line ups to those playing in that session.
However, the colours for each session will be designated before the start of the week. If you’re not going home before the session, bring both colours, and you can always change into the other after the session.
If you’re not going to be online after 5 pm, SMS me, the person in charge of the session, or any other participant, to find out your colour.
For next week, I’m considering an alternative order for the team assignments:
Team A: 1st, 4th, 6th, 7th, 10th
Team B: 2nd, 3rd, 5th, 8th, 9th
Unranked players joining us for the first time have been treated as the lowest ranked for purposes of team assignment. However, as the new players are sometimes quite good, it skews the strength of the team. From next week onwards, I’m considering assigning unranked players as the seventh or eighth player.
Apart from Tom, the other team selectors under the line up selection method never seemed particularly interested in selecting their teams:
• George: 20 October 2008, 10 November 2008, 24 November 2008
• Clive: 17 November 2008, 24 November 2008.
• Gilbert: 1 December 2008
[...] maintain player stats, I started keeping player stats in September 2008, dating back to 1 September 2008. The player rankings [...]